Microcapsules Made of Weak Polyelectrolytes: Templating and Stimuli

5888

Langmuir 2006, 22, 5888-5893

Microcapsules Made of Weak Polyelectrolytes: Templating and Stimuli-Responsive Properties Tatjana Mauser,*,† Christophe De´jugnat,† Helmuth Mo¨hwald,† and Gleb B. Sukhorukov‡ Max Planck Institute of Colloids and Interfaces, Am Mu¨hlenberg 1, 14424 Potsdam, Germany, and IRC/Department of Materials, Queen Mary UniVersity of London, Mile End Road, E1 4NS London, United Kingdom ReceiVed January 10, 2006. In Final Form: April 5, 2006 Hollow microcapsules composed of the weak polyelectrolytes poly(allylamine hydrochloride) (PAH) and poly(methacrylic acid) (PMA) are templated on silicon oxide particles using the layer-by-layer adsorption. The colloidal template is removed with a buffer system of hydrofluoric acid and ammonium fluoride. With this buffer system, the template can be dissolved in mild pH conditions, where the polymeric layers are still stable. The morphology and the thickness of the resulting capsules are investigated with atomic force microscopy. The resulting hollow capsules show pH-dependent properties. The shells are stable over a broad pH range and swell and immediately dissolve for pH values below 2.3 and above 11. If the molecular weight of the poly(methacrylic acid) is increased, the enhanced entanglement of the polymers results in a reversible swelling of the capsules at low and at high pH. The swelling degree is probed with confocal laser scanning microscopy. In addition to the pH-dependent size variations, the different ionization degree of poly(methacrylic acid) as a function of pH is used for the selective binding of calcium ions.

Introduction Stimuli-sensitive microstructures are important devices for new functional materials such as microsensors and drug-delivery systems. They exhibit a distinct and reversible change of properties in response to an external stimulus. Interesting examples of such structures are microcapsules composed of weak polyelectrolytes with pH-responsive functional groups. Because the linear charge density along the polymer backbone is a function of pH, such capsules are sensitive to a variation of the pH of the surrounding medium. Since the layer-by-layer (LbL) adsorption has been developed by Decher and co-workers,1,2 detailed studies have investigated the pH dependent characteristics of weak polyelectrolytes forming multilayers on flat substrates. As the charge of these polymers depends on the pH of the medium, the electrostatic interactions within the multilayer can easily be tuned. Most experiments concentrated on planar films of the weak polyacid poly(acrylic acid) (PAA) in combination with the weak polybase poly(allylamine hydrochloride) (PAH).3-8 However, only little is known about covered colloids9-11 or microcapsules with weak polyelectrolyte shells.12-15 Particularly microcapsules with pH-responsive properties are of interest for many applications, as the pH-dependent changes in morphology, size, and * Corresponding author. E-mail: [email protected]. Tel: +49-331-567-9235. Fax: +49-331-567-9202. † Max Planck Institute of Colloids and Interfaces. ‡ Queen Mary University of London. (1) Decher, G.; Hong, J. D.; Schmitt, J. Thin Solid Films 1992, 210, 831-835. (2) Decher, G. Science 1997, 277, 1232-1237. (3) Choi, J.; Rubner, M. F. Macromolecules 2005, 38, 116-124. (4) Mendelsohn, J. D.; Barrett, C. J.; Chan, V. V.; Pal, A. J.; Mayes, A. M.; Rubner, M. F. Langmuir 2000, 16, 5017-5023. (5) Burke, S. E.; Barrett, C. J. Pure App. Chem. 2004, 76, 1387-1398. (6) Kovacevic, D.; van der Burgh, S.; de Keizer, A.; Cohen Stuart, M. A. J. Phys. Chem. B 2003, 107, 7998-8002. (7) Kharlampieva, E.; Sukhishvili, S. A. Langmuir 2003, 19, 1235-1243. (8) Fery, A.; Scholer, B.; Cassagneau, T.; Caruso, F. Langmuir 2001, 17, 3779-3783. (9) Burke, S. E.; Barrett, C. J. Langmuir 2003, 19, 3297-3303. (10) Schach, R.; Hommel, H.; Van Damme, H.; Dejardin, P.; Amsterdamsky, C. Langmuir 2004, 20, 3173-3179. (11) Kato, N.; Schuetz, P.; Fery, A.; Caruso, F. Macromolecules 2002, 35, 9780-9787. (12) Gao, C. Y.; Mohwald, H.; Shen, J. C. AdV. Mater. 2003, 15, 930-933.

permeability make them ideal tools for many applications.16,17 The main differences between the adsorption process on planar substrates and colloidal templates arise from the requirement to maintain colloidal stability of the particles during layering and from the subsequent removal of the colloidal template under conditions where the polymeric shell is still stable.18 This premise is limiting for weak polyelectrolyte shells, as the polymeric layers are not stable in extreme conditions. Therefore, the choice of a suitable core material and the dissolution protocol are crucial steps in the formation of weak polyelectrolyte capsules. Kato et al. used melamine formaldehyde (MF) cores and the template was dissolved in acidic conditions at pH 1, leading to PAH/PAA capsules.11 Gao and co-workers studied the stability of the same polymer system on MF as a function of the exposure to 0.1 M HCl.12 However, recent research showed that the molecular weight of the MF oligomers, products of acidic hydrolysis, depends on the age of MF particles and varies from 4 to 14 kDa. Such high molecular weight compounds can affect the integrity of the capsule wall during the dissolution process. Furthermore MF oligomers are positively charged due to amino groups of melamine and can form a complex with negatively charged wall components, making the amount of MF bound to the capsule wall hard to control. These properties of the core material can interfere with the stability and characteristics of the hollow weak polyelectrolyte capsules.19 Another frequently used organic core material is polystyrene (PS), which can be dissolved with tetrahydrofuran or dimethylformamide. It was shown lately that these organic solvents lead to a structuring effect of the multilayer, thereby stabilizing (13) Schuetz, P.; Caruso, F. AdV. Funct. Mater. 2003, 13, 929-937. (14) Mauser, T.; Dejugnat, C.; Sukhorukov, G. B. Macromol. Rapid Commun. 2004, 25, 1781-1785. (15) Tong, W.; Gao, C.; Mo¨hwald, H. Macromolecules 2006, 39, 335-340. (16) Antipov, A. A.; Sukhorukov, G. B. AdV. Colloid Interface Sci. 2004, 111, 49-61. (17) Peyratout, C. S.; Dahne, L. Angew. Chem., Int. Ed. 2004, 43, 3762-3783. (18) Donath, E.; Sukhorukov, G. B.; Caruso, F.; Davis, S. A.; Mohwald, H. Angew. Chem., Int. Ed. Engl. 1998, 37, 2202-2205. (19) Gao, C.; Moya, S.; Lichtenfeld, H.; Casoli, A.; Fiedler, H. Macromol. Mater. Eng. 2001, 286, 355.

10.1021/la060088f CCC: $33.50 © 2006 American Chemical Society Published on Web 05/18/2006

Microcapsules Made of Weak Polyelectrolytes

layers also in extreme conditions.20 As the core removal process may affect the shell structure, the material of the template particle can influence properties such as permeability and mechanical characteristics of the resulting capsules. Since many applications depend on the change of wall properties or variations in the arrangement of the polymers, templates not affecting the chemical composition of the capsule wall or its mechanical properties are needed. Therefore, several inorganic particles have been used as templates. Colloidal CaCO3 particles can easily be dissolved in mild conditions with complexing agents, and the ions formed during dissolution do not interact with most polyelectrolytes. However, these CaCO3 cores are polydisperse and have a porous structure, leading to a huge inner surface that favors the formation of a polymeric matrix in the capsule interior, which can be a disadvantage, especially for studying the permeability of the shells.21 Another inorganic core material is silicon oxide, but to date the acidic conditions for the dissolution with hydrofluoric acid (HF) only allowed to use it as a template for strong polyelectrolyte capsules22-24 or for hydrogen bonded multilayers that are stable in acidic conditions.25,26 Schu¨tz and Caruso focused on multilayers of weak polyelectrolytes adsorbed on SiO2 cores, but only after cross-linking of the multilayers the core material was removed with 1 M HF.13 This work is driven by the aim to obtain a quantitative control of structural changes of capsule walls which is mandatory for practical applications. Therefore, parameters have to be varied in a systematic way as well as the system preparation considering the limitations mentioned above. We use SiO2 particles as templates for the deposition of weak polyelectrolyte multilayers and present a new method to dissolve the particles in mild pH conditions. The resulting hollow capsules consist of the weak polyelectrolytes poly(methacrylic acid) (PMA) and PAH. We also investigate their pH dependent stability and stimuliresponsive behavior. As the ionization of the polymers is regulated by the acidity of the solution, uncomplexed moieties within the film can be used for selective reactions. Rubner and co-workers as well as Schu¨tz and Caruso used multilayers of weak polyelectrolytes as nanoreactors for the syntheses of nanoparticles in flat films and on covered colloids. The anchoring points of these reactions were uncompensated carboxylate or ammonium groups that could be obtained by post assembly changes of pH.27,28 We probe the different degree of ionization of the carboxylic groups of PMA by the pH dependent attachment of calcium ions. Experimental Section Materials. Poly(allylamine hydrochloride) (PAH; Mw ) 70 kDa), poly(methacrylic acid) (PMA; Mw ) 75.1 kDa and Mw ) 790 kDa), HCl, NaOH, NaCl, ethylendiaminetetraacetic acid disodium salt (EDTA), CaCl2, H2O2, NH3, NH4F, and HF were purchased from Sigma-Aldrich (Germany). Monodisperse silicon oxide particles with a mean diameter of 4.48 ( 0.26 µm were obtained from Microparticles (20) Dejugnat, C.; Sukhorukov, G. B. Langmuir 2004, 20, 7265-7269. (21) Volodkin, D. V.; Petrov, A. I.; Prevot, M.; Sukhorukov, G. B. Langmuir 2004, 20, 3398-3406. (22) Adalsteinsson, T.; Dong, W.-F.; Scho¨nhoff, M. J. Phys. Chem. B 2004, 108, 20056-20063. (23) Itoh, Y.; Matsusaki, M.; Kida, T.; Akashi, M. Chem. Lett. 2004, 33, 1552-1553. (24) Ko¨hler, K.; Shchukin, D. G.; Mo¨hwald, H.; Sukhorukov, G. B. J. Phys. Chem. B 2005, 109, 18250-18259. (25) Yang, S.; Zhang, Y.; Yuan, G.; Zhang, X.; Xu, J. Macromolecules 2004, 37, 10059-10062. (26) Zhang, Y.; Guan, Y.; Yang, S.; Xu, J.; Han, C. C. AdV. Mater. 2003, 15, 832-835. (27) Joly, S.; Kane, R.; Radzilowski, L.; Wang, T.; Wu, A.; Cohen, R. E.; Thomas, E. L.; Rubner, M. F. Langmuir 2000, 16, 1354-1359. (28) Schuetz, P.; Caruso, F. Chem. Mater. 2004, 16, 3066-3073.

Langmuir, Vol. 22, No. 13, 2006 5889 (Germany). All chemicals were used without any further purification. PAH labeled with tetramethylrhodamine isothiocyanate (PAHTRITC) was synthesized as described elsewhere.29 Water from a three-stage Millipore Milli-Q Plus 185 purification system was used in all experiments. Capsule Preparation. Prior to the deposition of polyelectrolytes, the surface of the silica particles was roughened by 20 min exposure to a mixture of 5:1:1 H2O:H2O2 (30%):NH3 (25%) to ensure a better attachment of the polymers to the bare SiO2 surface. Then the particles were washed with water to reach neutral pH. For all deposition steps, 2 mg/mL polymer solutions with a NaCl concentration of 0.2 M were used at pH 5. The adsorption step was 15 min followed by centrifugation at 1400g for 1 min to remove the supernatant. Each deposition step was followed by triple washing with water at pH 5 to remove excess polymer. After the desired number of layers were deposited, the silica particles were dissolved with a buffer system of 0.2 M NH4F and HF at pH 4.5. To ensure a complete core removal the buffer system was used five times for 1 h. The resulting hollow polymeric capsules were washed 5 times with water at the same pH to remove HF, SiF62- and NH4F. Energy-Dispersive X-ray Analysis (EDX). EDX analysis of the capsules was performed with a Zeiss DSM 940 scanning electron microscope. Atomic Force Microscopy (AFM). The thickness measurements of PAH/PMA capsules in dried state were performed using a Nanoscope III Multimode AFM (Digital Instruments Inc., U.S.A.) operating in tapping mode. The samples were prepared by applying a drop of the capsule solution to a freshly cleaved mica surface and drying it in air. To ensure a better attachment of the negatively charged shells, the mica surface was precoated with poly(ethylene imine) (PEI). The single wall thickness of a capsule was determined as half of the height of the collapsed flat regions of dried shells. At least 20 profiles of different capsules were analyzed, and the mean thickness differences between the mica surface and the lowest region of the shells were averaged. Confocal Laser Scanning Microscopy (CLSM). Confocal images of PAH/PMA capsules in solution were obtained with a Leica TCS SP confocal scanning system (Leica, Germany) equipped with a 100×/1.4-0.7 oil immersion objective lens. The polyelectrolyte multilayers were visualized by incorporation of a rhodamine labeled polymer (PAH-TRITC) during capsule preparation. The diameter of the capsules was determined with the Leica confocal software (LCS, version 2.5). pH Measurements. For pH studies the pH of the capsule solutions was adjusted with 0.1 M HCl and 0.1 M NaOH. All measurements were performed with a calibrated microprocessor pH-meter WTW pH 539 (WTW, Germany) using a combined glass/reference electrode.

Results and Discussion Dissolution of SiO2 with HF. Silica particles are commonly dissolved with diluted aqueous solutions of hydrofluoric acid. As HF is a weak acid with a pKa of 3.2 at 25 °C,30 the pH of its aqueous solutions is a function of the HF concentration. Some studies reported the use of SiO2 particles as templates for the fabrication of hollow microshells by the LbL process. In all cases, the dissolution process was accomplished in rather acidic conditions which are not suitable for weak polyelectrolytes. Itoh and co-workers used 46% HF for the dissolution step,23 Schu¨tz et al. dissolved the cores with 1 M HF, and Adalsteinsson et al. worked with a concentration of 0.5 M HF.22 The lowest HF concentration for the template dissolution reported by our group was 0.1 M,24 resulting in a solution pH of 2.1. At this pH, the electrostatic interactions within the multilayers of weak polyelectrolytes are too reduced to keep the integrity of the shells. (29) Ibarz, G.; Da¨hne, L.; Donath, E.; Mo¨hwald, H. AdV. Mater. 2001, 13, 1324-1327. (30) Perrin, D. D. Ionization Constants of Inorganic Acids and Bases in Aqueous Solution; 2nd ed.; Pergamon: Oxford, 1982.

5890 Langmuir, Vol. 22, No. 13, 2006

Mauser et al.

Figure 2. Thickness of the capsule wall as measured by AFM for (PAH/PMA75kDa)n (closed symbols) and (PAH/PMA790kDa)4 (open symbol).

Figure 1. AFM images of dried capsules templated on SiO2. (A) (PAH/PMA75kDa)4, dissolution with 0.2 M NH4F at pH 5, maximum height (hm) ) 400 nm. (B) (PAH/PMA75kDa)4, dissolution with 0.2 M NH4F at pH 4.5, hm ) 300 nm. (C) (PAH/PMA75kDa)8, dissolution with 0.2 M NH4F at pH 4.5, hm ) 600 nm. (D) (PAH/PMA790kDa)4, dissolution with 0.2 M NH4F at pH 4.5, hm ) 300 nm.

Thus, the pH of the medium has to be increased. Theoretically, it is possible to work with a more dilute HF solution, but this would result in 0.4 L of aqueous HF solution for the dissolution of 10 mg of SiO2 at pH 3. A better choice is to use a buffer system prepared of HF and NH4F. According to the HendersonHasselbach equation the pH of the buffer system is only related to the pKa of the acid and the relative concentrations of HF and NH4F. This allows us to work in mild pH conditions with an increased HF concentration. The chemical equations (1) and (2) illustrate the interplay of the creation of HF in the buffer equilibrium and the consumption of HF during the dissolution of SiO2.

NH4F + H+ T HF + NH4+

(1)

SiO2 + 6 HF T H2SiF6 + 2 H2O

(2)

The combination of the two reactions ensures a constant amount of hydrofluoric acid and a constant pH of the solution which is critical for both the multilayer stability as well as colloidal stability during the core dissolution process. AFM images of dried PAH/PMA capsules after core removal using a 0.2 M NH4F solution at different pH values are displayed in Figure 1. At pH 4 and below, no stable capsules could be obtained, as the electrostatic interactions between the polymers are too low to provide the integrity of the shells during the dissolution step. In Figure 1A, the dissolution of the core at pH 5 for capsules made of 8 polymer layers is shown. The shell structure is fuzzy and the mean height of more than 100 nm also suggests that there is still silica gel trapped inside the capsules. An example of a (PAH/PMA)4 capsule with the core dissolution at pH 4.5 is given in Figure 1B. Dissolving the template at a pH of 4.5 led to smooth and thin hollow capsules until a maximum of 14 layers. If more than 14 polymer layers were adsorbed onto the silicon oxide, complete core removal could not be achieved under the above-mentioned conditions. An AFM image of PAH/

PMA capsules made of 16 layers is shown in Figure 1C. The blurred structure of the shell and the increased height profile suggest that some silica was still present in the capsules, probably because the shells were too thick or too dense to allow the core material to penetrate. Increasing the molecular weight of PMA to 790 kDa on the other hand did not have any influence on the dissolution process, and stable capsules of 8 layers of PAH and high molecular weight PMA are displayed in Figure 1D. For the capsules shown in Figure 1, panels B and D, no silicon could be detected in EDX measurements, ensuring the complete dissolution of the core. AFM measurements were performed on dried capsules to estimate the thickness of their walls. The results are displayed in Figure 2. For capsules made of low molecular weight PMA, the thickness increases linearly from 8 to 14 layers with a resulting bilayer thickness in the range of 3.0 ( 0.2 nm. The thickness of capsules prepared with high molecular weight PMA is 3.4 ( 0.3 nm per bilayer. The small differences in the thickness of the shells could be due to a larger entanglement of the polymers when the molecular weight is increased, leading to an increased thickness of each layer. These thickness values are lower than the reported 5-5.5 nm per bilayer found for most of the strong polyelectrolyte systems that were adsorbed with an ionic strength of 0.5.24,31 The reduced thickness can be explained by a reduced salt concentration of 0.2 M in comparison to 0.5 M for most of the adsorptions involving strong polyelectrolytes, leading to thinner polymer layers. Another influencing variable is the pH dependent thickness behavior of weak polyelectrolyte systems that was studied in detail by Shiratori et al.32 For PAH, the apparent pK (ionization degree of 50%) is shifted from 8.6 in solution to 10.8 in a complex with PMA. For PMA, the pKapp is shifted from 6.8 in water to 3.9 in a complex with PAH.14 As both polymers are adsorbed at pH values below (PAH) and above (PMA) their respective pKapp in a multilayer they yield thin films such as strong polyelectrolytes. pH-Dependent Behavior of (PAH/PMA75kDa)4 Capsules. Weak polyelectrolyte capsules templated on SiO2 have the same characteristic pH-dependent stability reported for all weak polyelectrolyte systems. They are stable over a broad pH regime but dissolve at extreme high or low pH. The stability of the capsules as a function of pH as well as some CLSM images of the dissolution process is highlighted in Figure 3. Shells made of 8 layers of PAH and low molecular weight PMA are stable within the pH range from 2.3 to 11. In this pH range, the size of the capsules is nearly constant with a mean diameter of 4.4 (31) Dubreuil, F.; Elsner, N.; Fery, A. Eur. Phys. J. E 2003, 12, 215-221. (32) Shiratori, S. S.; Rubner, M. F. Macromolecules 2000, 33, 4213-4219.

Microcapsules Made of Weak Polyelectrolytes

Langmuir, Vol. 22, No. 13, 2006 5891

Figure 4. Diameter of (PAH/PMA790kDa)4 capsules as a function of pH. The average diameter was calculated from 20 individual capsules. The hatched areas indicate regions where the capsules are dissolved.

Figure 3. (A) Diameter of (PAH/PMA75kDa)4 capsules as a function of pH. The average diameter was calculated from 20 individual capsules. The hatched areas indicate regions where the capsules are dissolved. (B) CLSM images of the dissolution process at low pH.

( 0.3 µm, corresponding to the size of the silicon oxide templates. If the pH is increased or decreased to extreme values, the shells swell and dissolve within seconds. The dissolution at low pH is shown in Figure 3B. The stability in the intermediate pH range can be explained by a global compensation of charges between the ammonium groups of PAH and the carboxylic groups of PMA. At low pH, the carboxylic groups get protonated and thereby loose their charges. Because the ionization of the ammonium groups is not reduced, this results in a lowered electrostatic interaction of the polymers as well as in an increased electrostatic repulsion between uncompensated positive charges of PAH. In addition small counterions can penetrate the layer structure to compensate the positive charges. The higher ionic concentration compared to the surrounding solution increases the osmotic pressure and results in the permeation of water into the membrane; thus, the membrane starts to swell. At some point, the electrostatic interaction is not strong enough to keep the multilayers stable, and the capsules swell and dissolve. At high pH, the situation is complementary. Similar effects have also been reported at high pH for the dissolution of (PAH/polystyrene sulfonate (PSS)) capsules templated on CaCO3 or MnCO3.20 Regardless of the number of deposited layers, it was not possible to stabilize the swollen capsules at low pH. Such a stabilization can be obtained for the same polymer system on CaCO3 particles because of intermolecular hydrophobic interactions of uncharged PMA that counteract the electrostatic repulsion.14 The differences in stability could be a result of the different nature of the core material. The large cavities in the CaCO3 particles result in a much higher amount of polymer material adsorbed in this case, leading to the formation of a polymeric matrix. Volodkin et al. reported an elevenfold increase in the surface area of porous CaCO3 in comparison to nonporous cores of the same size.21 pH-Dependent Behavior of (PAH/PMA790kDa)4 Capsules. (PAH/PMA)4 capsules prepared from high molecular weight PMA show a different pH dependent stability at the edges of the stability range than capsules with low molecular weight PMA.

Figure 5. Reversibility of the swelling of (PAH/PMA790kDa)4 capsules at low and at high pH. The average capsule diameter is calculated from 20 individual capsules.

The capsule size as a function of pH is shown in Figure 4. The capsules are stable within the pH range from 2.3 to 11.1 and dissolve in extreme conditions. From pH 2.4 to 10.4, the mean diameter is nearly constant with 4.4 ( 0.3 µm. Below pH 2.4 the capsules swell by 20% to a mean diameter of 5.5 ( 0.6 µm. In the pH range from 10.4 to 11.1 they swell by 35% to a diameter of 6.2 ( 0.6 µm. Both swollen structures are stable for several days. At low pH, the swelling occurs over only 0.1 pH unit, whereas at high pH, the increase in size is a more continuous transition over 0.7 pH units. One parameter influencing the stable swollen state at low pH can be attributed to intermolecular hydrophobic stabilization of uncharged PMA that is accompanied by a conformational change of the polymer backbone from coil to globule.33-35 In addition, uncharged polycarboxylic acids are stabilized by hydrogen bonds that can stabilize polyelectrolyte multilayers in acidic media.36 Furthermore, one also has to consider the polymeric entanglement as an important factor in the stabilization process, because capsules prepared of low molecular weight PMA do not show any stabilization of the swollen state. At high pH, the entanglement of polymers seems to be the only stabilization, as no other efficient counteracting forces to the electrostatic repulsion are known for uncharged polyallylamine. The reversibility of the swelling as a function of pH is displayed in Figure 5. Swollen capsules at pH 2.3 and 11.1 were washed (33) Mandel, M.; Leyte, J. C.; Stadhouder, M. G. J. Phys. Chem. 1967, 71, 640-649. (34) Arnold, R. J. Colloid Sci. 1957, 12, 549-556. (35) Anufrieva, E. V.; Birshtein, T. M.; Nekrasova, T. N.; Ptitsyn, O. B.; Sheveleva, T. V. T. J. Polym. Sci., Polym. Symp. 1968, 16, 3519-3531. (36) Izumrudov, V.; Sukhishvili, S. A. Langmuir 2003, 19, 5188-5191.

5892 Langmuir, Vol. 22, No. 13, 2006

Mauser et al.

Figure 6. CLSM images of the effect of calcium ions on (PAH/PMA790kDa)4 in basic pH. Capsules in water at pH 7 (A) and at pH 11.1 (E). Capsules in 10 mM CaCl2 at pH 7 (B), at pH 10 (C) and in 10-4 M CaCl2 at pH 11.1 (D). Experimentally the process has been proceeded from either A f B f C f D or A f E f D. Table 1. Stability and Swelling of (PAH/PMA790kDa)4 Capsules at Low pH as a Function of the Concentration of Calcium Ions in the Solution CaCl2 concentration, mmol/L

pH of the onset of capsules swelling

0 1 5 10 20

2.3 2.4 2.6 2.7 2.9

with water. After 1 h, the size of these capsules was compared with the size of the original ones. The diameter of capsules that were previously swollen at low pH decreased to 4.5 ( 0.2 µm, whereas capsules that were first exposed to a solution of pH 11.1 only showed a decrease in the size to 5.2 ( 0.3 µm. A possible explanation for the differences in the reversibility of the swelling at low and at high pH could be the different polymer conformations at low and at high pH. When the pH is changed after the capsules are swollen, the repulsion between neighboring functional groups is reduced and the electrostatic attraction between the different layers is increased. A change in entanglement of the layers can counteract the shrinking of the shell, resulting in a size that is a bit larger than that of the original capsules. Investigation of the Role of Ca2+ on the Multilayer Stability. The functional groups of weak polyelectrolytes are protonated or deprotonated depending on the pH of the solution; therefore, they either participate in the electrostatic interactions with the other polymer or are available for selective reactions at the uncomplexed moieties. The potential of such reactions was probed by the addition of Ca2+ to the shells at different pH values. Ca2+ is known to bind COO- groups,37,38 leading to the formation of complexes that can precipitate in solution. The effect of Ca2+ binding at low pH is demonstrated in Table 1. The increase of the calcium concentration leads to a decrease in the stability of the capsules at low pH and the onset of the swelling of the shells at a slightly higher pH. This allows for a selective tuning of the stability of the multilayers as a function of added calcium and (37) Eichenbaum, G. M.; Kiser, P. F.; Shah, D.; Meuer, W. P.; Needham, D.; Simon, S. A. Macromolecules 2000, 33, 4087-4093. (38) Dejugnat, C.; Halozan, D.; Sukhorukov, G. B. Macromol. Rapid Commun. 2005, 26, 961-967.

Scheme 1. Mechanism of the Binding of Calcium Ions to Carboxylic Groups in Basic pH

pH. The reduced stability can be interpreted by the competitive binding of Ca2+, ammonium groups of PAH, and H+ to the carboxylic groups, thereby reducing the electrostatic interactions between PAH and PMA which stabilize the multilayers. The effect of calcium in the basic pH range is illustrated in the CLSM pictures in Figure 6. The exposure of capsules to a solution of 10 mM CaCl2 increases the aggregation between the shells, as the outermost layer is negatively charged (deprotonated PMA) and thereby a bridging of Ca2+ between different capsules is favored. Furthermore, the capsules shrink as the pH increases, leading to shrunken particles at a pH of 11.1 (Figure 6, from A f B f C f D). If calcium chloride in a concentration of 10-4 M is added to swollen capsules at pH 11.1 a shrinking of the structures to small particles is induced (Figure 6, from A f E f D). The addition of HCl or EDTA to the shrunken structures does not swell the capsules back to their original size. A mechanism for the Ca-binding is proposed in Scheme 1. The NH4+ groups of PAH are deprotonated in the basic region, thereby releasing carboxylic groups out of electrostatic interactions. As calcium ions are present in the solution, they bind to the carboxylic groups and form a complex. Since there are only a few electrostatic interactions between the polymers left at pH 11.1, a small amount of calcium is sufficient to form a precipitate. The addition of EDTA or HCl reduces either the amount of Ca2+ by complexation or the available binding sites for calcium, but there is no shape memory in the system that would restore the original size. The pH dependent binding effect of calcium shows the versatility of the system to be used for reactions that are restricted to free functional groups within the multilayer.

Conclusions We reported on a new approach for the dissolution of silicon oxide templates for the formation of weak polyelectrolyte

Microcapsules Made of Weak Polyelectrolytes

capsules. The application of a buffer system allowed a high concentration of HF in combination with an increased pH. The resulting capsules are stable over a wide pH range and dissolve in extreme pH conditions. When the entanglement between the polymer chains is large enough, a stable swollen state is observed before the capsules are dissolved. The swelling process is reversible. Furthermore, as the ionization degree of PMA is tuned by pH, the unbound carboxylic groups are available as anchoring sites for the complexation of Ca2+ ions. The calcium-binding reduces the stability of the shells at low pH and leads to a shrinking

Langmuir, Vol. 22, No. 13, 2006 5893

of the capsules in the basic pH region. The reversible swelling at low and at high pH in combination with the pH dependent calcium response make this system a smart stimuli-responsive microreactor with many potential applications. Acknowledgment. The authors thank A. Heilig for the AFM measurements and Dr. J. Hartmann for the EDX-analysis. This work was supported by the 6th FP EU Project STREP001428 “Nanocapsules for Targeted Controlled Delivery of Chemicals”. LA060088F