Structural Evolution of Polyelectrolyte Complex Core Micelles and

Nov 11, 2014 - Polyelectrolyte complex formation, both in the form of solid phase precipitation and fluid phase coacervate formation, has been studied...
0 downloads 14 Views 1MB Size
Download PDF
Article pubs.acs.org/Macromolecules

Structural Evolution of Polyelectrolyte Complex Core Micelles and Ordered-Phase Bulk Materials Daniel V. Krogstad,†,‡,∇ Nathaniel A. Lynd,‡,○ Daigo Miyajima,‡,§ Jeffrey Gopez,‡,∥ Craig J. Hawker,†,‡,§ Edward J. Kramer,†,‡,∥ and Matthew V. Tirrell*,⊥,# †

Department of Materials, ‡Materials Research Laboratory, §Department of Chemistry and Biochemistry, and ∥Department of Chemical Engineering, University of California, Santa Barbara 93106, United States ⊥ Institute for Molecular Engineering, The University of Chicago, Chicago, Illinois 60637, United States # Argonne National Laboratory, Lemont, Illinois 60439, United States S Supporting Information *

ABSTRACT: The kinetics of formation and structural evolution of novel polyelectrolyte complex materials formed by the assembly of water-soluble di- and triblock copolymers, with one neutral block and one (or two) cationic or anionic blocks, have been investigated. Comparison was made between the assembly of ABA and AB′ copolymers in which A represents the ionic blocks and B and B′ are the neutral poly(ethylene oxide) blocks. The degree of polymerization of B was twice that of B′ and the ionic A blocks were of equal degrees of polymerization in all polymers. The mechanism and speed of the assembly process, and the organization of these domains, was probed using dynamic mechanical spectroscopy and small-angle X-ray scattering (SAXS). SAXS revealed that the equilibrium morphologies of both the diblock copolymer and the triblock copolymer materials were generally qualitatively the same with some apparent quantitative differences in phase boundaries, possibly attributable to lack of full equilibration. Slow kinetics and difficulties in reaching equilibrium phase structures, especially in triblock materials, is a principal message of this article. Detailed analysis of the SAXS data revealed that the triblock copolymer materials formed ordered phases via a nucleation and growth pathway and that the addition of small amounts (∼20%) of corresponding diblock copolymers increased the rate of structure formation and enhanced several key physical properties.

■

INTRODUCTION Soft, self-assembling, self-repairing materials have many potential applications, especially, though not exclusively, in biomedical applications.1,2 A particular advantage of such materials is the capability of developing mechanical properties that are tunable to different steady states, and switchable dynamically, over orders of magnitude. Tunability is typically achieved by variations in molecular weight, concentration or strength of intermolecular bonding. Switchability refers to a dynamic response to changes in environmental conditions or imposed fields. Both characteristics are useful in biomaterials. In the case of drug delivery applications, designer materials may be locally injected to an inflamed site and slowly release a drug cargo. In regenerative medicine, injectable scaffolds may be useful. In order to be considered injectable, soft, self-assembling self-repairing materials are desirable as they can be injected through a syringe and then form an elastic gel in situ to stabilize the injected material.1 To assess the usefulness of a material for applications of this sort, it is necessary to investigate the kinetics and dynamics of assembly into a material of sufficient rigidity. In this work, we explore the kinetics of structure and © 2014 American Chemical Society

modulus development in electrostatically self-assembled block copolymers. The equilibrium phases of neutral block copolymers assembled due to dispersion forces are reasonably well understood; however, generally less is known about the kinetics of phase separation and ordering. Like all materials that exhibit ordered structures, block copolymer microphase separation has been described as proceeding through nucleation and growth or spinodal decomposition mechanisms.3−8 Nucleation and growth occur when the disordered phase is metastable with respect to the ordered phase.8 During classical nucleation and growth, small critical nuclei are formed within the disordered phase, during what is often termed the incubation period. These nuclei then grow larger until they fill the entire sample volume. At this point, the ordered phase can only increase its perfection further through defect annihilation.3−7 On the other Received: August 28, 2014 Revised: October 20, 2014 Published: November 11, 2014 8026

dx.doi.org/10.1021/ma5017852 | Macromolecules 2014, 47, 8026−8032

Macromolecules

Article

Figure 1. Phase diagrams developed for the (a) diblock and the (b) triblock copolymer materials showing the equilibrium phases at varying polymer and salt concentrations. The same phases are present for both systems; however, the two-phase regions are smaller and the BCC phase is larger for the diblock copolymers. The phase diagrams have been developed from SAXS data. The lines have been added to guide the eye. Part b has been reproduced from a previous publication.12

hand, during spinodal decomposition, composition fluctuations occur simultaneously throughout the entire medium. The kinetics of block copolymer ordering have typically been studied experimentally by quenching the samples from the disordered state to an ordered state and observing the structural changes through scattering experiments. Experimental investigation of ordering in the spherical body-centered cubic (BCC) and hexagonally packed cylinders (HEX) morphologies indicate that most block copolymer systems form ordered phases through nucleation and growth mechanisms.3−7,9,10 The time-scale of order formation in these samples varies significantly from seconds to hours. There have been a few reports of block copolymer systems that undergo spinodal decomposition.8,9 Spinodal decomposition is observed in scattering experiments by the immediate presence of a diffuse scattering ring and then subsequent sharpening of the ring with time. In contrast to kinetic studies of assembly in neutral block copolymer solutions and melts, very little has been done to study the kinetics of electrostatic assembly of charged block copolymer material systems. For such materials, the presence of a large amount of water (60−99%), could potentially have significant effects on both the pathway and the time scale of the ordering. All of the studies of ordering phenomena on uncharged polymer systems pertain to situations where the ordering was driven by balances of van der Waals interactions among the polymer segments. In previous work,11,12 we have developed and characterized a block copolymer system where micellar assembly and ordering phenomena are driven by electrostatic interactions rather than by van der Waals forces. These publications concerned the synthesis and efforts to characterize the equilibrium structures that evolve in aqueous systems consisting of water-soluble block copolymers with uncharged blocks and polyelectrolyte blocks. When mixed together in water, the polyelectrolyte end-blocks interact and form polyelectrolyte complex domains that act as physical cross-links in the networked material as has been described previously.11,12 Polyelectrolyte complex formation, both in the form of solid phase precipitation and fluid phase coacervate formation, has been studied fairly extensively, though, until recently, not very quantitatively, both in homopolymer systems13−17 and in block copolymer micellar suspensions.18−28 However, little attention has been paid to the kinetics of formation of the complex

domains, and to the resulting larger scale ordering process at higher copolymer concentrations. In this study, we investigated the ordering kinetics in the same chemical system studied previously, namely, functionalized poly[(allyl glycidyl ether)-b(ethylene oxide)-b-(allyl glycidyl ether)] (P(AGE)-b-P(EO)-bP(AGE)) triblock copolymers and (P(AGE)-b-P(EO)) diblock copolymers. The AGE repeat units were functionalized postpolymerization with either sulfonate or guanidinium groups to form copolymers with oppositely charged polyelectrolyte end-blocks. The kinetics of assembly and ordering of the triblock copolymers, analogous diblock copolymers, and mixtures of the two were studied using dynamic mechanical spectroscopy and SAXS.

■

MATERIALS AND METHODS

Polymer Synthesis and Functionalization. The triblock copolymers were synthesized by anionic ring-opening polymerization of allyl glycidyl ether (AGE) from a PEO-diol macroinitiator as in earlier publications.11,12,29 The diblock copolymers were synthesized by sequential anionic polymerization starting with the initiation of AGE from benzyl alcohol titrated with potassium naphthalenide in the absence of solvent at 30 °C. Under these conditions, isomerization and subsequent loss of the functional allyl groups was eliminated.30 Once the AGE monomer was completely consumed, tetrahydrofuran was added followed by the addition of ethylene oxide (EO). The EO was allowed to polymerize overnight at 45 °C to produce the final diblock copolymer. The composition and molecular weight of the resultant PAGE-b-PEO diblock copolymer was defined by the polymerization stoichiometry. The AGE repeat-units on both the triblock and diblock copolymers were functionalized using thiol−ene click chemistry as has been described previously.11 The base P(AGE)31-b-P(EO)455-bP(AGE)31 triblock copolymer had a number-average molecular weight of 27 000 Da (1H NMR spectroscopy), and a polydispersity index of 1.14 (size exclusion chromatography). The P(AGE)30-b-P(EO)216 diblock copolymer had a number-average molecular weight of 12 950 Da, very close to one-half the molecular weight of the triblock, and a polydispersity index of 1.06. Complete functionalization of the block copolymers with guanidinium and sulfonate groups was demonstrated by 1H NMR spectroscopy. Dynamic Mechanical Spectroscopy. Samples for dynamic mechanical spectroscopy were prepared by dissolving the oppositely charged polyelectrolytes separately in water at the intended polymer concentration. The sulfonate-functionalized polymer solution was added with a pipet to the guanidinium-functionalized polymer solution and the solution was immediately mixed with a vortex for 1 min and placed on the rheometer for testing. Testing was performed on a Rheometrics Scientific Ares II rheometer using the parallel plate 8027

dx.doi.org/10.1021/ma5017852 | Macromolecules 2014, 47, 8026−8032

Macromolecules

Article

geometry (25 mm diameter). During a typical experiment, three tests were run: a time sweep, a frequency sweep and a strain sweep. For the time sweeps, the materials were measured at 1 Hz and 1% strain every 10 s for the duration of the experiment. Immediately after, frequency sweeps were performed from 0.005 to 8 Hz at 1% strain. Lastly, a strain sweep from 0.02 to 400% strain at 1 Hz was used to ensure that the previous experiments were performed in the linear viscoelastic regime. Small Angle X-ray Scattering. Materials for the small-angle X-ray scattering (SAXS) experiments were prepared in the same manner as the dynamic mechanical spectroscopy samples. The materials were then injected into 1.5 mm boron enriched quartz capillary tubes using a syringe. Both ends of the capillary were flame-sealed. Most of the SAXS experiments were performed at beamline 8-ID-E at the Advanced Photon Source, Argonne National Laboratory using 7.35 keV X-rays and a detector distance of 2.18 m. Some of the SAXS experiments used for the development of the diblock copolymer phase diagram were performed at beamline 1−4 at the Stanford Synchrotron Radiation Lightsource using 7.1 keV X-rays and a detector distance of 2.1 m.

lower critical micelle concentration in diblocks compared to triblocks. The salt concentrations at which the ordered phases were present were higher for the diblock copolymers, which potentially indicates that the diblock micelles were more stable with respect to ordered phase formation than the corresponding triblock derivatives. Lastly, the BCC region extends to higher polymer concentrations for the diblock copolymers, likely due to the PEO chains in the diblock micelle being less extended than the fraction of PEO chains that can act as bridges between complex domains in the triblock copolymer system. As a result, the number density in the diblock micelles was larger. Consequently, the BCC phase existed in the diblock systems at polymer concentrations where the hexagonal phase existed for the triblock copolymers. This was confirmed by examining the spherical domain spacing of the diblock and triblock copolymers at varying polymer concentrations (Figure S1, Supporting Information). It was seen that the domain spacing of the two systems were approximately equal until about 16−18 wt %; after which, the spacing of the diblock copolymers became smaller relative to that for similar triblock copolymer systems. Kinetics of Ordering. Some of the differences in the phase diagrams of the diblock and triblock copolymers, gave rise to the hypothesis that the triblock copolymer architecture could inhibit equilibration of the ordering transition through entanglements and bridging configurations of the PEO midblocks. In other words, the translation of a single complex domain may require the rearrangement of a much larger number of polymer configurations for the connected triblock copolymer domains. To test this hypothesis, the kinetics of ordered phase formation were studied, at a fixed total polymer concentration, using dynamic mechanical spectroscopy and SAXS. The oppositely charged polyelectrolyte solutions were mixed at a polymer concentration of 20 wt % (a composition at which the final ordered structure of both diblock and triblock materials was BCC), immediately placed on the rheometer, and the storage (G′) and loss (G″) moduli were monitored at fixed frequency (1 Hz) and strain (1%) for 500 min (Figure 2). Initially, G′ increased rapidly for both systems until the slope of the moduli changed in an intermediate regime and finally plateaued. The modulus of the triblock copolymer material increased to 17% of the final plateau modulus before the experiment could be started (approximately 5 min after mixing), the slope decreased dramatically after approximately 75 min and a final plateau was observed at 400 min. These results were in contrast to the initial modulus being 60% for the diblock copolymer materials, which showed a decreased slope at approximately 70 min with a plateau reached at a time of 120 min. (The slight decrease in the diblock G′ seen after 200 min is not believed to be real but rather a result of difficulties in maintaining a perfectly stable rheological experiment over several hours.) From these experiments, it was readily apparent that the diblock copolymers develop their structure significantly faster than the corresponding triblock copolymers. Additionally, G′ for the diblock copolymers was 54% larger than G′ for the triblock copolymer after 500 min and the increased modulus for the diblock copolymer samples indicates that the triblock copolymers were less ordered, even after 500 min of equilibration. SAXS was used to interrogate further the microstructural reasons underlying the differences in kinetics of ordering between diblock copolymer and triblock copolymer materials.

■

RESULTS AND DISCUSSION Block Copolymer Phase Diagrams. The PAGE-b-PEO diblock copolymer system was prepared and its phase behavior was characterized in order to compare it with the triblock copolymer system reported previously.11,12 These charged diblock copolymers formed micelles through the ionic complexation of the oppositely charged polyelectrolyte blocks into complex domains. The effects of polymer and salt concentration on the complex organization were investigated using small-angle X-ray scattering (SAXS) and were shown to have significant effects on the size, shape and the structural organization of these domains. From these data, a phase diagram for the diblock copolymer materials was developed (Figure 1a) and compared to the previously reported triblock copolymer material phase diagram12 (Figure 1b). For diblock copolymer samples with no added NaCl, at 12 wt % polymer and below, they formed disordered spherical micelles (DIS), between 14 and 30 wt %, they formed spheres on a BCC lattice (BCC), and at 35 wt % and above, they formed cylinders packed on a hexagonal lattice (HEX). The effects of the salt concentration on the self-assembly of the diblock copolymers were similar to the corresponding triblock copolymer materials. Increasing salt concentration leads to disruption of the nanostructure; the salt concentration required to disrupt the ionic associations increased with increasing polymer concentration. Comparison of the phase diagrams for the diblock and triblock copolymers revealed three distinctive features for the diblock copolymers: the apparent two-phase coexistence (i.e., DIS/BCC) regions were significantly smaller, a polyelectolyte solution phase was not observed, even under conditions where a homogeneous solution appeared to be obtained in triblock copolymers, and the region of stability for the BCC phase was significantly larger. The smaller two-phase region may be due to the ease or speed of ordering in the diblock copolymer sample. In other words, based on work described later in this paper, ordering is particular slow in the triblock systems. It cannot yet be assured that Figure 1 represents fully equilibrated phase behavior, particularly with respect to the apparent coexistence regions, especially in the triblock system. Progress toward understanding the kinetics of approach to equilibrium is the principal aim of this paper. The absence of a clear solution phase in the diblock copolymer system at low concentration may be the result of a 8028

dx.doi.org/10.1021/ma5017852 | Macromolecules 2014, 47, 8026−8032

Macromolecules

Article

Figure 2. Dynamic mechanical spectroscopy data showing the evolution of G′ and G″ with time for both the diblock and the triblock copolymer materials. The diblock copolymers not only equilibrated faster, but G′ for the diblock copolymer was significantly larger over the entire time sweep. Both systems were tested at 20 wt % and 0 M NaCl which corresponds to the BCC phase. The breaks in the curves represent brief periods where frequency sweeps were performed.

By collecting SAXS patterns on both samples periodically over a 12-h period, structural changes that correlated with the evolution of viscoelasticity could be observed (Figure 3). For the triblock copolymers, SAXS revealed that the materials did not initially form an ordered structure; however, approximately 30 min after mixing, the complex domains started organizing on a BCC lattice. These peaks continued to increase in magnitude until 60 min, after which, the peaks remained relatively constant in intensity. Further structural evolution was evident in the narrowing of the reflections and a shift toward lower q values. In contrast, the diblock copolymers formed well-ordered BCC structures almost immediately and with time, the first order peak shifted to lower q values and narrowed. While qualitative structural changes could be extracted from the SAXS data, quantitative analysis revealed significantly more information about the assembly process. The primary peak was fit with a Gaussian function and the full-width half-maximum (fwhm) was measured and used as a self-consistent assessment of the long-range order of the materials. In this analysis, the Scherrer equation was used in order to estimate the translational correlation length (D). The Scherrer equation states that the width of the first peak is directly proportional to the correlation length through the equation D = 2π/(fwhm). The results of this analysis on the diblock and the triblock copolymer materials can be seen in Figure 4. The diblock copolymer had larger correlation lengths over the entire 12-h study. After 12 h, the diblock copolymers had a onedimensional correlation length of approximately 10 000 Å (ca. 1 μm) and the triblock copolymer had a correlation length of approximately 5000 Å (ca. 0.5 μm). Both samples had a lattice spacing of about 290 Å; therefore, the correlation length corresponds to ordered BCC grains of 35 and 18 unit cells across for the diblock and triblock copolymer materials, respectively. In volumetric terms, the ordered diblock copolymer BCC grains were about eight times larger than the ordered triblock copolymer BCC grains. By combining the dynamic mechanical spectroscopy and the SAXS data for the triblock copolymers, it can be concluded that the majority of the complex domains were formed and aligned

Figure 3. Small angle X-ray scattering was utilized to investigate the structural evolution with time for both the (a) triblock and the (b) diblock copolymer materials. (a) It can be seen that for the triblock copolymers, the 5, 7.5, 10, and 15 min samples are completely overlapping and the BCC structural peaks were not initially observed. However, starting at about 20−30 min, the second and third order BCC peaks started to form (inset). With time, these peaks increased and the width of the first order peak decreased. The formation of the second- and third-order peaks are more clearly visible in the inset showing the 15, 20, 30, 45, and 60 min patterns. (b) For the diblock copolymers, the BCC structure formed quicker and was observed even at 5 min. Both samples had a polymer concentration of 20 wt % and a salt concentration of 0 M NaCl. The arrows indicate the direction of changes in the SAXS patterns with time.

on a BCC lattice within the first 75 min. Additionally, the shape of the growth profile of the ordered BCC grains for the triblock copolymers in Figure 4 was consistent with a nucleation and growth pathway as is illustrated in Figure 4. Shortly after mixing, a disordered arrangement of spherical complex domains exists (time A), and only at approximately 30 min (time B), were critical BCC nuclei stabilized. Once stabilized, the ordered BCC grains grew rapidly in size and perfection (time C) until the grains filled most of the sample (time D). However, small changes in the correlation length and the modulus (from dynamic mechanical spectroscopy) continued for the remainder of the experiments. Additionally, the change in q* from the SAXS data indicated that the complex domain spacing increased during the ordering process. Therefore, we hypothesized that after 75 min, further growth primarily occurred through defect annihilation at the grain boundaries. For both the diblock and triblock copolymer systems, the correlation length profiles did not completely plateau within 12 8029

dx.doi.org/10.1021/ma5017852 | Macromolecules 2014, 47, 8026−8032

Macromolecules

Article

copolymer and the triblock copolymer underling the different kinetics of structural evolution in these materials. In the diblock copolymer system, the complex core micelles are not physically connected to their neighboring complex domains, and therefore structural rearrangement to a highly ordered and defect-free state was relatively rapid. However, in the triblock copolymer system, there was a large fraction of configurations where one or more triblock copolymers formed a bridge between two neighboring complex domains. Annealing to a highly ordered state required the dynamic rearrangement of these bridging configurations to allow the complex domains to relax to their lowest energy locations. Because of the architectural restrictions in the triblock copolymer system, the structure appears to continue to evolve for at least 13.5 months after initial formation of the ordered material. Future studies will explore the possible effects of annealing conditions on structural evolution in these systems. While extracting the correlation length from the SAXS data provides interesting insights into the material ordering, there are caveats that accompany this analysis. The Scherrer method assumes that the width of the primary peak was proportional to the correlation length. However, there were other contributions to peak width: defects within the crystallites, a potential distribution of lattice spacing (paracrystalline disorder), and instrumental factors such as beam profile, can all broaden the scattering peaks. Therefore, this simple method only provides an estimate of the lower bound of the translational correlation length. To minimize instrumental contributions, all of the experiments were performed on the same beamline at the same time to ensure that self-consistent comparisons among the samples were meaningful. Other factors that contribute to peak width can be thought of collectively as contributions due to the disorder of the samples; therefore, the correlation length can be considered representative of the cumulative degree of order. Rheology of Material Formation from Mixtures of Diblock and Triblock Copolymers. Prior work has shown that the triblock copolymer materials have interesting mechanical properties, namely substantial elastic rigidity even with 80% water content;12 however, the long equilibration times observed in Figure 4 could be a limiting factor for many applications. In order to decrease the equilibration times and overcome this potential limitation, the effect of small amounts of diblock copolymers added to the triblock copolymer materials was studied. Spontak et al. have previously shown that the modulus of ordered block copolymer materials could be increased by adding a small percentage of diblock copolymers to a triblock copolymer gel.31 They determined that the corona molecules of the diblock copolymers shrunk at the core−corona interface, which forced the midblock of the triblock copolymer, through steric repulsion, to be extended and promote bridging configurations between neighboring domains. Therefore, it was hypothesized that not only would adding small amounts of diblock copolymers into these functionalized PAGE−PEO−PAGE triblock copolymer complex materials likely decrease the equilibration time necessary to form these bridges, but it would also potentially increase the modulus by increasing the number of bridging configurations. Dynamic mechanical spectroscopy was therefore performed on 20 wt % polymer materials in which the composition ranged from 100% triblock copolymer to 100% diblock copolymer with frequency sweeps (Figure 6), time sweeps (Figure S2) and strain sweeps

Figure 4. Translational correlation length was estimated from the SAXS data using the Scherrer method for the 20 wt % diblock (red symbols) and the 20 wt % triblock copolymer materials (blue symbols). The kinetics of ordering were faster and the overall correlation length was larger for the diblock copolymers. For the triblock copolymer materials, the ordering clearly occurred through a nucleation and growth pathway as shown schematically to the right. At time point A, individual complex domains are formed, but they are not arranged on any lattice. At time point B, the critical BCC nuclei have formed. At time C, these nuclei are growing quickly to form larger (or more perfect) ordered BCC grains until time point D in which the vast majority of the complex domains were organized on a BCC lattice. Further increases in the correlation length were due to defect annihilation. The circles in the schematic represent both the PEO and the PAGE domains.

h. Therefore, increased annealing times were needed in order to investigate the long-term structural evolution within the materials, and to ascertain if they had reached an equilibrated state. Figure 5 shows the SAXS patterns for triblock copolymer

Figure 5. SAXS patterns for 20 wt % triblock copolymer materials. The gels were kept in sealed capillaries for between 3 h and 13.5 months. The data show that not only do the peaks get narrower with increased annealing time, but the peaks shifted to higher q, indicating that the domain spacing decreased.

materials that were annealed at room temperature for up to 13.5 months in flame-sealed capillary tubes. Significantly, these data show that the polyelectrolyte complex materials continue to evolve structurally even over the course of a year, with the lattice reflections continuing to narrow and q* increasing. These results indicated that the BCC grains grew and/or became more ordered and the complex domain spacing decreased. These results therefore suggest that the gels were kinetically trapped immediately after the initial formation of the material with the architectural differences between the diblock 8030

dx.doi.org/10.1021/ma5017852 | Macromolecules 2014, 47, 8026−8032

Macromolecules

Article

diblock copolymer gels were significantly weaker. The combination of these three experiments shows that the 80% triblock copolymer materials appears to have the best combination of mechanical properties. To confirm these results (Figure 7), materials consisting of 80% triblock, 50% triblock, and 20% triblock copolymers were

(Figure S3) being detailed for all of these samples in the Supporting Information.

Figure 6. Dynamic mechanical spectroscopy frequency sweep at 1% strain for 20 wt % materials after 4.25 h of equilibration consisting of mixtures of diblock and triblock copolymers (labeled as % triblock copolymer). From the frequency sweeps, it can be seen that the high diblock copolymer concentration materials have the flattest spectra, indicating that they are more gel-like. They also have higher moduli at low frequencies, however, at the highest frequencies, the 80% triblock copolymer material has the highest modulus. The lines are drawn to guide the eye.

Figure 7. Translational correlation length was calculated from the SAXS data using the Scherrer method for the 20 wt % materials formed from a mixture of diblock and triblock copolymer materials. Adding a small amount of diblock copolymers into the triblock copolymer materials (80% triblock) increased both the ordering kinetics and the correlation length of the pure triblock copolymer materials. For the 50% triblock sample, the ordering kinetics were slower than any of the other samples. The ordering kinetics and the correlation length of the 20% triblock material was nearly identical to the pure diblock copolymer sample.

Frequency sweeps (Figure 6) were performed to obtain insight into the changes in rheological response as a function of the diblock/triblock copolymer ratio. Interestingly, the pure diblock copolymer material had the highest G′ at all frequencies below 2 Hz, however, at higher frequencies, several of the high percentage triblock copolymer materials had a larger modulus. Additionally, as the triblock copolymer concentration was increased, the slope of G′ versus ω increased significantly. This change in frequency response is potentially due to the equilibrium kinetics of the different samples; the higher percentage triblock copolymer samples had slower equilibration and did not have as well-developed long-range order. Without the long-range order, the materials flowed over long time periods, as was indicated by the low frequency data. However, the high frequency data described very short time scales and thus relied heavily on the short-range order. It was no surprise that at these time scales, the samples that had a high fraction of triblock copolymers had higher moduli since they could form elastic bridges between neighboring complex domains. Interestingly, the 80% triblock sample had the highest G′, indicating that, like the system described by Spontak et al., mixing a small percentage of diblock copolymers with triblock copolymers increased the modulus of the materials. Time sweeps were performed and confirmed the results from the frequency sweeps (Figure S2). Again, it was observed that the 80% triblock sample had a local maximum in both the modulus and the equilibration time indicating that there was some benefit to the amount of diblock copolymers added. The strain sweep (Figure S3) showed that for all of the samples with at least 50% triblock copolymers, the loss of linear viscoelasticity occurred at approximately 20% strain; however, as the diblock copolymer concentration was increased beyond 50%, the critical strain decreased to a minimum of 3% for the pure diblock copolymer sample. Thus, even though the samples had a gel-like response at the low strain (1%) that was used for the frequency (Figure 6) and time sweeps (Figure S2), the

formed and studied using SAXS and analyzed using the Scherrer method. The SAXS results were in agreement with the dynamic mechanical spectroscopy observation. The 80% triblock sample had slightly faster equilibration time than the pure triblock sample and it had a significantly larger correlation length. From these results, it is clear that the 80% triblock copolymer material leads to improved performance with a higher G′, faster equilibration, and the samples have a larger correlation length than the pure triblock copolymer materials at the same breakdown strain.

■

CONCLUSIONS The kinetics of triblock and diblock copolymer materials have been studied using dynamic mechanical spectroscopy and SAXS. We determined that the triblock copolymer materials needed longer equilibration times as compared to the corresponding diblock copolymer materials with the latter forming ordered structures within 5 min and their mechanical properties plateauing at around 120 min. In contrast, the triblock copolymer counterparts took 75 min to form wellordered structures and over 400 min before the mechanical properties stabilized. Additionally, through careful characterization of the materials with SAXS and dynamic mechanical spectroscopy, it was determined that the triblock copolymer materials continued to evolve in structure for at least a year (longer studies were not performed). It was concluded that while the triblock copolymer samples formed through a nucleation and growth pathway and achieved short-range order after a few hours, the samples continued to evolve for 8031

dx.doi.org/10.1021/ma5017852 | Macromolecules 2014, 47, 8026−8032

Macromolecules

Article

(3) Dai, H. J.; Balsara, N. P.; Garetz, B. A.; Newstein, M. C. Phys. Rev. Lett. 1996, 77, 3677−3680. (4) Newstein, M. C.; Garetz, B. A.; Balsara, N. P.; Chang, M. Y.; Dai, H. J. Macromolecules 1998, 31, 64−76. (5) Winter, H. H.; Scott, D. B.; Gronski, W.; Okamoto, S.; Hashimoto, T. Macromolecules 1993, 26, 7236−7244. (6) Harkless, C. R.; Singh, M. A.; Nagler, S. E.; Stephenson, G. B.; Jordansweet, J. L. Phys. Rev. Lett. 1990, 64, 2285−2288. (7) Stuhn, B.; Vilesov, A.; Zachmann, H. G. Macromolecules 1994, 27, 3560−3565. (8) Balsara, N. P.; Garetz, B. A.; Newstein, M. C.; Bauer, B. J.; Prosa, T. J. Macromolecules 1998, 31, 7668−7675. (9) Balsara, N. P.; Lin, C.; Hammouda, B. Phys. Rev. Lett. 1996, 77, 3847−3850. (10) Sakamoto, N.; Hashimoto, T. Macromolecules 1998, 31, 8493− 8502. (11) Hunt, J. N.; Feldman, K. E.; Lynd, N. A.; Deek, J.; Campos, L. M.; Spruell, J. M.; Hernandez, B. M.; Kramer, E. J.; Hawker, C. J. Adv. Mater. 2011, 23, 2327−2331. (12) Krogstad, D. V.; Lynd, N. A.; Choi, S.-H.; Spruell, J. M.; Hawker, C. J.; Kramer, E. J.; Tirrell, M. V. Macromolecules 2013, 46, 1512−1518. (13) Stuart, M. A. C.; Hofs, B.; Voets, I. K.; de Keizer, A. Curr. Opin. Colloid Interface Sci. 2005, 10, 30−36. (14) Hwang, D. S.; Zeng, H.; Srivastava, A.; Krogstad, D. V.; Tirrell, M.; Israelachvili, J. N.; Waite, J. H. Soft Matter 2010, 6, 3232−3236. (15) Srivastava, A.; Waite, J. H.; Stucky, G. D.; Mikhailovsky, A. Macromolecules 2009, 42, 2168−2176. (16) Chollakup, R.; Smitthipong, W.; Eisenbach, C. D.; Tirrell, M. Macromolecules 2010, 43, 2518−2528. (17) Priftis, D.; Tirrell, M. Soft Matter 2012, 8, 9396−9405. (18) Harada, A.; Kataoka, K. Macromolecules 1995, 28, 5294−5299. (19) Harada, A.; Kataoka, K. Science 1999, 283, 65−67. (20) Harada, A.; Kataoka, K. Langmuir 1999, 15, 4208−4212. (21) Harada, A.; Kataoka, K. Macromolecules 2003, 36, 4995−5001. (22) Harada, A.; Kataoka, K. Soft Matter 2008, 4, 162−167. (23) Lemmers, M.; Sprakel, J.; Voets, I. K.; van der Gucht, J.; Stuart, M. A. C. Angew. Chem., Int. Ed. 2010, 49, 708−711. (24) Lemmers, M.; Voets, I. K.; Stuart, M. A. C.; van der Gucht, J. Soft Matter 2011, 7, 1378−1389. (25) Stuart, M. A. C.; Besseling, N. A. M.; Fokkink, R. G. Langmuir 1998, 14, 6846−6849. (26) van der Burgh, S.; de Keizer, A.; Stuart, M. A. C. Langmuir 2004, 20, 1073−1084. (27) Voets, I. K.; de Keizer, A.; Stuart, M. A. C.; Justynska, J.; Schlaad, H. Macromolecules 2007, 40, 2158−2164. (28) Voets, I. K.; de Keizer, A.; Stuart, M. A. C. Adv. Colloid Interface Sci. 2009, 147−48, 300−318. (29) Lundberg, P.; Lynd, N. A.; Zhang, Y. N.; Zeng, X. H.; Krogstad, D. V.; Paffen, T.; Malkoch, M.; Nystrom, A. M.; Hawker, C. J. Soft Matter 2013, 9, 82−89. (30) Lee, B. F.; Kade, M. J.; Chute, J. A.; Gupta, N.; Campos, L. M.; Fredrickson, G. H.; Kramer, E. J.; Lynd, N. A.; Hawker, C. J. J. Polym. Sci., Part A: Polym. Chem. 2011, 49, 4498−4504. (31) Spontak, R. J.; Wilder, E. A.; Smith, S. D. Langmuir 2001, 17, 2294−2297.

months by increasing long-range order, likely through defect annihilation. Lastly, it was shown that when small amounts of diblock copolymers were mixed in with triblock copolymers, the materials equilibrated faster and that materials with an 80% triblock copolymer mixture had significantly increased mechanical properties than the pure triblock copolymer materials. There remains a significant amount of interesting work to be done in this and related systems to understand the thermodynamics and kinetics of phase evolution and the effects of controllable, external influences on these phenomena.

■

ASSOCIATED CONTENT

S Supporting Information *

Figures S1, S2, and S3, showing domain spacing, dynamic mechanical spectroscopy time sweep, and dynamic mechanical spectroscopy strain sweep, and a discussion of kinetics of formation and the structural evolution of electrostatically selfassembled polyelectrolyte complex materials. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*(M.V.T.) E-mail: [email protected]. Telephone: 773834-2001. Fax: 773-834-7756. Present Addresses ∇

(D.V.K.) Illinois Applied Research Institute, University of Illinois at Urbana−Champaign. ○ (N.A.L.) McKetta Department of Chemical Engineering, University of Texas at Austin. Author Contributions

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 The synthesis and rheology work reported here was partially supported by the MRSEC Program of the National Science Foundation under Award No. DMR 1121053 (D.V.K., D.M., J.G., N.A.L., C.J.H., and E.J.K.). SAXS work, as well as experimental interpretation and writing, were supported by the University of Chicago (D.V.K., M.V.T.), and by the Laboratory Directed Research and Development Program of the Argonne National Laboratory under U.S. Department of Energy Contract No. DE-AC02-06CH11357 (M.V.T.). The authors would like to thank Dr. Joseph Strzalka at Argonne National Laboratory for his help and support with the SAXS experiments. This research used resources of the Advanced Photon Source, a U.S. Department of Energy (DOE) Office of Science User Facility operated for the DOE Office of Science by Argonne National Laboratory under Contract No. DE-AC0206CH11357. Use of the Stanford Synchrotron Radiation Lightsource, SLAC National Accelerator Laboratory, is supported by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences under Contract No. DE-AC02-76SF00515.

■

REFERENCES

(1) Guvendiren, M.; Lu, H.; Burdick, J. Soft Matter 2012, 8, 260− 272. (2) Mano, J. F. Adv. Eng. Mater. 2008, 10, 515−527. 8032

dx.doi.org/10.1021/ma5017852 | Macromolecules 2014, 47, 8026−8032