Control of Corona Composition and Morphology in Aggregates of

Apr 13, 2014 - The symbols for the morphologies are given on the left side of each ..... symbols) and corona composition (expressed by the color codes...
0 downloads 0 Views 6MB Size
Download PDF
Article pubs.acs.org/Langmuir

Control of Corona Composition and Morphology in Aggregates of Mixtures of PS‑b‑PAA and PS‑b‑P4VP Diblock Copolymers: Effects of pH and Block Length Renata Vyhnalkova,† Axel H. E. Müller,‡ and Adi Eisenberg*,† †

Department of Chemistry, McGill University, Otto Maass Building, 801 Sherbrooke Street West, Montréal, Québec, Canada H3A 2K6 ‡ Institut für Organische Chemie, Johannes Gutenberg-Universität Mainz, 55099 Mainz, Germany S Supporting Information *

ABSTRACT: The corona compositions and morphologies in aggregates of mixtures of amphiphilic polystyrene-blockpoly(acrylic acid) (PS-b-PAA) and polystyrene-block-poly(4vinylpyridine) (PS-b-P4VP) diblock copolymers are influenced by controllable assembly parameters such as the hydrophilic block length and solution pH. The morphologies and corona compositions of the aggregates were investigated by transmission electron microscopy and electrophoretic mobility, respectively. When mineral acids or bases are present during aggregate formation, they can exert a strong influence on the corona composition. Morphology changes were also seen with changing pH, as well as changes in corona composition, specifically for vesicles. Because of complications introduced by the presence of ions, the general hypothesis that the external corona of the vesicles is composed of the longer chains, while the shorter chains form the inner corona, which is valid only in mixtures containing only nonionic chains without any additives (no acids or bases) or within a well-defined narrow pH range. In addition to the numerical block lengths and the pH, the solubility of the hydrophilic blocks can also influence the morphology and as well as the interfacial composition of vesicles; as the numerically longer chains become less soluble, they can contract and move to the interior, while the numerically shorter but more soluble chains go to the external corona. A remarkable morphological feature of the pH continuum is that for some compositions vesicles are observed in four distinct pH regions, separated by pH ranges in which other morphologies dominate. The effect of pH and microion content on coil dimensions of the PVP and PAA chains in the block copolymers is most likely responsible for the observed behavior.

1. INTRODUCTION

Block copolymer vesicles and other aggregates have many potential applications in such fields as catalysis, pharmacology, and cosmetics. For some applications, an understanding of the equilibrium nature of vesicles and control of vesicle parameters would be very useful. In general, the factors influencing the properties of liposomes (i.e., small-molecule vesicles) have received considerable attention.24,25 Vesicles prepared from block copolymers, on the other hand, have been studied much less extensively, especially with regard to their equilibrium status. It was shown previously4,26 that vesicles are equilibrium structures under a range of conditions since their sizes and many of their other properties can be changed reversibly by changing the composition of the solvent mixture or other parameters of the solution in which they are prepared. A thermodynamic curvature-stabilization mechanism for block copolymer vesicles has been proposed.27 Stabilization

Asymmetric amphiphilic block copolymers can self-assemble in selective solvents to give aggregates with a wide range of morphologies. Since 1995, when the first report dealing with “crew cut” aggregates appeared, 1 a large number of morphologies were identified and investigated by many groups.2−23 Not only small spherical micelles (SSMs), rods, vesicles, and bicontinuous structures as well as inverted structures were observed but also a range of mixed, combined, and much more complex aggregates, with all of them containing a phase-separated insoluble core or wall and a soluble corona. Thermodynamically, the parameters which influence the morphology are chain stretching in the core, corona repulsion, and interfacial energy,2,9 and it is the interplay of these factors which determines the final morphology. Experimentally accessible morphogenetics include, among others, polymer block lengths and composition, nature of the common solvent, initial polymer concentration, type and concentration of added ions, method of preparation, presence of homopolymers, and polydispersity.4−6,9 © 2014 American Chemical Society

Received: February 21, 2014 Revised: April 3, 2014 Published: April 13, 2014 5031

dx.doi.org/10.1021/la500712b | Langmuir 2014, 30, 5031−5040

Langmuir

Article

at high pH. Surprisingly, at extreme pH (prep) values, the opposite behavior is observed in that at high pH (prep) the external vesicular corona was composed of P4VP chains while at low pH (prep) the PAA chains were forming the external corona. It was also expected, based on previous studies,26,27 that in mixtures of block copolymers in the absence of polyelectrolyte (PE) effects the corona of the vesicles would be composed of the longer chains while the shorter ones would occupy the inner corona. However, contrary to what has been noted in previous studies, it was found that in some cases the numerically longer chains can be made to contract and go to the inside while the numerically shorter chain goes to the outside. Morphological changes are frequently induced by very small changes in the environment.3−6,9 The present work explores the morphological consequences of changes in factors such as block length, pH, coil dimension, and salt concentration. Even a change in only the pH, which changes the coil dimensions, can have a major effect on the morphology. As an example of the interplay of effects, one should note that the changes in pH vary the coil dimensions, which, in turn, modify the polymer concentration in the region close to the interface. Therefore, it should not come as a surprise that a systematic variation in the pH changes the relative importance of several morphogenic parameters so that the system can undergo multiple morphological changes. Frequently, the specific trigger for morphological transformations is impossible to identify among the many contributing factors. In this work, only the unambiguous ones will be described. While the present system is extremely complicated with regard to both morphology and corona composition, it is worth exploring because it illustrates very clearly the complex interactions between the PE effect and the effects of block length and solubility (especially in vesicles) as they influence the corona composition and morphology. In some ranges, very small changes in one of the variables can have major consequences, while in other regions the morphology and corona compositions are relatively insensitive. The reasons for such behavior are not always clear, and the elucidation is in the early stages so that only ex post facto and not a priori explanations can be offered at this time. The present study underscores the extreme complexity of the behavior and the complex nature of the interplay among block length, degree of ionic character, and solubility on the morphology and corona composition and illustrates the extreme sensitivity of the system to the external stimuli under some circumstances.

consists of preferential segregation of the shorter hydrophilic corona chains to the inside of the vesicles and of the longer chains to the outside and was proven by the use of fluorescently labeled diblock copolymers. Segregation increases the corona repulsion on the outside of the vesicle relative to that on the inside and provides thermodynamic stabilization of the curvature. It was also found that the degree of segregation is dependent on the size of the vesicles and is reversible in response to changes in size.26 On the basis of the previously mentioned observations, Luo and Eisenberg27 explored the segregation of a mixture of polystyrene diblock copolymers containing both PS-b-PAA and PS-b-P4VP. Although one of these blocks is cationic and the other one is anionic and would thus be expected to interact strongly with each other,29−31 it was shown that they can be preferentially segregated, one to the inside and the other to the outside of the vesicles, respectively, provided that the different hydrophilic blocks are of different lengths. One aspect of the evidence of segregation consisted of a comparison of the ζ potential of vesicles of only PS310-b-P4VP33 or of only PS310-bPAA44 to that of vesicles prepared from mixed copolymers of PS310-b-P4VP33 with PS310-b-PAA11. The ζ potential of the mixed systems was identical with that of vesicles prepared from pure PS-b-P4VP, proving that the longer P4VP chains were on the outside corona.26 Liu and Eisenberg28 also studied the morphologies of bisamphiphilic triblock PAA26-b-PS890-b-P4VP40 terpolymers as a function of pH in DMF/THF/H2O mixtures; the corona composition on the external interface of the aggregates was characterized by ζ-potential measurements. Starting at pH (prep) 1 and increasing gradually to pH (prep) 14, the aggregate morphologies change progressively from vesicles (pH (prep) 1), to solid spherical or rod-shaped aggregates (pH (prep) 3−11), and back to vesicles (pH (prep) 14). Vesicles prepared at pH (prep) 1 contained poly(vinylpyridinium) P4VP chains on the outside and poly(acrylic acid) PAA chains on the inside, while those prepared from the same triblock at pH (prep) 14 contained PAA outside and P4VP inside. The segregation is based on the difference in repulsive interactions within the PAA or P4VP corona under different pH conditions. The authors28 concluded that at low pH (prep) the curvature is stabilized through increased repulsive interactions between the protonated ionic P4VP chains on the outside relative to the less repulsive interactions between the protonated, nonionic PAA chains on the inside. At pH (prep) 14, by contrast, PAA is preferentially segregated to the outside. In parallel to the work with triblock terpolymers,28 the goal of the present study is to expand the understanding of the relative importance of the factors that govern the balance of forces, which determine the inside and outside interface composition in vesicles of block copolymer mixtures (i.e., pH (prep) vs block length). In the present work, in the case of mixtures of PS-b-PAA and PS-b-P4VP block copolymers, either the PAA or the P4VP or a mixture of the two can be directed to the outside if the vesicles are prepared under appropriate conditions; the interface compositions are dependent on relative coil dimensions under assembly conditions. On the basis of previous work27 in which the block lengths were approximately the same, at low pH (prep) the corona was expected to consist of P4VP, and as the pH (prep) of the system increases, one would expect the PAA to form the outside corona because of ionization effects accompanying the protonation of P4VP at low pH and the deprotonation of PAA

2. EXPERIMENTAL SECTION Diblock copolymers of PS-b-PAA and PS-b-P4VP of identical number ratio (1:1) of PAA to P4VP chains were first dissolved in DMF, with an initial concentration of each copolymer of 0.5 wt %. The total initial concentration of the copolymers was always kept constant (1 wt %), as was the rate of water addition (2 wt %/min). Inspired by the previous works and the fact that DMF is a good solvent in which to prepare vesicles from PS-b-PAA and PS-b-P4VP block copolymers, DMF was chosen as a common solvent in this study. DMF was also extensively used as a solvent for PS-b-PAA and PS-b-P4VP in number of other works from our group. The samples were stirred overnight to ensure thorough mixing. For samples involving the addition of acid or base (HCl or NaOH), a calculated amount needed to reach a given molar ratio, R, of HCl/4VP or NaOH/AA was added to the copolymer solutions. The solutions containing additives (HCl or NaOH) were again mixed overnight to reach equilibrium. Deionized water was then 5032

dx.doi.org/10.1021/la500712b | Langmuir 2014, 30, 5031−5040

Langmuir

Article

Figure 1. Schematic representation, symbols, and TEM micrographs of the aggregate morphologies found in the mixtures of PS-b-PAA and PS-bP4VP: (A) small spherical micelles (SSMs), (B) rodlike micelles, (C) vesicles, (D) large compound micelles (LCMs), (E) linked elongated vesicles, and (F) large compound vesicles (LCVs). The symbols for the morphologies are given on the left side of each micrograph.

Figure 2. Schematic representation of mixtures of PS-b-PAA and PS-b-P4VP diblock copolymers: (A) diblock copolymers prior to mixing; (B) diblock copolymers after mixing in DMF, indicating the formation of the IPEC, which is collapsed; (C) self-assembly of SSM after water addition indicating the collapsed PS core and the collapsed and swollen IPEC region; and (D) dry state of the SSM. added dropwise with stirring to the copolymer solutions at a rate of 2.0 wt %/min until the water content reached 50 wt %. After measuring the pH (prep) (apparent pH, value as measured for the mixed solvent system in which the aggregates were prepared), we quenched the samples in a 6-fold larger amount of water to freeze the morphology. The samples were then dialyzed against Milli-Q water at pH 3 to 4 for 3 days to keep the colloid solutions from precipitating and to remove all of the solvent from the solutions. At such high water content, the aggregates do not undergo any morphological changes during dialysis. Further experimental details are given in the Supporting Information (SI).

These include the morphologies shown in Figure 1, where they are given in the form of a schematic representation, a symbol, and a sample TEM micrograph for each of the morphologies. In relation to the elongated vesicles, schematically shown in Figure 1 E, it should be noted that one can distinguish two types of elongated vesicles which are frequently seen together in the same micrograph. These are sequentially linked vesicles or tubular vesicles, which are sometimes linked to spherical vesicles in one unit. Both of these types of structures are referred to as elongated vesicles in the present study. Figure 2 gives a schematic representation of one of the obtained morphologies, the SSMs. The figure shows the formation of an interpolyelectrolyte complex (IPEC)3,29,30 in DMF (2B). Figure 2C indicates self-assembly in water, showing the collapsed PS core as well as the IPEC, which now is collapsed but still swollen (in green), and the excess PAA in the

3. RESULTS AND DISCUSSION 3.1. Morphologies Observed in Mixtures of PS-b-PAA and PS-b-P4VP Block Copolymers. Various morphologies were observed in the present study of mixtures of PS-b-PAA and PS-b-P4VP block copolymers at various pH (prep) values. 5033

dx.doi.org/10.1021/la500712b | Langmuir 2014, 30, 5031−5040

Langmuir

Article

solvent (blue segment outside of the green region). Finally, on drying, the PS core is surrounded by a shell of IPEC, which is further surrounded by a shell of excess PAA (or P4VP if that is present in excess). The most common morphologies found in the mixture of two block copolymers include SSMs and vesicles. The corona composition for each of the observed morphologies was examined by ζ-potential measurements (for vesicles, only the external interfacial corona was studied) as described in the SI section. The aggregates identified as SSMs in this blended system are somewhat larger than SSMs prepared from unmixed AB diblocks with a hydrophobic length of around 300 styrene units. The SSMs (see schematic representation in Figures 1A and 2) fall in the range of 30 to 45 nm in diameter, while SSMs composed of unblended diblocks based on a PS of around 300 units are ca. 25 nm in diameter with small variations depending on the hydrophilic block length. The reason for the increase in size is most probably due to the formation of paired strands of P4VP and PAA as shown schematically in Figure 2. These paired strands are expected to be hydrophobic, in parallel to the hydrophobicity of IPECs. Therefore, in the blend systems, the hydrophobic paired strands would be expected to aggregate further to form a hydrophobic shell around the core because of their expected immiscibility with PS and thus to enlarge the observed hydrophobic core size relative to that of the unblended PS core micelles. The expected size of the shell can be estimated crudely by considering that the paired strand of P4VP and PAA forms a hydrophobic ladder pair which is as long as the length of the shorter chain in the combination. For example, in the case of the P4VP12 and PAA27 mixture, the hydrophobic pair will be 12 paired units long, with the remaining 15 units of PAA providing the soluble segment. A 12-unit paired strand would be expected to have a maximum length close to that of a fully stretched planar zigzag (i.e., close to 3 nm (12 × 0.25 nm)). If the paired strands were to assume a random coil configuration, which is unlikely for a ladderlike pair, the dimensions would shrink to 0.9 nm (120.5 × 0.25 nm). The typical micellar dimensions based on unblended diblocks containing 300 styrene units are around 25 nm. If the paired strands were to assume their maximum planar zigzag length of 3 nm, then the increase in the core diameter would be 6 nm (2 × 3 nm). However, a planar zigzag conformation, even for a ladder, is not likely. If the shape of the ladder segment were that of a random coil, the end-to-end distance would shrink to 1.8 nm (2 × 0.9 nm). The actual increase in the core dimensions is likely to be somewhere between those two values but probably closer to that of the planar zigzag then that of the random coil. 3.2. Solubility of PAANa and P4VPCl in DMF/Water Mixtures. The details of the solubility behavior of the two key materials (i.e., PAANa and P4VPCl) in mixtures of DMF and water are crucial to our understanding of the self-assembly and corona composition. For this reason, detailed studies were performed on the absorbance versus composition behavior of PAANa and P4VPCl dissolved in either DMF or water as a function of the added amount of the other solvent component. Figure 3 illustrates the effect of water content on the absorbance of poly(sodium acrylate)−PAANa (for an equimolar ratio of acid to sodium). The sodium salt of poly(acrylic acid) was prepared from the homopolymer of Mw = 2000 (nmax = 28). In the first part of the experiment, the homopolymer of PAA (0.8 wt %) was dissolved in DMF, an equimolar amount of

Figure 3. Absorbance of PAANa in water and in DMF.

NaOH was added, and the absorbance was monitored (point A in Figure 3). It is worth recalling that PAA is soluble in DMF but PAANa is insoluble (see Table 2 SI in SI). The intermediate absorbance (0.95) at point A probably reflects the insolubility of NaOH and whatever PAANa may be formed. As the water content increases (while the DMF content decreases), NaOH becomes more soluble and more PAANa can be formed. However, PAANa is insoluble in DMF; therefore, as more PAANa is formed, the solution becomes increasingly turbid. At point B (17 wt % of water), the absorbance has reached a maximum value reflecting the increasing but still soluble amount of PAANa. As the water content increases toward and through point C, the solubility of PAANa improves continuously and the absorbance decreases. At a water content of about 48 wt %, PAANa becomes completely soluble and the solution is transparent. In the second part of the experiment, the homopolymer of PAA (0.8 wt %) was dissolved in water and an equimolar amount of NaOH was added. A very low absorbance was observed at point K due to the solubility of PAANa in water (see Table 2 SI in the SI section). The absorbance remains unchanged up to point L (58 wt % water−42 wt % DMF content). At point L, the solubility limit is reached for that particular molecular weight and polymer concentration. As the water concentration decreases further, toward and through point M, the solubility of PAANa decreases and absorbance increases. The absorbance behavior of an aqueous solution of PAANa is consistent with the results of the solubility study of PAANa in DMF as well as with the behavior of the various block copolymer combinations of PAA in DMF, all of which are included in Table 3 SI of the SI; also included is a detailed explanation of the tabulated results. A solubility study of the homopolymers, copolymers, and their salts in water and DMF was performed to understand the behavior of the mixtures in the extreme pH regions. It is reasonable to expect that the ionized PAA should stay inside the aggregate because of its insolubility, while the soluble P4VP forms the corona. As illustrated in Figure 3, there is a hysteresis loop indicated by the light-gray area. This loop is possibly related to kinetic factors and/or particle size effects. The results of the effect of water content on the absorbance of poly(vinylpyridinium chloride) P4VPCl (molar ratio of HCl/4VP = 1) are described in Figure 4. P4VPCl was prepared from a homopolymer of P4VP of Mw = 2700 (nmax = 26). 5034

dx.doi.org/10.1021/la500712b | Langmuir 2014, 30, 5031−5040

Langmuir

Article

the water content increases toward point C, the solubility improves and the absorbance decreases. At a water content of 5 wt % (point C), the sample completely dissolves and remains soluble at all higher water contents. In the second part of the experiment, the homopolymer of P4VP (0.8 wt %) was dissolved in water and an equimolar amount of HCl was added. Very low absorbance was measured in pure water due to the solubility of the P4VPCl (solubility Table 3 SI in the SI). The absorbance remained unchanged up to point C (95.0 wt % DMF). The behavior of P4VPCl in DMF and water, as discussed above, is consistent with the insolubility of P4VPCl in DMF and its solubility in water; these results are described in Table 4 SI in the SI, along with those of the solubility study of PS310-bP4VP33, of P4VP homopolymer, and of P4VP+Cl − in water and DMF. 3.3. Morphologies and Corona Compositions in a Mixture of PS300-b-PAA11 and PS310-b-P4VP 12. We investigated nine different mixtures of PS-b-PAA and PS-bP4VP with different lengths. However, for brevity, only one of the nine combinations was selected for a detailed discussion here because it has the overall features which characterize the system. The detailed results for the remaining eight combinations are given in the SI. The chosen combination represents mixtures of short PAA and short P4VP chains (upper right corner of the composition diagram in the left upper corner of Figure 5). The overall conclusions of the study,

Figure 4. Absorbance of P4VP+Cl− in water and in DMF.

In the first part of the experiment, the P4VP homopolymer (0.8 wt %) was dissolved in DMF, an equimolar amount of HCl was added, and the absorbance was monitored (point A in Figure 4 (point A was measured experimentally)). P4VP is soluble in DMF, but P4VPCl is insoluble (see Table 3 SI in the SI). The absorbance at point A (2.7) reflects the insolubility of P4VPCl in DMF. More P4VPCl is formed with increasing water content; therefore, the solution becomes increasingly turbid as one moves toward point B. Point B (at about 2 wt % water) reflects the same phenomena as point B in Figure 4. As

Figure 5. Schematic representation of the pH (prep) effect on the morphology and corona composition for mixtures of PS313-b-PAA11 and PS310-bP4VP12 in DMF/water (1:1), with an AA/4VP ratio of 1.048. The ζ-potential measurements and TEM images are taken after quenching from DMF/ water (1:1) (w/w) to pure water. 5035

dx.doi.org/10.1021/la500712b | Langmuir 2014, 30, 5031−5040

Langmuir

Article

results of Luo and Eisenberg,26,27 one would expect that at very low pH (prep) (i.e., at high HCl/4VP ratios, at a large excess of HCl) the vinylpyridine should be quaternized, while the acrylic acid remains protonated and is therefore nonionic. Since the vinylpyridine would thus be a PE, its coil dimensions would presumably be larger than those of the PAA of a similar degree of polymerization; the P4VP should therefore be outside in any vesicles that are formed. The possibility of the screening of the charges by excess HCl should also be mentioned. However, it should be borne in mind that the entire solution is accessible to the HCl microions, not just the immediate vicinity of the chains. Therefore, the screening effect is not expected to be large. As can be seen clearly (bottom rows of Figure 5), according to the ζ-potential results, the P4VP is not on the outside. Some other effect must, therefore, be operative. The reversal of the expected corona composition can be understood in the following way: A mixture of PAA and HCl is soluble in DMF (see Table 3 SI). Low concentrations of HCl would not be expected to perturb the PAA coil dimensions appreciably. On the other hand, the addition of HCl to P4VP yields quaternized poly(vinylpyridinium) chloride, which is not soluble in DMF (Figure 4) but eventually becomes soluble as one adds water (at approximately 12% (w/w)). The coil dimensions would be expected to increase as the water concentration increases. At the water concentration at which vesicles form, the PAA coil dimensions must have been larger than those of the quaternized poly(vinylpyridinium) chloride because the latter is barely soluble in DMF at these low water concentrations and because the outside corona consists mostly of PAA chains. Eventually, when enough water has been added, the quaternized poly(vinylpyridinium) chloride becomes highly soluble and the coil dimensions become “normal”, but by then the vesicles are already formed with mostly PAA chains in the external corona. While the quaternized poly(vinylpyridinium) chloride is not soluble in DMF, it is clearly not so insoluble as to form cross-links and thus give inverse micelles; therefore, vesicles are formed. According to the ζ-potential results, the corona of the vesicles is composed of mostly PAA, implying that P4VP must form the inner corona of the vesicles. It should be recalled that the wall of the vesicles is composed of PS chains, which become morphologically frozen after quenching in pure water. The case of greatest excess for NaOH (R = 3.0, micrograph J) is discussed next. The mixture of PAA and NaOH (i.e., PAANa) is not soluble in DMF (Table 3 SI). Therefore, regardless of which aggregate is formed, P4VP will be on the outside due to its solubility in DMF. The ionized PAA stays inside or at the interface of the aggregate. In this case, however, because of the extreme insolubility of the sodium polyacrylate in DMF, reverse micelles, presumably with PAANa in the core, are the most stable aggregate morphology formed. Upon water addition, the reverse micelles aggregate to form LCMs, the cores of which are composed of the aggregated PAANa chains within each reverse micelle. According to the ζ-potential results, the corona of the LCMs is composed of P4VP chains, which are, even in the presence of NaOH, soluble enough in DMF and DMF/water mixtures to remain on the outside of the aggregate. The region in which the corona composition changes from PAA to a mixture of PAA and P4VP, or even to solely P4VP, will be referred to as the transition region. According to Figure 5, the transition region for the present system, as deduced from

however, are based on the results for all nine block copolymer combinations. An example of a data table of the type obtained for each of the nine rectangles of Table 1 SI in the Supporting Information is illustrated in Figure 5. It contains an overview of the results of the first example of the study of the pH (prep) effect on the morphology and the composition of the external surfaces of the aggregates in DMF/water (1:1). The first line of the figure, above the micrographs, specifies the experimental parameter, the effect of which is being explored on the morphology and the external corona composition. The upper left-hand corner of Figure 5 (to the left of the micrographs) contains the schematic reminder (see SI), in which the cell representing the specific mixture under study is shaded in gray. The upper part of Figure 5 shows a small section of each of the micrographs (marked with letters A to J and in some figures to O), obtained after quenching the samples from a solution of 50% (w/w) water and subsequent dialysis into pure water. These letters are used to identify the micrograph as well as to refer the reader to the relevant experimental conditions and results related to the specific samples which are listed in the column below each micrograph. The top row of the table gives the molar ratio of HCl/4VP (on the left) or NaOH/AA repeat units (on the right), with 0 representing the absence of additives. The pH (prep) values (i.e., the pH at which the samples were prepared), as obtained from the pH meter, are given in the next row (below the ratios), in the row with gray shading. In this and all subsequent tables, the gray shading will be utilized for the row devoted to the parameter that is being explored, in this case, pH (prep). It should be noted that the pH (prep) was measured in the mixed solvent containing 50% (w/w) of each component. The next row in the table is devoted to a schematic description of the observed aggregate morphologies, using symbols for the morphologies (Figure 1) and including the size distribution and standard deviation (SD) in nanometers. This row can be divided into subunits, depending on the number of morphologies which were observed on each micrograph. The most frequently observed morphology (if there is more than one) is always placed on the top line in the third row. The wall thickness of the vesicles is not shown in the table because it did not vary in any systematic way since it depends only on the DP and PS, which are constant; the average value (±SD) was 27 ± 3 nm. The ζ-potential results for the aggregates, which had been prepared at various pH values and quenched in Milli-Q water, are given in the next two rows of the figure. The pH values (3.0 or 4.2) (on the far left of the fourth to seventh rows in Figure 5) are those at which the ζ potentials were measured. It had been found26 that the ζ potentials as a function of pH could be correlated with the corona composition as shown in Figure 2 SI. The composition of the corona, as deduced from measurements at either of the two pH values, is shown in the bottom two rows. It is clear from the micrographs that in the region HCl/4VP ≥ 1 (micrographs A−D), vesicles are the dominant morphology. However, at lower HCl/4VP ratios (F) or in pure DMF (G) and also with increasing NaOH/AA values (H− J), the morphologies shift to a coexistence of simple micelles and LCMs or to only LCMs. The case of the highest HCl/4VP ratio (R = 10.7) (micrograph A) is discussed first. Extrapolating from the 5036

dx.doi.org/10.1021/la500712b | Langmuir 2014, 30, 5031−5040

Langmuir

Article

Figure 6. Schematic representation of the pH (prep) and block length effect on the morphology (expressed by schematic symbols) and corona composition (expressed by the color codes) of PS-b-PAA and PS-b-P4VP mixtures in DMF/water (1:1) at pH 3.0. The PAA block length is kept constant, and the P4VP block length varies as illustrated by gray shading on the composition grids beside each subtable. The corona composition is identified by the following colors: pure PAA, blue; mostly PAA, green; mixture of PAA and P4VP, yellow (transition region); mostly P4VP, orange; and pure P4VP, pink.

ζ-potential measurements at pH 3.0 and 4.2, is located at an HCl/4VP ratio of about 3.0 (B). In the transition region, the coil dimensions of PAA and P4VP should be comparable, since the corona of the aggregate is composed of both. It should be noted that, in this case, vesicles are found on both sides of the transition region. The morphologies and the corona compositions of the aggregates over the entire range of R values (i.e., between 10.7 HCl/4VP and 3.0 of NaOH/AA) can now be discussed. At the highest HCl/4VP ratio (i.e., 10.7 (A)), the PAA, being nonionic, is least affected by the presence of salt and therefore stays outside of the vesicles and forms the external corona. By contrast, quaternized P4VP is insoluble in DMF and becomes marginally soluble only above 12% (w/w) water; therefore, it forms the inner corona of the vesicle. The interface compositions must reflect the relative coil dimensions under conditions at which vesicles are formed. With decreasing R values, the solubility of the quaternized poly(vinylpyridinium) chloride chains increases; in the transition region (HCl/4VP ratio ca. 3.0 (B)), the coil dimensions become equal, with P4VP dominant at all higher pH(prep) values. Below the transition region, at HCl/4VP ratios 3.0), the PAA chains were expected to be present in the corona; however, according to the ζ-potential results, the corona is, in all combinations, composed of P4VP chains. Such behavior can be explained considering the solubility of P4VPCl and PAANa in DMF. As is clear from Figures 6 and 14 SI, which give general corona composition trends (not only the extreme cases), as the pH (prep) increases, the corona composition changes from PAA, to mostly PAA, to a mixture of PAA and P4VP, to mostly P4VP, and finally to only P4VP chains. Such behavior is consistent over all nine block copolymer combinations and, as mentioned above, is in contrast to what was expected on the basis of the block length effect.26,27 In samples without additives (at R = 0) and at constant PAA block lengths (Figures 6 and 14 SI), the P4VP content of the corona increases with increasing P4VP block length, which is the expected trend.26,27 It is also clear that at a constant PAA block length of 11 units (top subtable), as the P4VP block length increases, the transition region (yellow color) moves toward higher pH values in the acidic region (higher ratios). At constant PAA block lengths of 27 and 45 units (second and third subtables), as the block length of P4VP increases, the transition region occurs at lower acidic ratios. At a PAA block length of 45 units (third subtable in Figures 6 and 14 SI) the range of the transition region tends to narrow as the P4VP block length increases.

4. CONCLUSIONS Mixtures of amphiphilic diblock copolymers produce aggregates of various morphologies, including vesicles with differing corona compositions on the interior and exterior interfaces. In the case of mixtures of two block copolymers with a common hydrophobic block, such as PS-b-PAA and PS-bP4VP, either the PAA or the PVP or a mixture of both can be directed to the outside interface of the vesicles. Knowledge of the solubility is crucial to understanding the behavior with regard to the morphology and corona composition. Using HCl as an additive during self-assembly, for the highest HCl/4VP ratio, the protonated P4VP would be expected to form the external corona of the aggregates. The encountered reversal of the expected corona composition is explained as follows: A mixture of PAA and HCl is soluble in DMF. HCl at low concentrations would not be expected to perturb the PAA coil dimensions appreciably. The PAA coil dimensions are larger than those of the (protonated) poly(vinylpyridinium) chloride because the latter is barely soluble in DMF at low water concentrations. Eventually, when enough water is added, the poly(vinylpyridinium) chloride becomes soluble and the coil dimensions become “normal”, but by then the vesicles are already formed with the PAA on the outside. In the case of micelles, the poly(vinylpyridinium) chains presumably form a coronal complex with the protruding soluble PAA chains in the solution. In the inverse case of highest NaOH/AA ratio during selfassembly, the external corona of the aggregates was expected to be composed of PAANa chains. Again, exactly the opposite was found to be the case. The stoichiometric mixture of PAA and NaOH (i.e., PAANa) is not soluble in DMF. Therefore,

■

ASSOCIATED CONTENT

S Supporting Information *

The SI contains a detailed description of the experimental aspects of the work, specifically an overview of the block copolymers used, the method of preparing the aggregates, the experimental details of pH measurement, the evaluation of the corona composition, transmission electron microscopy, electrophoretic mobility and absorbance. An experimental design is described (i.e., the compositions of the mixtures and of the relationship between electrophoretic mobility and corona composition. The scheme of data presentation is also given, along with an overview of the solubility of PAA and P4VP. A detailed presentation of the results of investigation of the pH effect in DMF/water (1:1) in eight block copolymer mixtures, specifically AA11/VP33, AA11/VP58, AA27/VP12, AA27/ VP33, AA27/VP58, AA45/VP12, AA45/VP33 and AA45/ VP58 is also given. A discussion of the reproducibility of the morphologies obtained under various experimental conditions and an overview of the results of the pH and block copolymer length on the morphology and corona composition at pH = 4.2 are given at the end of SI. This material is available free of charge via the Internet at http://pubs.acs.org/.

■

AUTHOR INFORMATION

Corresponding Author

*Tel: 514-398-6934. Fax: 514-398-3797. E-mail: adi. [email protected]. 5039

dx.doi.org/10.1021/la500712b | Langmuir 2014, 30, 5031−5040

Langmuir

Article

Author Contributions

(14) Zhu, J.; Zhang, S.; Zhang, K.; Wang, X.; Mays, J. M.; Wooley, K. L.; Pochan, D. J. Disk-Cylinder and Disk-Sphere Nanoparticles via a Block Copolymer Blend Solution Construction. Nat. Commun. 2013, 4, 2297. (15) Zhang, M.; Hu, Y.; Hassan, Y.; Zhou, H.; Moozeh, K.; Scholes, G. D.; Winnik, M. A. Slow Morphology Evolution of Block Copolymer−Quantum Dot Hybrid Networks in Solution. Soft Matter 2013, 9, 8887−8896. (16) Adams, D. J.; Kitchen, C.; Adams, S.; Furzeland, S.; Atkins, D.; Schuetz, P.; Fernyhough, C. M.; Tzokova, N.; Ryan, A. J.; Buttler, M. F. On the Mechanism of Formation of Vesicles from Poly(ethylene oxide)-block-poly(caprolactone) copolymers. Soft Matter 2009, 5, 3086−3096. (17) Rajagopal, K.; Mahmud, A.; Pajerowski, C. J. D.; Brown, A. E. X.; Loverde, S. M.; Discher, D. E. Curvature-Coupled Hydration of Semicrystalline Polymer Amphiphiles Yields flexible Worm Micelles but Favors Rigid Vesicles: Polycaprolactone-Based Block Copolymers. Macromolecules 2010, 43, 9736−9746. (18) Zhu, Y.; Tong, W.; Gao, C.; Mohwald, H. Assembly of Polymeric Micelles into Hollow Microcapsules with Extraordinary Stability against Extreme pH Conditions. Langmuir 2008, 24, 7810− 7816. (19) Du, J.; O’Reilly, R. K. Advances and Challenges in Smart and Functional Polymer Vesicles. Soft Matter 2009, 5, 3544−3561. (20) Alexandridis, P.; Lindman, B. Amphiphilic Block Copolymers: SelfAssembly and Applications; Elsevier: Amsterdam, 2000. (21) Förster, S.; Antonietti, M. Amphiphilic Block Copolymers in Structure-Controlled Nanomaterial Hybrids. Adv. Mater. 1998, 10, 195−217. (22) Bockstaller, M. R.; Mickiewicz, R. A.; Thomas, E. L. Block Copolymer Nanocomposites: Perspectives for Tailored Functional Materials. Adv. Mater. 2005, 17, 1331−1349. (23) Balacz, A. C.; Emrick, T.; Russel, T. P. Nanoparticle Polymer Composites: Where Two Small Worlds Meet. Science 2006, 314, 1107−1110. (24) Kaler, E. W.; Murthy, A. K.; Rodriguez, B. E.; Zasadzinski, J. A. N. Spontaneous Vesicle Formation in Aqueous Mixtures of SingleTailed Surfactants. Science 1989, 245, 1371−1374. (25) Safran, S. A.; Pincus, P.; Andelman, D. Theory of Spontaneous Vesicle Formation in Surfactant Mixtures. Science 1990, 248, 354−356. (26) Luo, L.; Eisenberg, A. One-Step Preparation of Block Copolymer Vesicles with Preferentially Segregated Acidic and Basic Corona Chains. Angew. Chem., Int. Ed. 2002, 41, 1001−1004. (27) Luo, L.; Eisenberg, A. Thermodynamic Stabilization Mechanism of Block Copolymer Vesicles. J. Am. Chem. Soc. 2001, 123, 1012− 1013. (28) Liu, F.; Eisenberg, A. Preparation and pH Triggered Inversion of Vesicles from Poly(acrylic Acid)-block-Polystyrene-block-Poly(4-vinyl Pyridine). J. Am. Chem. Soc. 2003, 125, 15059−15064. (29) Kabanov, V. A. Polyelectrolyte Complexes in Solution and in Bulk. Russ. Chem. Rev. 2005, 74, 3−20. (30) Kabanov, V. A.; Zezin, A. B. Soluble Interpolymeric Complexes as a New Class of Synthetic Polyelectrolytes. Pure Appl. Chem. 1984, 56, 343−354. (31) Riess, G. Micellization of Block Copolymers. Prog. Polym. Sci. 2003, 28, 1107−1170.

The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We would like to thank to Dr. Hongwei Shen for synthesizing PS-b-P4VP diblock copolymers, Dr. Futian Liu for synthesizing PS-b-PAA diblock and Dr. Tony Azzam for synthesizing the homopolymer of poly(4-vinylpyridine) used to determine the solubility and to measure absorbance curves. We would also like to thank NSERC for providing financial support in this study. Axel Müller acknowledges a visiting professorship sponsored by NSERC. Financial support by NSERC is gratefully acknowledged.

■

ABBREVIATIONS PS-b-PAA, Polystyrene-block-poly(acrylic acid); PS-b-P4VP, polystyrene-block-poly(vinylpyridine); PAANa, sodium acrylate; P4VPCl, poly(vinylpyridinium chloride); SSMs, small spherical micelles; LCMs, large compound micelles; THF, tetrahydrofuran; TEM, transmission electron microscopy; SI, Supporting Information; IPEC, interpolyelectrolyte complex

■

REFERENCES

(1) Zhang, L. F.; Eisenberg, A. Multiple Morphologies of “Crew-Cut” Aggregates of Polystyrene-b-poly(acrylic acid) Block Copolymers. Science 1995, 268, 1728−1731. (2) Zhang, L.; Eisenberg, A. Multiple Morphologies and Characteristics of “Crew-Cut” Micelle-like Aggregates of Polystyrene-b-poly(acrylic acid) Diblock Copolymers in Aqueous Solutions. J. Am. Chem. Soc. 1996, 118, 3168−3181. (3) Betthausen, E.; Drechsler, M.; Förtsch, M.; Schacher, F. H.; Müller, A. H. E. Dual Stimuli-Responsive Multicompartment Micelles from Triblock Terpolymers with Tunable Hydrophilicity. Soft Matter 2011, 7, 8880−8891. (4) Discher, B. M.; Won, Y.-Y.; Ege, D. S.; Lee, J. C.-M.; Bates, F. S.; Discher, D. E.; Hammer, D. A. Polymersomes: Tough Vesicles Made from Diblock Copolymers. Science 1999, 284, 1143−1146. (5) Torchilin, V. P. Micellar Nanocarriers: Pharmaceutical Perspectives. Pharm. Res. 2007, 24, 1−16. (6) Forster, S.; Abetz, V.; Müller, A. H. E. Polyelectrolyte Block Copolymer Micelles. Adv. Polym. Sci. 2004, 166, 173−210. (7) Giacomelli, C.; Schmidt, V.; Borsali, R. Specific Interactions Improve the Loading Capacity of Block Copolymer Micelles in Aqueous Media. Langmuir 2007, 23, 6947−6955. (8) Kwon, A. V.; Kataoka, K. Block copolymer micelles as longcirculating drug vehicles. Adv. Drug Delivery Rev. 1995, 16, 295−309. (9) Borisov, O. V.; Zhulina, E. B.; Leermakers, F. A. M.; Müller, A. H. E. Adv. Polym. Sci. 2011, 241, 57−129. (10) Kabanov, A. V.; Alakhov, V. Y.; Alexandris, P.; Lindman, B. Amphiphilic Block Copolymers: Self Assembly and Applications; Elsevier: Amsterdam, 1997. (11) Gindy, M. E.; Panagiotopoulos, A. Z.; Prud’homme, R. K. Composite Block Copolymer Stabilized Nanoparticles: Simultaneous Encapsulation of Organic Actives and Inorganic Nanostructures. Langmuir 2008, 24, 83−90. (12) Jain, S.; Bates, F. S. On the Origins of Morphological Complexity in Block Copolymer Surfactants. Science 2003, 300, 460−464. (13) Wilhelm, M.; Zhao, Ch.-L.; Wang, Y.; Xu, R.; Winnik, M. A.; Mura, J.-L.; Riess, G.; Croucher, M. D. Poly(styrene-ethylene oxide) Block Copolymer Micelle Formation in Water: a Fluorescence Probe Study. Macromolecules 1991, 24, 1033−1040. 5040

dx.doi.org/10.1021/la500712b | Langmuir 2014, 30, 5031−5040