Formation and Stability of Water-Soluble, Molecular Polyelectrolyte

---

pubs.acs.org/Langmuir © 2009 American Chemical Society

Formation and Stability of Water-Soluble, Molecular Polyelectrolyte Complexes: Effects of Charge Density, Mixing Ratio, and Polyelectrolyte Concentration Alexander Shovsky,*,† Imre Varga,†,^ Ricardas Makuska,‡ and Per M. Claesson§ † Department of Chemistry, Surface and Corrosion Science, Royal Institute of Technology, Drottning Kristinas :: vag 51, SE-100 44 Stockholm, Sweden, ‡Department of Polymer Chemistry, Vilnius University, Naugarduko 24, LT-03225 Vilnius, Lithuania, §YKI, Institute for Surface Chemistry, P.O. Box 5607, SE-114 86 Stockholm, :: :: Sweden, and ^Institute of Chemistry, Eotvos University, P.O. Box 32, 1518 Budapest 112, Hungary

Received January 9, 2009. Revised Manuscript Received March 10, 2009 The formation of complexes with stoichiometric (1:1) as well as nonstoichiometric (2:1) and (1:2) compositions between oppositely charged synthetic polyelectrolytes carrying strong ionic groups and significantly different molecular weights is reported in this contribution. Poly(sodium styrenesulfonate) (NaPSS) was used as polyanion, and a range of copolymers with various molar ratios of the poly(methacryloxyethyltrimethylammonium) chloride, poly(METAC), and the nonionic poly(ethylene oxide) ether methacrylate, poly(PEO45MEMA), were used as polycations. Formation and stability of PECs have been investigated by dynamic and static light scattering (LS), turbidity, and electrophoretic mobility measurements as a function of polyelectrolyte solution concentration, charge density of the cationic polyelectrolyte, and mixing ratio. The data obtained demonstrate that in the absence of PEO45 side chains the 100% charged polymer (polyMETAC) formed insoluble PECs with PSS that precipitate from solution when exact stoichiometry is achieved. In nonstoichiometric complexes (1:2) and (2:1) large colloidally stable aggregates were formed. The presence of even a relatively small amount of PEO45 side chains (25%) in the cationic copolymer was sufficient for preventing precipitation of the formed stoichiometric and nonstoichiometric complexes. These PEC’s are sterically stabilized by the PEO45 chains. By further increasing the PEO45 side-chain content (50 and 75%) of the cationic copolymer, small, water-soluble molecular complexes could be formed. The data suggest that PSS molecules and the charged backbone of the cationic brush form a compact core, and with sufficiently high PEO45 chain density (above 25%) molecular complexes are formed that are stable over prolonged times.

Introduction Mixing a polyanion with a polycation in aqueous media generally leads to spontaneous self-assembly of nanoparticles known as of polyelectrolyte complexes (PECs).1,2 The association between oppositely charged polyelectrolytes occurs due to strong electrostatic interactions, mainly driven by the increase in entropy due to the release of low-molecular-weight counterions. However, other intermolecular interactions, such as hydrogen bonding,3 hydrophobic, charge-transfer, and van der Waals interactions, can also play a significant role. Naturally occurring PECs play an essential role in nature where many charged macromolecules and supramolecular assemblies (e.g., proteins, polysaccharides, and cells) are dispersed in body fluids and tissues without macroscopic precipitation.4 They are essential in many biological processes, for instance, in complex DNA-protein interactions occurring in cell machineries.4,5 Synthetic, stoichiometric PECs were first prepared at the beginning of the 1960s by Michaels et al.6-8 They form when *To whom correspondence should be addressed. (1) Michaels, A. S.; Mir, L.; Schneider, N. S. J. Phys. Chem. 1965, 69 1447–1455. (2) Philipp, B.; Dautzenberg, H.; Linow, K.-J.; Kotz, J.; Dawydoff, W. Prog. Polym. Sci. 1989, 14, 91. (3) Sotiropoulou, M.; Oberdisse, J.; Staikos, G. Macromolecules 2006, 39 3065–3070. (4) Kabanov, A. V.; Gebhart, C. L.; Bronich, T. K.; Vinogradov, S. V. Abstr. Am. Chem. Soc. 2002, 224, U471–U471. (5) Kabanov, A. V.; Kabanov, V. A. Bioconjugate Chem. 1995, 6, 7–20. (6) Michaels, A. S.; Miekka, R. G. J. Phys. Chem. 1961, 65, 1765–1773. (7) Michaels, A. S. Ind. Eng. Chem. 1965, 57, 32–40. (8) Michaels, A. S.; Falkenstein, G. L.; Schneider, N. S. J. Phys. Chem. 1965, 69, 1456–1465.

Langmuir 2009, 25(11), 6113–6121

oppositely charged polyelectrolytes are mixed in a 1:1 charge ratio. Since they are not soluble in any known solvent, they are ideal materials in several large-scale industrial applications e.g. as membrane materials, coatings, binders, and flocculants. Later a considerable interest developed for the preparation of synthetic, nonstoichiometric PECs, dispersed in aqueous solution. These complexes are prospective candidates in various biotechnological and pharmaceutical applications including the encapsulation of drugs, proteins, and DNA.9-11 Systematic studies of the parameters influencing the formation and structure of nonstoichiometric PECs were independently undertaken by the groups of Tsuchida,12 Dautzenberg,2,13 and Kabanov.5,14 The molecular weight of the polyelectrolytes, the concentration of the polyelectrolyte solutions, the mixing ratio of the polyelectrolytes, the salt content of the medium, and the polymer charge density were varied in order to study the nature of complex formation between oppositely charged polyelectrolyte chains. A generally accepted classification of nonstoichiometric, synthetic PECs has emerged from these studies. They are divided into two categories. (i) Highly aggregated complexes: these PECs are large nonequilibrium aggregates of several polyelectrolyte chains. The formed PEC particles are stabilized by the polyion in excess that charges the PEC surface and prevents macroscopic (9) Behr, J.-P. Bioconjugate Chem. 1994, 5, 382–389. (10) Harada, A.; Kataoka, K. Macromolecules 1995, 28, 5294–5299. (11) Harada, A.; Kataoka, K. Science 1999, 283, 65–67. (12) Eishun Tsuchida, Y. O. K. S. J. Polym. Sci., Part A: Polym. Chem. 1972, 10, 3397–3404. (13) Karibyants, N.; Dautzenberg, H. Langmuir 1998, 14, 4427–4434. (14) Kabanov, V. A. Usp. Khim. 2005, 74, 5–23.

Published on Web 4/16/2009

DOI: 10.1021/la804189w 6113

Article

Shovsky et al.

precipitation. (ii) Water-soluble molecular complexes: the formation of water-soluble PECs is an equilibrium phenomenon that can occur when the following special conditions are met:15-17 (1) one component has weak ionic groups, (2) there is a significant difference in the molecular weights of the oppositely charged chains, (3) the mixture contains a high excess of the long-chain component, and (4) some salt is present in the system. The formation of soluble complexes is governed by thermodynamic equilibrium and results in a uniform distribution of the shortchain component among the chains of the oppositely charged long-chain component. If one or more of the above conditions are not met, complex formation results in highly aggregated complex particles in the colloidal range. The structure of PECs of relatively short chains has been thoroughly investigated theoretically by Linse and co-workers.18,19 The results of their investigations have led to the formulations ;of four rules describing the structure and composition of PECs: (i) Coexistence of oppositely charged PECs is avoided. (ii) The cluster excess charge density divided by surface area is minimized to reduce the electrostatic repulsion among excess charges. (iii) The total cluster surface area is minimized to reduce the loss of electrostatic correlation energy. (Positive charges tend to be surrounded by negative charges, but this tendency decreases at the surface). (iv) The number of clusters is maximized to gain translational energy. These rules are in conflict with each other, and consequently predominance of one rule over another is observed under different conditions. They can thus not without difficulty be used to predict the outcome when mixing two polyelectrolytes, but they do provide the guidelines for understanding changes in PEC structure with charge stoichiometry ratio and salt concentration. Experimentally, the situation is complicated by the fact that trapped nonequilibrium states often are encountered as evidenced by observations that e.g. the properties and size of the formed PECs depend on the mixing conditions.20,21 Stoichiometric complexes are (by definition) electroneutral since the charges of the components are mutually neutralized; thus, they lose their colloid stability and precipitate from solution provided no nonelectrostatic repulsive forces exist between the PECs.22 It has recently been shown that the phase separation in stoichiometric mixtures can be prevented if a hydrophilic nonionic block or side chain (e.g., poly(ethylene oxide) (PEO)) is attached to at least one of the polyelectrolytes.23-27 In these cases, the aggregates have a well-defined micellar core-shell structure. The core of the micelle consists of the polyelectrolyte complex, and it is surrounded by a hydrophilic corona of the nonionic blocks. This type of complex is known as a blockionomer complex (BIC). Such core-shell supramolecular structures are, for instance, formed by mixing poly(ethylene oxide)-block-poly (15) Tsuchida, E. J. Macromol. Sci., Pure Appl. Chem. 1994, A31, 1–15. (16) Tsuchida, E.; Abe, K.; Honma, M. Macromolecules 1976, 9, 112–117. (17) Zintchenko, A.; Rother, G.; Dautzenberg, H. Langmuir 2003, 19 2507–2513. (18) Hayashi, Y.; Ullner, M.; Linse, P. J. Phys. Chem. B 2003, 107, 8198–8207. (19) Ryden, J.; Ullner, M.; Linse, P. J. Chem. Phys. 2005, 123. (20) Reihs, T.; Muller, M.; Lunkwitz, K. Colloids Surf., A 2003, 212, 79. (21) Chen, J.; Heitmann, J. A.; Hubbe, M. A. Colloids Surf., A 2003, 223, 215. (22) Kudlay, A.; de la Cruz, M. O. J. Chem. Phys. 2004, 120, 404–412. (23) Sotiropoulou, M.; Cincu, C.; Bokias, G.; Staikos, G. Polymer 2004, 45, 1563. (24) Harada, A.; Okada, M.; Li, J.; Kamachi, M. Macromolecules 1995, 28, 8406–8411. (25) Gohy, J. F.; Varshney, S. K.; Jerome, R. Macromolecules 2001, 34 2745–2747. (26) Kabanov, A. V.; Bronich, T. K.; Kabanov, V. A.; Yu, K.; Eisenberg, A. Macromolecules 1996, 29, 6797–6802. (27) Cohen Stuart, M. A.; Besseling, N. A. M.; Fokkink, R. G. Langmuir 1998, 14, 6846–6849.

6114

DOI: 10.1021/la804189w

(R,β-aspartic acid) and poly(ethylene oxide)-block-poly(L-lysine) diblocks, and these PECs have been shown to present chain length recognition properties.11 The size of these assemblies was found to be in the range of some decades of nanometers. Recently, the reversible formation of water-soluble BICs between sodium poly(4-styrenesulfonate) and a poly(2-vinylpyridinium)block- poly(ethylene oxide) copolymer has been reported.28 For the supramolecular assemblies formed at pH = 3 in dilute solutions, various structures have been observed, extending from spherical to rodlike micelles and vesicles.25 We note that the formation of water-soluble PECs between block ionomers and enzymes29,30 or oligonucleotides31-33 has been proposed for potential applications in enzyme entrapment or gene therapy,4 respectively. Formation of water-soluble PECs involving polyelectrolytes of a comb-type structure are scarce, and there are only a few reports on this topic.23,34-36 The basic concept behind the possible water solubility of such complexes is similar to that used when BICs are employed. Brush polyelectrolytes (also known as comb polyelectrolytes) belong to a class of branched macromolecular structures possessing a high density of side chains on a linear polymeric main chain. Similar architectures can be obtained by forming complexes with a linear polyelectrolyte and a short (ionic-nonionic) diblock polyelectrolyte of opposite charge.37,38 This polyelectrolyte architecture is useful for achieving colloidal stability where both steric and electrostatic stabilization mechanisms can be achieved.39,40 Thus, the molecular design of such comb-type copolymers consisting of various polyelectrolyte backbones and water-soluble nonionic side chains provides a flexible strategy to prepare a large variety of novel water-soluble stoichiometric PECs. Moreover, the features of the complexes can be tuned at will by the comb polymer architecture (e.g., density of the charged backbone and length and graft density of nonionic side chains). Recently, Sotiropoulou et al.23 reported formation of stoichiometric water-soluble PECs formed upon mixing of dilute solutions of poly(sodium acrylate-co-sodium 2-acrylamido-2-methyl1-propanesulfonate)-graft-poly(N,N-dimethylacrylamide) and poly(diallyldimethylammonium chloride). Core-shell assemblies of several decades of nanometers in size were revealed. The stabilization of these nanoparticles was achieved by the nonionic grafted poly(N,N-dimethylacrylamine) side chains. In this report we describe a systematic investigation of the influence of copolymer architecture and polyelectrolyte (28) Gohy, J. F.; Varshney, S. K.; Antoun, S.; Jerome, R. Macromolecules 2000, 33, 9298–9305. (29) Harada, A.; Kataoka, K. Macromolecules 1998, 31, 288–294. (30) Harada, A.; Kataoka, K. Langmuir 1999, 15, 4208–4212. (31) Kabanov, A. V.; Vinogradov, S. V.; Suzdaltseva, Y. G.; Alakhov, V. Y. Bioconjugate Chem. 1995, 6, 639–643. (32) Vinogradov, S. V.; Bronich, T. K.; Kabanov, A. V. Bioconjugate Chem. 1998, 9, 805–812. (33) Vinogradov, S. V.; Batrakova, E. V.; Li, S.; Kabanov, A. V. J. Drug Targeting 2004, 12, 517–526. (34) Sato, A.; Choi, S. W.; Hirai, M.; Yamayoshi, A.; Moriyarna, R.; Yamano, T.; Takagi, M.; Kano, A.; Shimamoto, A.; Maruyama, A. J. Controlled Release 2007, 122, 209–216. (35) Sato, Y.; Moriyama, R.; Choi, S. W.; Kano, A.; Maruyama, A. Langmuir 2007, 23, 65–69. (36) Choi, S. W.; Kano, A.; Maruyama, A. . Nucleic Acids Res. 2008, 36, 342-351. (37) Gus’kova, O. A.; Pavlov, A. S.; Khalatur, P. G.; Khokhlov, A. R. J. Phys. Chem. B 2007, 111, 8360–8368. (38) Kramarenko, E. Y.; Pevnaya, O. S.; Khokhlov, A. R. J. Chem. Phys. 2005, 122, 084902. (39) Naderi, A.; Iruthayaraj, J.; Pettersson, T.; Makuska, R.; Claesson, P. M. Langmuir 2008, 24, 6676–6682. (40) Naderi, A.; Iruthayaraj, J.; Vareikis, A.; Makuska, R.; Claesson, P. M. Langmuir 2007, 23, 12222–12232.

Langmuir 2009, 25(11), 6113–6121

Shovsky et al.

Article

Table 1. Molecular Characteristics of Samples Employed for PEC Formationa brush polymer (-charge density)

Rg Mw (kg mol-1) (nm)

RH (nm)

Rg/ RH B

660 41 15 2.6 0.9 PEO45MEMA:METAC-25 680 40 17 2.4 1.4 PEO45MEMA:METAC-50 520 35 20 1.8 2.3 PEO45MEMA:METAC-75 a Data extracted from SLS measurements:40 Mw = weight-averaged molecular weight, Rg = radius of gyration, B = second virial coefficient (104 cm3 mol g-2), and RH = hydrodynamic radius, evaluated from DLS measurements.

concentration on the formation and stability of stoichiometric (1:1) and nonstoichiometric (1:2) and (2:1) PECs generated with a brush polycation and a much smaller polyanion in aqueous medium. The properties of the generated PEC particles (in aqueous solutions) were determined by a combination of different methods, including static and dynamic light scattering, turbidity, and electrophoretic mobility techniques. Steric stabilization of the stoichiometric PEC particles due to the presence of PEO45 side chains in the polycation macromolecule was revealed. The effect of side-chain density of the brush polyelectrolyte on the stability of the aggregates is discussed.

Experimental Section Materials. Polymers. Commercially available poly(sodium

styrenesulfonate) (NaPSS) standard (Mw = 4300 g mol-1, Mw/Mn = 1.2) was used as polyanion. The material was purchased from Fluka and used as received. Poly(methacryloxyethyl trimethylammonium) chloride, poly(METAC), and its copolymer with poly(ethylene oxide) methyl ether methacrylate, PEO45MEMA, with different molar ratios were employed as polycations. Henceforth, PEO45MEMA:METAC-X represents the general abbreviation of these brush copolymers. The subscript 45 refers to the number of ethylene oxide units in the side chains, and X denotes the molar percentage of charged units in the copolymer macromolecule (X = 100, 75, 50, 25). The synthesis of the copolymers was performed by free-radical copolymerization, which resulted in close to random copolymers, having a polydispersity index of around 2-3, typical for polymers prepared by this method. Details of the synthetic procedure as well as the characterization of the materials have been reported elsewhere.40 Some physical characteristics of the polycations have been summarized in Table 1, and the molecular structures of the monomer units are schematically depicted in Figure 1. Sample Preparation. The water used in all experiments was first pretreated with a Milli-RO 10 Plus unit and then further purified with a Milli-Q PLUS 185 system. The resistivity after the treatment was 18.2 MΩ cm, and the total organic carbon content of the water did not exceed 2 ppb. Stock solutions of the polycations were prepared in 5 mM sodium chloride (Fluka) solution with concentration equal to 5000 ppm. NaPSS stock solutions were prepared in 5 mM NaCl in such a manner that the concentrations of the anionic charges from PSS were equal to the concentrations of cationic charges from the bottle-brush polyelectrolytes in their stock solutions. The polyelectrolyte stock solutions were vigorously stirred for ∼24 h and thereafter filtered employing a 0.1 μm inorganic membrane filters (Whatman, Anotop 25). Solutions of the cationic polyelectrolyte with concentrations of 200, 400, 800, 3200, and 4800 ppm were prepared by gradual dilution of the corresponding stock solutions using 5 mM NaCl aqueous solution. NaPSS solutions were prepared in the same manner. The sample preparation and the measurements were performed at room temperature (∼25 C). Mixing Procedure. The PECs were prepared by means of an automatic mixing process, using a programmable infusion pump Langmuir 2009, 25(11), 6113–6121

Figure 1. Schematic representation of the molecular structure of the monomer units in the polyelectrolytes used: (a) NaPSS, (b) METAC, and (c) PEO45MEMA. (model PHD200, Harvard Apparatus). The mixing is based on the simultaneous injection of the two polyelectrolyte solutions (2.5 cm3 from each) via small diameter tubes (d = 1 mm) into 5 cm3 of continuously stirred 5 mM NaCl solution. The applied infusion rate was 1 cm3/min. The final volume of the PEC containing solution was 10 cm3; thus, the contents of the cationic bottle-brush polyelecrolytes in the final mixtures were 50, 100, 200, 800, and 1200 ppm. Experimental Methods. Laser Light Scattering. The properties of the individual polyelectrolytes and the PECs were determined by means of laser light scattering. Static light scattering (SLS) and dynamic light scattering (DLS) measurements were conducted with a Brookhaven Instruments (USA) device, which consists of a BI-200SM goniometer and a BI9000AT digital autocorrelator. A water-cooled argon ion laser, Lexel 95 model 2, was used as light source. The laser was used at a wavelength of 514.5 nm and emitted vertically polarized light at a maximum power of 840 mW. In DLS mode the signal analyzer was used in a “multi-tau” mode; i.e., the time axis was logarithmically spaced to span the required correlation time range. The autocorrelation functions were measured at an angle of 90 in 218 channels using a 100 μm pinhole size. The measured autocorrelation functions were analyzed by the CONTIN and the second-order cumulant methods. The samples were found to be polydisperse, having a wide monomodal size distribution. Despite the polydispersity of the samples, the mean hydrodynamic diameter (the first cumulant of the second order cumulant expansion) was used for the presentation of the changes in the hydrodynamic size distribution. This can be done since neither the composition of the copolymer nor the complex formation changed significantly the character of the size distribution, as evidenced by the CONTIN analysis. For polydisperse systems, the first cumulant represents the z-average of the diffusion coefficient.41 At finite concentrations and q values, an apparent diffusion coefficient is obtained (Dapp); by using the Einstein-Stokes equation, the apparent hydrodynamic radius of the PEC particles could be determined RH ¼

kT 6πηDapp

where k is the Boltzmann constant, T the absolute temperature, η the viscosity of the medium, and RH the hydrodynamic radius. The autocorrelation functions were analyzed with the software supplied by Brookhaven Instruments. Thirty consecutive measurements were averaged for each PEC sample. In SLS mode the measurements were performed at 25 C at 17 different angles in the angular range 30 e θ e 155 and for five different polymer concentrations (prepared in aqueous (41) Brown, J. C.; Pusey, P. N.; Goodwin, J. W.; Ottewill, R. H. J. Phys. A: Math. Gen. 1975, 8, 664–682.

DOI: 10.1021/la804189w 6115

Article

Shovsky et al.

5 mM NaCl solution). The excess Rayleigh ratios (ΔR(θ,c)) were determined by standard procedures using toluene as the reference with the absolute Rayleigh ratio (Rtoluene= 24.2 cm-1 for vertically polarized light and vertically polarized analyzer42).
 Iðθ, cÞsolution -Iðθ, cÞsolvent nsolution 2 ΔRðθ, cÞ ¼ RðθÞtoluene Iðθ, cÞtoluene ntoluene ð1Þ where n is the refractive index and I denotes scattering intensities corrected for the detector dead time reflections and scattering volume. Static light scattering data are usually analyzed in terms of the classical Zimm equation, yielding the weight-averaged molecular weight (Mw), the z-mean of the square of the radius of gyration (R2g,z), and the second virial coefficient (A2) as the result of the analysis:
!
 16π2 n0 2 Rg 2 Kc 1 2 θ ¼ þ 2Bc 1þ sin ΔRðθ, cÞ Mw 2 3λ0 2

ð2Þ

where λ0 is the wavelength of the laser beam in vacuum, n0 is the refractive index of the solvent, and K is the optical constant given as K = 4π2n20(dn/dc)2/NAλ40, with dn/dc being the refractive index increment of the polymer (determined with an interferometric differential refractometer, Optilab DSP from Wyatt Technology), c the polymer concentration in mg/mL, and NA Avogadro’s number. As an alternative, the mean-squared radius of gyration ÆR2gæ can be determined by the Guinier equation (3) from the angular dependence of the scattering intensity measured at finite polymer concentrations: ln ΔRðqÞ ¼ ln ΔRðq0 Þ -q2

ÆRg 2 æ 3

ð3Þ

where q is the scattering vector defined as q = (4πn0/λ0) sin(θ/2). More detailed discussions on light scattering theory are available in several textbooks.43 Turbidimetry. The turbidity of the samples (τ) was determined by a Lambda 2 UV/vis spectrometer (Perkin-Elmer, UK). The measurements were done at λ = 400 nm, at room temperature using a 10 mm path-length cell. The turbidity values were determined directly after the sample preparation, and the measurements were repeated after 1 day and 2 weeks to investigate the stability of the formed complexes. The turbidity is determined from the attenuation of light passing through the sample: τ ¼ 2:303lg

I0 -1 l I

where I and I0 are the intensity of the incident and transmitted light, respectively, and l is the path length of the measuring cell. The turbidity of a highly diluted dispersion can be given as τ ¼ KMc

ð4Þ

where K is the optical constant defined in the section describing light scattering. It should be noted that K is proportional to the square of the optical contrast (dn/dc), which means that the turbidity of a sample can change due density changes of the dispersed particles. M is the weight (volume) of the dispersed particles, and c is the mass concentration of the dispersion. As indicated by eq 4, the turbidity can provide information on the aggregation of particles at a given concentration provided (42) Moreels, E.; Ceuninck, W. D.; Finsy, R. J. Chem. Phys. 1987, 86, 618–623. (43) Chu, B. Laser Light Scattering; Academic Press: Boston, 1991.

6116

DOI: 10.1021/la804189w

the optical contrast (density) of the particles does not change due to their aggregation. Electrophoretic Mobility Measurements. The electrophoretic mobility of the aggregates was measured using a Zetasizer 2000 device (Malvern Instruments, UK) in 5 mM NaCl solution at T = 25 C. The instrument was calibrated prior to measurements using Malvern Zeta Potential transfer standard (code DTS-1050). Finally we note that each point, in all plots/curves presented in this paper, is an entity that has been mixed and separately prepared and therefore is not the result of the continuous increase of the polyelectrolyte or PEC concentration in one particular sample.

Results and Discussion Characterization of the Brush Polyelectrolytes. The characteristics of the brush polyelectrolytes (weight-average molecular weight (Mw), z-average second virial coefficient (B), and z-average radius of gyration (Rg)) have previously been determined in 500 mM aqueous NaCl solution using Zimm plots.40 These results are summarized in Table 1. The B values obtained for the PEO45-rich copolymers are about a factor of 2-3 lower than those reported in the literature for linear poly (ethylene oxide) with the same molecular weight as the investigated brush copolymers.40 This is suggested to be a consequence of the different solution conformations of the linear and brush PEO.40 Linear PEO has a random coil solution structure, whereas PEO45MEMA:METAC-X adopts rodlike conformations in solution, as demonstrated by small-angle neutron scattering (SANS) and small-angle X-ray scattering (SAXS) measurements.44,45 In the current series of measurements the PECs were formed and characterized in 5 mM NaCl solution; we therefore also measured the radii of gyration of the brush polymers at this ionic strength using Guinier plots. The Rg values determined in 5 mM NaCl were found to be identical within the experimental error to the values determined previously in 500 mM salt solutions (Table 1). The salt-independent values of Rg are consistent with the stiff nature of these polymers. Similarly, the hydrodynamic radius of the PEO45MEMA:METAC-X copolymers, determined by DLS, were found to be independent of ionic strength in the measured NaCl concentration range of 5-500 mM. The CONTIN analyses of the DLS data (polyelectrolytes in 5 mM aqueous NaCl solution) showed a broad, monomodal size distribution. The value of the polydispersity index PDI = 0.25 (determined from the cumulant analyses) was found to be independent of the PEO45 content in the brush copolymers, indicating the polydisperse nature of all the samples. The hydrodynamic radii (Rh) as well as the characteristic ratios, F = Rg/RH, evaluated from the scattering data are summarized in Table 1. The Rg/RH ratio gives information about the global shape of the particles. Hydrodynamic theory46 shows that F changes from infinity to 0.775 when the polymer structure changes from a long rod to a sphere. For a random coil structure, F assumes values from 1.3 to 1.5,46 whereas F is expected to be >2 for rodlike particles. Thus, these data are consistent with a stiff rodlike conformation of the copolymers as suggested by SAXS and SANS data.44,45 (44) Bastardo, L. A.; Iruthayaraj, J.; Lundin, M.; Dedinaite, A.; Vareikis, A.; Makuska, R.; van der Wal, A.; Furo, I.; Garamus, V. M.; Claesson, P. M. J. Colloid Interface Sci. 2007, 312, 21. (45) Dedinaite, A.; Bastardo, L.; Oliveira, C. P.; Pedersen, J. S.; Claesson, P. M.; Vareikis, A.; Makuska, R. Proc. Baltic Polym. Symp. 2007, 112– 116. (46) Konishi, T.; Yoshizaki, T.; Yamakawa, H. Macromolecules 1991, 24 5614–5622.

Langmuir 2009, 25(11), 6113–6121

Shovsky et al.

Figure 2. Turbidity vs concentration of poly(METAC) mixed with NaPSS at different mixing ratios of cationic polyelectrolyte charges to anionic polyelectrolyte charges added to the solution: (9) (1:2) and (2) (2:1). The dotted curves are plotted as guides for the eye.

Characterization of the Polyelectrolyte Complexes. For the sake of simplicity, in the following paragraphs we will classify the complexes according to the PEO45 content: PEO45free is designated for poly(METAC)/NaPSS, PEO45-poor for PEO45MEMA:METAC-75/NaPSS, and PEO45-rich for PEO45MEMA:METAC-50/NaPSS and PEO45MEMA:METAC-25/ NaPSS. The complexes prepared upon mixing of the polyelectrolyte solutions with equivalent amounts of opposite charges will be further referred to as stoichiometric or electroneutral (1:1). The complexes prepared upon mixing of the polyelectrolyte solutions with double equivalent (amount) of one charge over the other charge will be designated as nonstoichiometric: (2:1) stands for the polycation-rich and (1:2) refers to for polyanion-rich complexes, respectively. Poly(METAC)/NaPSS. The preparation of nonstoichiometric complexes leads to the formation of turbid colloidal systems with suspended poly(METAC)/NaPSS particles. The solutions are turbid to the naked eye even at the lowest investigated polyelectrolyte concentration (50 ppm), in both the polycation-rich (2:1) and the polyanion-rich (1:2) cases. With increasing polyelectrolyte concentration the turbidity steeply increases (see Figure 2), and the samples become opaque. It is interesting to note that the polyanion-rich solutions (1:2) are always more turbid than the polycation-rich ones (2:1) at a given concentration (Figure 2). This is related to the larger conversion of the polycation to PECs, which gives rise either to larger PEC concentration or to formation of larger PEC particles or both. The colloidal stability of the nonstoichiometric PECs was probed by repeating the turbidity measurements after storage for 2 weeks. The measured turbidity values were found to be identical, within the experimental error, to those measured immediately after sample preparation. Since the poly(METAC) does not contain any hydrophilic side chains, it can be concluded that the formed nonstochiometric complexes are stabilized electrostatically by the excess polyelectrolyte. Consistently, it was found that the complexes could be precipitated by increasing the NaCl concentration to 100 mM. Mixing the polyelectrolyte solutions in a ratio that contains equivalent amounts of opposite charges (positively charged poly (METAC) and negatively charged NaPSS) leads to formation of a two-phase system of supernatant liquid and precipitated Langmuir 2009, 25(11), 6113–6121

Article

Figure 3. Turbidity vs concentration of PEO45MEMA:METAC75 mixed with NaPSS at different mixing ratios of cationic polyelectrolyte charges to anionic polyelectrolyte charges added to the solution: (9) (1:2), (O) (1:1), and (2) (2:1). The solid lines are plotted as guides for the eye.

stoichiometric PECs. The aggregation behavior of the (1:1) poly(METAC)/NaPSS was found to be very sensitive to the exact stoichiometry. A slight shift from (1:1) stoichiometry, either in positive or negative direction, leads to formation of turbid colloidal solutions with suspended poly(METAC)/ NaPSS particles stabilized by repulsive electrostatic doublelayer forces. The behavior of the nonstoichiometric and stoichiometric PEO45-free systems was expected and consistent with the results of previous studies of PEC formation for analogous linear polyelectrolytes.14 Thus, further characterizations of poly (METAC)/NaPSS complexes were not carried out. PEO45MEMA:METAC-75/NaPSS. Mixing PEO45MEMA:METAC-75 with NaPSS results in solutions that are less turbid compared to those obtained by mixing poly(METAC) and NaPSS. In the concentration range 50-200 ppm the PEO45MEMA:METAC-75/NaPSS solutions were found to be optically transparent to the naked eye, but nonzero turbidity is evident from the turbidity measurements. The turbidity increases linearly with polyelectrolyte concentration (Figure 3), which implies that the size of the formed complexes (M) does not change significantly with polymer concentration (c) (see eq 4). Dynamic light scattering measurements were also performed, and the CONTIN analysis of the measured autocorrelation functions indicated a broad, monomodal size distribution of the complexes with a polydispersity index (0.25) that was essentially the same as for the pure copolymer. This suggests that the PSS molecules are distributed uniformly among the brush polyelectrolytes. The RH values of the PEO45-poor complexes increase with polyelectrolyte concentration, as shown in Figure 4. This seems to contradict to the turbidity data, which implied a constant complex size with increasing polyelectrolyte concentration. This issue can be resolved if the individual compact complexes with increasing polyelectrolyte concentration becomes hydrodynamically coupled. In such a case the scattering intensity, and thus the turbidity, is determined by the compact core of the individual molecular complexes whereas the hydrodynamic size reflects that of the coupled cluster. For any given concentration the turbidity of the PEO45-poor complexes decreases in the order of (1:2) > (1:1) > (2:1) (Figure 3), and the hydrodynamic radius decreases in the same manner (Figure 4). The hydrodynamic size of the positively DOI: 10.1021/la804189w 6117

Article

Shovsky et al.

Figure 4. Hydrodynamic radius (RH) vs PEO45MEMA:METAC75 concentration (C) for PEO45MEMA:METAC-75/NaPSS at different mixing ratios of cationic polyelectrolyte charges to anionic polyelectrolyte charges added to the solution: (9) (1:2), (O) (1:1), and (2) (2:1). The curved lines are plotted as guides for the eye.

charged (2:1) and neutral (1:1) complexes remains unchanged during storage for 2 weeks, indicating formation of colloidally stable complexes. However, the size of the negatively charged (1:2) PECs steadily increases for several days due to aggregation until precipitation occurs (Figure 5). At this point the hydrodynamic size of the aggregates still present in the supernatant decreases significantly. A similar evolution with time has been observed for mixtures of cationic polyelectrolytes and anionic surfactants.47,48 The PEO45-poor PECs were also characterized by electrophoretic mobility measurements, as shown in Figure 6. A positive mean value of 1.25 ( μm cm /(V s)) was observed for the polycation-rich complexes (2:1), a slightly positive mobility was found for the stoichiometric complexes, and a significant negative mobility value (-1.1 μm cm /(V s)) was detected for the polyanion-rich systems. These electrophoretic mobility data are consistent with the change in mixing ratio, but they cannot explain the observed trends in aggregation behavior and long-term stability. We note that addition of NaCl to concentrations above 100 mM leads to precipitation at all three mixing ratios. This indicates that the stability of 2:1 and 1:1 complexes in 5 mM NaCl is due to a combination of electrostatic repulsion and steric repulsion imposed by the nonionic and hydrophilic side chains of the brush polyelectrolyte. This means that the 25% side-chain density is not sufficient to prevent aggregation of complexes during the complex formation stage, but in 5 mM NaCl it is large enough to, together with electrostatic repulsion, hinder the complexes from merging and forming large compact aggregate particles, indicating that the aggregate structure is different during the preparation process compared to after the preparation stage. In order to interpret the effect of mixing ratio on the turbidity, hydrodynamic size, and long-term stability, the details of the mixing protocol have to be considered. As described in the Experimental Section, the component polyelectrolytes are fed slowly and simultaneously into a continuously stirred NaCl solution. This means that in the case of the cationic polymer-rich :: (47) Naderi, A.; Claesson, P. M.; Bergstrom, M.; Dedinaite, A. Colloids Surf., A 2005, 253, 83. (48) Naderi, A.; Claesson, P. M. J. Dispersion Sci. Technol. 2005, 26, 329–340.

6118

DOI: 10.1021/la804189w

Figure 5. Hydrodynamic radius (RH) as a function of time for complexes formed PEO45MEMA:METAC-75 and NaPSS at different mixing ratios of cationic polyelectrolyte charges to anionic polyelectrolyte charges added to the solution: (9) (1:2), (O) (1:1), and (2) (2:1). The complex concentrations were (a) 100 ppm and (b) 800 ppm. The decrease in size observed after long time for the 1:2 complexes is a result of precipitation that only leaves small aggregates in the supernatant.

and the stoichiometric samples the added PSS molecules can react not only with the brush polyelectrolyte added at the same time to the solution but also with positively charged complexes that already are present in the mixture. If the association between an already formed complex and PSS leads to formation of an overcharged complex, then this complex can further react with a free brush polyelectrolyte or with an already formed positive complex leading to aggregate formation. The probability of this process increases with increasing polyelectrolyte concentration and with increasing NaPSS concentration, explaining why the stoichiometric mixtures show higher turbidity and larger hydrodynamic size than the samples rich in the cationic brush. If PSS is added in excess to the reaction mixture (1:2 complexes), then the formed complexes are negatively charged. This means that the probability that a freshly added cationic brush reacts with an overcharged complex is much larger than at the other two mixing ratios, which explains why most intensive aggregation occurs with excess NaPSS. Langmuir 2009, 25(11), 6113–6121

Shovsky et al.

Figure 6. Electrophoretic mobility vs PEO45MEMA:METAC-75 concentration (C) for PEO45MEMA:METAC-75/NaPSS at different mixing ratios of cationic polyelectrolyte charges to anionic polyelectrolyte charges added to the solution: (2) (2:1), (O) (1:1), and (9) (1:2). The horizontal lines represent the mean values.

The above argument explains the increasing extent of aggregation with increasing amount of PSS in term of PSS bridge formation between the brush polyelectrolytes during the PEC formation. However, Figure 5 indicates that the mixing ratio also has a strong influence on the long-term stability of the formed complexes. In principle, the lack of long-term stability can be explained by insufficient PEO brush density in the complexes to prevent slow aggregation. However, the most striking feature of Figures 5 and 6 is that while close to electroneutral complexes prepared by 1-1 mixing show longterm stability, the negatively charged 1-2 complexes formed in the presence of excess PSS slowly aggregate/precipitate. The fact the electroneutral complexes are stable on the investigated time scale indicates that the complexes have high enough PEO brush density to provide steric stabilization. Since the excess charge of the complex formed in the excess PSS should provide further electrostatic stabilization, one must conclude that some additional factors must lead to loss of stability. A possible interpretation of these observations can be given by considering the structure of the formed complexes. In the case of the positively charged and the stoichiometric complexes, the energetically most favorable complex structure is obtained when the whole PSS chain is located in the close vicinity of cationic backbone and thus buried among the PEO brushes (Figure 7a). On the other hand, in the presence of excess PSS, the system can gain entropy if only a fraction of the individual PSS chains interact with the cationic backbone, and the rest of the PSS chain extends from the complex core (Figure 7b). This facilitates bridging between complexes and thus promotes aggregation and finally precipitation (Figure 7c) during storage of the samples. To promote this process and avoid a large excess charge in one aggregate, it is an advantage if some PSS can leave the negatively charged aggregate. This is reasonable since it has been shown for highly aggregated polyelectrolyte complexes in the presence of a small amount of inert electrolyte that the short chain component of the PEC can exchange between the bulk and the complex due to local charge density fluctuations.13 PEO45MEMA:METAC-50/NaPSS and PEO45MEMA: METAC-25/NaPSS. The stoichiometric (1:1) and nonstoichiometric (2:1 and 1:2) mixing of PEO45MEMA:METAC-50 or PEO45MEMA:METAC-25 with NaPSS leads to formation of Langmuir 2009, 25(11), 6113–6121

Article

Figure 7. Schematic illustration of the PEO45MEMA:METAC-75/ NaPSS complex structure: black curve and sign (+) represent METAC-75, blue curve represents PEO45, and red curve represents PSS; (a) PSS is buried among the PEO brushes, (b) extended PSS chains from complex core, and (c) PSS chains bridging between complexes.

optically transparent solutions. An increase in the polyelectrolyte concentration showed no effect on the turbidity. Further, addition of NaCl up to a concentration of 1 M did not cause any increase in turbidity or precipitation. This means that the fractions of 50% and 75% of neutral PEO45 side chains in the brush copolymer is either sufficient to achieve complete steric stabilization of the complex particles and inhibit their further aggregation to larger particles or that it prevents complex formation. To distinguish these possibilities, static and dynamic light scattering as well as electrophoretic mobility measurements were performed. In Figure 8 the Guinier plots of the stoichiometric PEO45-rich complexes can be compared with those for the pure polymers. We note that the slopes in presence of NaPSS are very similar to those observed for the pure brush polyelectrolytes. Thus, the addition of NaPSS to the brush polyelectrolytes has only a minor effect on the radii of gyration. The radii of gyration determined in presence of NaPSS (35 and 38 nm for solutions containing the 50% and 25% charge density polymers, respectively) are slightly smaller than the ones calculated for the corresponding brush polymers in the absence of NaPSS (see Table 1). On the other hand, the scattering intensities of the solutions containing both the cationic brush polyelectrolyte and NaPSS are much larger than those of the corresponding solutions without NaPSS. Since these measurements were done at identical cationic polyelectrolyte concentrations (the small NaPSS alone does not contribute significantly to the scattering) and the radii of gyration do not change significantly upon addition of NaPSS, the increased scattering intensities must reflect an increased scattering contrast (dn/dc), which confirms formation of polyelectrolyte complexes. The slight decrease of the radii of gyration on complex formation is consistent with the assumption that the NaPSS polymer is incorporated into the core of the brush, in close vicinity of the positively charged backbone. This structure can explain not only the decreasing radius of gyration but also the practically unchanged hydrodynamic size (see below) and the increased scattering intensity that is related to the higher optical contrast of the core. DOI: 10.1021/la804189w 6119

Article

Shovsky et al.

Figure 8. Guinier plots for PEO45-rich brush copolymers in the absence (solid symbols) and presence of NaPSS (open symbols). The solid lines represent least-squares fits to the data points.

Figure 9. Electrophoretic mobility vs PEO45MEMA:METAC-50 concentration (C) for PEO45MEMA:METAC-50/NaPSS at different mixing ratios of cationic polyelectrolyte charges to anionic polyelectrolyte charges added to the solution: (2) (2:1), (O) (1:1), and (9) (1:2). The horizontal lines represent the mean values.

An additional evidence of the complex formation in the PEO45-rich systems is provided by electrophoretic mobility measurements. As shown in Figures 9 and 10, the mobility of the stoichiometric complexes is as expected essentially zero. Furthermore, the electrophoretic mobility was found to be positive when a 2:1 mixing ratio was used (polycation-rich systems), whereas it was negative when the applied mixing ratio was 1:2 (polyanion-rich solutions). The mobility was found to be concentration independent for all mixing ratios. The hydrodynamic radii of the PEO45-rich complexes are plotted as a function of polyelectrolyte concentration in Figures 11 and 12. The size of the formed complexes is concentration independent for both PEO45-rich systems. The RH values observed in the case of (2:1) mixing ratio is practically equal to the hydrodynamic size of the pure brush polymer, supporting the suggestion that the PSS chain is located close to the backbone of the cationic brush polyelectrolyte. The increase in PSS concentration to achieve a mixing ratio of 1:1 has only a minor effect on the hydrodynamic size; the observed increase is comparable to the error of the DLS measurements. However, further increase in PSS content to achieve 6120

DOI: 10.1021/la804189w

Figure 10. Electrophoretic mobility vs PEO45MEMA:METAC25 concentration (C) for PEO45MEMA:METAC-25/NaPSS at different mixing ratios of cationic polyelectrolyte charges to anionic polyelectrolyte charges added to the solution: (2) (2:1), (O) (1:1), and (9) (1:2). The horizontal lines represent the mean values.

Figure 11. Hydrodynamic radius (RH) vs PEO45MEMA:METAC-50 concentration (C) for PEO45MEMA:METAC-50/NaPSS at different mixing ratios of cationic polyelectrolyte charges to anionic polyelectrolyte charges added to the solution: (9) (1:2), (O) (1:1), and (2) (2:1). The horizontal lines represent the mean values.

a mixing ratio of 1:2 (anionic polyelectrolyte-rich solutions) significantly increases the hydrodynamic size of the complexes (by 30%), which is assigned to some extended PSS chains. Finally, it should be emphasized that these complexes were stable for an extended period of time (at least several months) independent of the mixing ratios. Clearly, the high side-chain density provides an efficient steric barrier that prevents flocculation. In conclusion, the relationship between PEC properties and polyelectrolyte architecture is considered. By comparing similarities and differences in properties of the investigated molecular bottle-brush complexes to previously reported contributions on PECs formed by pairs of charged (i) linear/linear polyelectrolytes,49 (ii) linear polyelectrolyte/ionic block copolymer,37,50 and (iii) ionic block copolymer/ionic block copolymer.11,51 (49) Webster, L.; Huglin, M. B.; Robb, I. D. Polymer 1997, 38, 1373. (50) Chelushkin, P. S.; Lysenko, E. A.; Bronich, T. K.; Eisenberg, A.; Kabanov, V. A.; Kabanov, A. V. J. Phys. Chem. B 2007, 111, 8419–8425. (51) Voets, I. K.; deKeizer, A.; CohenStuart, M. A.; Justynska, J.; Schlaad, H. Macromolecules 2007, 40, 2158–2164.

Langmuir 2009, 25(11), 6113–6121

Shovsky et al.

Article

of the charged neutralized polyelectrolyte complex, and it is surrounded by a hydrophilic corona of the nonionic blocks, similarly to that of bottle-brush PECs. However, the differences is that bottle-brush PECs represent rodlike objects, composed of only one cationic polyelectrolyte, when the charge density, X, of the PEO45MEMA:METAC-X/NaPSS bottle brush is 25 e X < 75.

Figure 12. Hydrodynamic radius (RH) vs PEO45MEMA:METAC-25 concentration (C) for PEO45MEMA:METAC-25/NaPSS at different mixing ratios of cationic polyelectrolyte charges to anionic polyelectrolyte charges added to the solution: (9) (1:2), (O) (1:1), and (2) (2:1). The horizontal lines represent the mean values.

(i) PECs composed of linear polyelectrolytes of opposite charge without hydrophilic side chains tend to either precipitate (stoichiometric) or form large charge stabilized aggregates (nonstoichiometric) in aqueous solution.49 Similar properties have been revealed for the bottle-brush PEO45MEMA:METAC-75/NaPSS complexes, when the amount of PEO45 side chains was 25%. Thus, this side-chain density is insufficient to prevent aggregation into multichain complexes during the PEC formation stage. (ii) PECs formed by one large linear polyelectrolyte and smaller ionic block copolymers can be characterized as sterically stabilized cylindrical brushes of core-shell morphology.38 The core of this complex consists of neutralized ionic chains, and the shell is composed of the neutral hydrophilic blocks. The structures of these nanoparticles are similar to those of the sterically stable molecular complexes reported in this contribution, namely PEO45MEMA:METAC-50/NaPSS and PEO45MEMA:METAC-25/NaPSS. (iii) PECs formed by assembly of block ionogenic copolymers (BICs)52 exhibit water-soluble stoichiometric complexes. Generally, they are spherical in shape and consist of several molecules of each component; hence, the size ranges from several tens to hundred nanometers.51 The core of the micelle consists (52) Oh, K. T.; Bronich, T. K.; Bromberg, L.; Hatton, T. A.; Kabanov, A. V. J. Controlled Release 2006, 115, 9–17.

Langmuir 2009, 25(11), 6113–6121

Conclusions The influence of PEO45 content in brush polymers on the formation and stability of PEC nanoparticles of shoichiometric and nonstoichiometric compositions has been investigated. Cationic brush copolymers and low molecular weight PSS were used for the complex formation. In agreement with the literature, in the absence of PEO45 side chains the 100% charged polymer (polyMETAC) formed insoluble PECs with PSS that precipitated from the solution when exact stoichiometry was achieved. In nonstoichiometric complexes (1:2) and (2:1) large colloidally stable aggregates were formed that could be precipitated by addition of salt. The presence of even a relatively small amount of PEO45 side chains (25%) in the cationic copolymer (PEO45MEMA:METAC75) was sufficient for preventing precipitation of the complexes in 5 mM NaCl. In this case turbid and stable (2:1 and 1:1 complex) or slowly flocculating (1:2 complexes) colloidal dispersions were formed, but also in this case flocculation was induced by salt addition which demonstrates the importance of both electrostatic and steric forces. The slow aggregation in 5 mM NaCl and excess PSS (1:2 complexes) is suggested to be due to bridging of PSS chains between aggregates, followed by expulsion of some of the PSS chains from the aggregates. Further increasing the PEO45 side-chain content (50 and 75%) of the cationic copolymer resulted in formation of small, watersoluble molecular complexes. The hydrodynamic size of the aggregates decreases in the order (1:2) > (1:1) ≈ (2:1), independent of concentration. It is suggested that PSS present in cationic and uncharged PECs are located close to the cationic backbone of the larger brush polyelectrolyte. However, in the negatively charged PECs a fraction of the PSS chains are extended away from the backbone, explaining the increased hydrodynamic size. Acknowledgment. The work has been supported by the European Commission in the sixth Framework Programme under The Marie Curie Intra-European Fellowships, Contract No. MEIF-CT-2006-024997, Project PE-NANOSTRUCTURES. A.S. and P.C. acknowledge financial support from the Swedish Research Council, VR.

DOI: 10.1021/la804189w 6121