Toughness Enhancement and Stick–Slip Tearing of Double-Network

Toughness Enhancement and Stick–Slip Tearing of Double-Network Hydrogels ... Graduate School of Science, Hokkaido University, Sapporo 060-0810, Japa...
0 downloads 6 Views 2MB Size
Article pubs.acs.org/Macromolecules

Toughness Enhancement and Stick−Slip Tearing of Double-Network Hydrogels in Poly(ethylene glycol) Solution Songmiao Liang,†,⊥ Jian Hu,‡ Zi Liang Wu,‡ Takayuki Kurokawa,†,§ and Jian Ping Gong*,† †

Faculty of Advanced Life Science, Hokkaido University, Sapporo 060-0810, Japan Division of Biological Sciences, Graduate School of Science, Hokkaido University, Sapporo 060-0810, Japan § Creative Research Institution, Hokkaido University, Sapporo 001-0021, Japan ‡

ABSTRACT: Using a thin film double-network (DN) hydrogels of 100 μm thickness, the behaviors of DN gels, from poly(2-acrylamido-2-methyl-1-propanesulfonic acid) (PAMPS) as the first network and polyacrylamide (PAAm) as the second network, swelled in poly(ethylene glycol) (PEG) aqueous solution of short molecular weight (Mw: 400 g/mol) are investigated. Comparing with the conventional DN gels that are swelled in water, the toughness of the DN gels swelled in PEG solution is largely enhanced, as revealed by the tensile test and tearing test. Furthermore, a transition in the crack propagation dynamics, from the steady-state mode in water to the stick−slip instability mode in PEG solution, is observed in the tearing test. The morphology around the crack front of the DN gels also changes with the presence of PEG, and characteristic triangular-like damage patterns are observed. Besides, spicule-like structures are observed on the fractured surface, either in water or in PEG solution. The PEG-induced behaviors are associated with the molecular interaction of PEG to polyacrylamide, the second ductile network of the DN gels.



INTRODUCTION Double-network (DN) hydrogels with an extraordinarily high toughness have been developed in our lab since 2003.1 DN gels are characterized by a special network structure consisting of two types of polymer components with contrast physical natures: the minor component is abundantly cross-linked polyelectrolytes (rigid skeleton), and the major component is comprised of poorly cross-linked neutral polymers (ductile substance). The former and the latter components are referred to as the first network and the second network, respectively, since the synthesis should be done in this order to realize this contrast structure. For example, DN gels from poly(2acrylamido-2-methyl-1-propanesulfonic acid) (PAMPS) as the first network and polyacrylamide (PAAm) as the second network showed superhigh tearing energy of 102−103 J/m2, which is 10−1000 times larger than that of the corresponding gels with a single network.2−4 Extensive studies have been focused on the toughening mechanism of the DN gels.3−11 It has been observed that, during the tensile process, the DN gels exhibit yielding and necking.5,8 The underlying yielding mechanism is the internal local damage of the brittle component before its macroscopic fracture.5 That is, softening of the gels occurred in the cost of large energy dissipation. This yielding phenomenon has also been found at the crack tip and contributes dominantly to the high tearing energy.6,8,11 Based on the necking phenomena and the large hysteresis of the DN gels, the theoretical prediction concerning the existence of the softened (damaged) zone at the crack tip vicinity and its dominant role to the propagation of the crack tip have been proposed by Brown and Tanaka.6,8 According to their models, © 2012 American Chemical Society

the damage zone was estimated on an order of hundreds of micrometer. These predictions were well in accordance with the experimental results that were obtained by the micromechanical measurement12 and the direct laser visualization.11 More recently, the surface morphology of the DN gels was investigated by coupling the ultrathin film technique and optical microscope.13 Both the tensile-induced necking region and the tearing-induced damage zone showed similar characteristic surface morphologies, which directly confirms that the sample experienced a tensile deformation at the crack front. According to the surface morphology, the experienced stress and strain at the crack front can be analyzed in terms of the preyielding zone (P-zone), yielding zone (Y-zone), and hardening zone (Hzone).13 From the above study, it is clarified that the rapture and/or the crack propagation of the DN gels occurs by two steps, namely, first the break of rigid skeleton, PAMPS network, and then the break of the ductile substance, PAAm network. The former mainly occurs in the necking region which leads to a dramatic decrease in the Young’s modulus of the gels, while the latter occurs at the hardening region which determines the ultimate fracture strain and stress as well as the final crack propagation of the gels. Accordingly, the toughening of the material is built on the delicate mechanical balance between the two networks, whereupon the extensibility and final fracture stress of the second ductile network should play a crucial role. Received: February 21, 2012 Revised: May 9, 2012 Published: May 17, 2012 4758

dx.doi.org/10.1021/ma300357f | Macromolecules 2012, 45, 4758−4763

Macromolecules

Article

2 sizes ((w) 100 ± 10 μm, (d) 7.5 mm, (l) 50 mm, the length of the initial notch is 20 mm), with a gel cutting machine (Dumbbell Co., Ltd.). The test was performed in model I geometry (Figure 2A). The two arms of the test sample were clamped, and the one was pulled upward at a velocity V, while the other was maintained stationary. The tearing force F was recorded. The tearing energy T is calculated from the average value of F at steady state during tearing and the gel thickness w according to the relation15

In this work, we report the discovery on the poly(ethylene glycol) (PEG)-induced toughness enhancement of PAMPS/ PAAm DN gels. Both the tensile and tearing tests have shown that DN gels swelled in PEG aqueous solution have a better mechanical performance than those swelled in pure water. Furthermore, the crack dynamics of the gels in PEG solution also changes. Based on the ultrathin film technique and optical microscope, we also investigate the morphology at the crack front and its correlation with the crack dynamics of the gels in PEG solution. We discuss these PEG-induced changes in terms of the molecular interactions between PEG and the second network PAAm.



T = 2F /w

(1)

Here, we adopt the traditional relation in the literature to estimate the tearing energy T, which is higher by a factor of 4 than that used in our early papers on DN gels.2−4 The cross-head velocity V, which related to the tearing velocity as V/2, was set at the range of 10−100 mm/ min. Each tearing energy T in Figure 3 was averaged from 5 parallel tests, and the “error bars” denote the deviation of the test values from their average value of the 5 parallel tests. Observation of the Structure at the Crack Tip. The samples were shaped as that used for the above tearing test. The crack tips of these samples were created by tearing them under the tearing velocity of 60 mm/min. The tearing distance was about 10 mm for each sample. After undergoing the pretearing treatment, the resultant samples were immersed in the corresponding solvent for reswelling and relaxation. During the observation process, the reswollen samples were placed on glass substrate and slightly stretched and fixed to make the crack opening visible. So the crack opening angles observed in Figure 4 is not directly related to the in situ ones or the completely free-standing ones. The fracture structure at the crack tip of the samples was visualized by a phase contrast microscope equipped with ×5, ×10, ×20, and ×40 objective lens.

EXPERIMENTAL SECTION

Materials. 2-Acrylamido-2-methylpropanesulfonic acid (AMPS; Tokyo Kasei Co., Ltd.), as monomer, was used as received. Acrylamide (AAm; Junsei Chemical Co. Ltd.), as monomer, and N,N′methylenebis(acrylamide) (MBAA; Tokyo Kasei Co., Ltd.), as crosslinker, were recrystallized from chloroform and ethanol, respectively, before use. 2-Oxoglutaric acid (OA; Wako Pure Chemical Industries, Ltd.), as radical initiator for the gelation reactions, was used as received. Poly(ethylene glycol) (PEG) (Wako Pure Chemical Industries, Ltd.) with molecular weight 400 g/mol (PEG-400) was used as received. Milli-Q (18.3 MΩ) water was used in all experiments. Synthesis of the Thin Film DN Gels. The procedure for creating the ultrathin film poly(2-acrylamido-2-methylpropanesulfonic acid)/ polyacrylamide (PAMPS/PAAm) double-network (DN) gels was performed by multistep UV-light-initiated polymerization (UVIP).14 Using two glass plates spaced with 50 μm polyethylene film, we first created the ultrathin PAMPS gels by the UVIP from a precursor aqueous containing 1 M AMPS, 0.04 M MBAA, and 0.001 M OA. Then, the PAMPS gels were immersed in AAm (4 M)/OA (1.67 × 10−4 M) aqueous solution, containing 0.08 M NaCl, to obtain a controlled-swelling state. Following this step, another UVIP to the partially swelled PAMPS gels was performed to in situ introduce PAAm chains to the PAMPS network. Finally, the enhanced PAMPS gels were washed until free from NaCl by distilled water. They were then immersed in AAm (2 M)/OA (1.67 × 10−4 M solution for 10 h, and the UVIP was applied to obtain the DN gels. The obtained DN gels were stored in water and 5−30 wt % PEG aqueous solution for 6 days to reach swelling equilibrium. The thickness of the swelled DN gels was 100 ± 10 μm. No obvious change in the thickness of the samples was observed when immersed in these solvents. The DN gels swelled by water and water/PEG mixture were named as DN-W and DN-P, respectively. According to the PEG content, DN-P gels were further differentiated as DN-P-5, DN-P-10, and DN-P-30, which are corresponding to the gels swelled by 5, 10, and 30 wt % PEG aqueous solutions, respectively. Synthesis of Thin Film PAAm Hydrogels. PAAm hydrogels were synthesized by using UV-initiated radical polymerization. Solutions containing 4 M AAm, 0.0004 M OA, and 0.004 M MBAA were injected into the glass chambers spaced with 50 μm polyethylene film. Polymerization was performed under UV light for 8 h. The obtained PAAm hydrogel films were then fully swelled by water, 5 wt %, and 10 wt % PEG solutions, respectively. Correspondingly, the swelled PAAm gels were named as PAAm-W, PAAm-P-5, and PAAmP-10. The thickness of the PAAm gels was about 70 μm. Tensile Test. Tensile mechanical properties of the DN gels and PAAm gels were measured with a commercial test machine (Tensilon RT-1150A, Orientec Co.) in air. The samples were cut into a dumbbell shape standardized as JIS-K6251-7 sizes (length 51 mm, width 4 mm, gauge length 20 mm) with a gel cutting machine (Dumbbell Co., Ltd.). The sample length between two clamps was ∼30 mm. The stress−strain curves were recorded while the sample gel was stretched at a constant velocity of 40 mm/min. Tearing Test. The tearing test was performed with a commercial test machine (Tensilon RTC-1150A, Orientec Co.) in air. The DN gels were cut into the shape, which has the standardized JIS-K6252 1/



RESULTS AND DISCUSSION Figure 1 shows the tensile stress−strain curves of the DN gels and the PAAm gels being swelled in water as well as in PEG aqueous solutions of various concentrations. Comparing with the behavior in water, both the DN gels and PAAm gels exhibit much enhanced fracture stress and strain in the presence of PEG, while their stress−strain curves hardly change at low strain regions, especially before necking. As the load is sustained by the brittle skeleton of PAMPS network at the initial deformation, and it transforms to the ductile substance of PAAm after necking, we preliminarily assume, from the result of Figure 1, that PEG molecules reinforce the second network of PAAm by increasing its extensibility to fracture. Figure 2B presents the tearing force F as a function of the displacement x for the samples swelled by water. The tearing force F of the DN-W gels, after the tearing induction period, keeps almost constant during the propagation of the crack. It reflects a relatively stable and continuous crack dynamics of the gels. This tearing behavior is well consistent with that observed in the bulk DN system2 and thus implies that the crack dynamics of the DN gels has little dependence on their thickness over the experimental range. Further, at the high tearing velocity, the average F slightly increases. However, this stable crack propagation disappears for the DN gels swelled in 5 wt % PEG solution. As shown in Figure 2C, a stick−slip like behavior in the F(x) curve of DN-P-5 is observed, especially at high velocity. For the curves performed at V = 60 and 80 mm/ min, two distinct processes, which denote a moderate, positiveslope increase and an abrupt, negative-slope decrease in F(x), are observed. When the slope of the F(x) curve is positive, the crack front remains stationary and the deformation around the front is increased by the pulling action. The decrease in F(x), on the other hand, corresponded to the quick propagation of the crack front. That is, the maximum force and minimum force correspond to respectively the crack initiation and arrest. 4759

dx.doi.org/10.1021/ma300357f | Macromolecules 2012, 45, 4758−4763

Macromolecules

Article

Similar stick−slip like behavior in tearing force is observed in the DN-P-10 sample (Figure 2D) (V = 60 mm/min), but with much shorter period and smaller amplitude than that of DN-P5. As shown by the inset in Figure 2D, the distance between the two neighboring F(x) peaks, namely, the step length a, ranges from 0.6 to 2.1 mm. This value is 5 times smaller than that observed for DN-P-5 sample of 2−10 mm. Furthermore, even in the single increasing period of F(x), for example, from 1 to 2, superimposed small stick−slips can be observed. In the positive-dependence region of F on x, the average slope (for example, 1 → 2 in inset of Figure 2D), k = ΔF/Δx, is unsusceptible to the PEG concentration. The average value of k for the DN-P-10 gel is about 0.36 N/m, which is only slightly higher than 0.28 of the DN-P-5 gel. Figure 3 presents the cross-head velocity dependence of the average tearing energy T of the gels in water and in PEG

Figure 1. Stress−strain curves of the thin film DN gel (A) and PAAm gel (B) swelled in water and in PEG solution with various concentrations: (a) DN-W, (b) DN-P-5, (c) DN-P-10. (d) PAAmW, (e) PAAm-P-5, (f) PAAm-P-10.

Figure 3. Cross-head velocity V dependence of tearing energy of DNW gel (●), DN-P-5 gel (■), and DN-P-10 gel (▲). The tearing velocity at the crack tip is half of the cross-head velocity V.

Figure 2. (A) Illustration of the tearing test geometry (model I). (B−D) Tearing force profiles F(x) against cross-head displacement x: (B) DN-W gels at the cross-head rate V of 20 mm/min (a), 60 mm/min (b), and 80 mm/min (c); (C) DN-P-5 gels at the cross-head rate V of 20 mm/min (a), 60 mm/min (b), and 80 mm/min (c); and (D) DN-P-10 gels at the cross-head rate V of 60 mm/min. The inset in (D) presents the magnification of the curve segment located within the tearing displacement of x = 51−60 mm. 4760

dx.doi.org/10.1021/ma300357f | Macromolecules 2012, 45, 4758−4763

Macromolecules

Article

Figure 4. Morphologies of the damage zone at the crack fronts of the DN-W gel (a, b), the DN-P-5 gel (c, d), and the DN-P-10 gel (e, f). The observation was performed on the reswollen samples after the tearing test.

solution. On the whole, T shows a moderate increase with the increase of V when V is not very high. This result suggests that, even in the DN gels, the tearing fracture energy is weakly dependent on the tearing velocity, which is in accordance with the results obtained in the bulk DN gels.2 In addition, there is 100−200 J/m2 increase in T with the addition of PEG in the DN gel, implying that PEG can enhance the fracture propagation resistance of the DN gels, in agreement with the result in Figure 1. The stick−slip instability of the tearing test has often been observed in the rubbers and thermoplastic elastomers16−18 and recently was also observed in the poly(vinyl alcohol) hydrogels.19 The stick−slip instability occurs when the tearing force (or fracture energy) for stationary cracks is larger than that for moving crack fronts.16 The stretching-induced formation of the tough and ordered structure (crystallization, etc.) at the crack front vicinity is accounted for the stick−slip tearing in these systems. However, these mechanisms are likely inapplicable for the DN-P gels as a result of the noncrystal/ glassy properties of the PAAm network and the two-step crack propagation process in the DN-P gels. As has been clarified, the fracture of the DN gels occurs by the break of the densely cross-linked PAMPS network at low strain (ε < 500%) and the break of the slightly cross-linked PAAm network at high strain (ε > 1000%).5 Thus, the final propagation of the crack is dominantly controlled by the second step, the PAAm breaking. This result, combined with the fact that in PEG solution the

toughness of pure PAAm gel is also enhanced, implies that the enhancement in the toughness and stick−slip behavior is associated with the interaction of PEG with PAAm in the DN gels. As reported in the literature, the viscosity of aqueous solution of PEG−PAAm mixture is lower than the ideal case, indicating that interaction between PEG and PAAm form compact structure.20 It is explained that PEG enhances the hydrogen bonding between intra/inter-PAAm chains by forming adsorption on the surface of the PAAm random coil.20 It is also reported that PAAm hydrogels polymerized in PEG aqueous solution forms heterogeneous structures.21,22 For example, PAAm gels synthesized in 17.5 wt % PEG-400 become turbid.21 In the present case, the PAAm network of the DN gels might also form some heterogeneous structures in the 5 and 10 wt % PEG-400 aqueous solution, although the heterogeneous structures are too small to be optically visible. Since the cross-linking density of the PAAm network was very low (5 × 10−2 mol % in relative to PAAm repeated units),23 the PAAm partial chains are considered to be favorable for the formation of the heterogeneous structures. We consider that these heterogeneous structures, if exist, can act as pining center to cause the stick−slip crack dynamics due to the enhanced fracture resistance in these structures. We should point out that the single PAAm hydrogel prepared with the same formulation as the second network of the DN gel was too weak to perform the tearing test either in water or in PEG solution in the present study. 4761

dx.doi.org/10.1021/ma300357f | Macromolecules 2012, 45, 4758−4763

Macromolecules

Article

Figure 5. Cross-sectional morphologies of the fractured surfaces for the DN-W gel (a−c) and the DN-P-5 gel (d−f).

presence of PEG both the tearing dynamics and the structure of the damage zone in the DN gels are substantially changed. However, the damage structure at the crack front strongly depends on the PEG concentration. As shown in Figure 4e, the crack front of the DN-P-10 gel has homogeneous wrinkle-like structure, similar to that of gel swelled in water. The wrinkles, comparing to that observed in the DN-W gel, appeared to be larger in both length and width. The overall width of the damage zone is about 600 μm, which is close to that of the DNP-5 gel and about twice larger than that of the DN-W gel. Therefore, the large difference in the damage zone morphologies between the DN-P-5 gel and the DN-P-10 gel well corresponds to the behaviors in tearing test shown in Figure 2. That is, the large amplitude and long periodicity of the stick−slip tearing force of DN-P-5 gel are related to the large multitriangular patterns, while the strengthened tearing energy T of the DN-P gels (Figure 3) arises from the expanded damage zone. The same fracture structure as that of DN-P-10 was also observed for the DN-P-30 gel. We next study the morphology of the fractured surfaces in the DN-W gel and the DN-P-5 gel. As shown in Figure 5, a comb-like structure, which is composed of many spicules, was observed on the fractured surfaces for both of the DN-W and DN-P-5 gels. In the case of the DN-W gel, the distribution and length of the spicules is more random and uneven as compared to that of the DN-P-5 gel. The average length of the spicules at the edge of the DN-W cross section is about 100 μm, which is about twice larger than ∼45 μm of the DN-P-5 gel. To clarify the origin of the comb-like structure, we further performed the same observation on the fractured surface of pure PAMPS gel and DN gels with different PAAm content (0.5−4 M). Similar comb-like structure is observed for all of the DN gels but not for pure PAMPS gel. Moreover, the spicule size increases with the increase of the PAAm content. This result suggests that the

Looking at the above difference in the crack dynamics between the DN-W gel and the DN-P gel, we further studied the damage structure at their crack front vicinity by a phase contrast optical microscope, and the results are shown in Figure 4. As seen in Figure 4a, at a low magnification, a smooth surface was observed even near the crack front of the DN-W gel. However, under a high magnification, a damage zone of ∼300 μm width with wrinkle-like region was observed at the crack front as shown in Figure 4b. The wrinkle-like structure is homogeneously distributed and mutually connected, which denotes a local necking behavior of the DN gels.13 Thus, the wrinkle-like region is the yielding zone (Y-zone) and the region close to the fractured surface is strain hardening zone (Hzone).13 It should be noted that the presence of the white and dark regions above the H-zone (Figure 4b) is due to the light scattering or reflection at the edge of the crack opening, and it denotes no any real structure. Strikingly, as seen in Figure 4c, the DN-P-5 gel shows a completely different morphology at the crack front, where a structure of large multitriangular patterns is observed close to the fractured surface even at the low magnification. The average width of the damage zone at the crack front is about 600 μm, much larger than that of the DN-W gel. Observation at high magnification reveals that, besides its relatively smooth and thin center, each triangular pattern has two borders with the branchlike structure (Figure 4d, region 2). The branches are considered as the precursor of the triangular damage arising from the break of the PAMPS network. No obvious damage structure is observed in the region between the two triangular regions, implying little damage of the PAMPS network in this region (Figure 4d, region 3). Moreover, wrinkle-like structure can be observed inside the triangular patterns (inset in Figure 4d). This wrinkle-like structure probably arises from the damage of the PAMPS. The above results indicate that in the 4762

dx.doi.org/10.1021/ma300357f | Macromolecules 2012, 45, 4758−4763

Macromolecules

Article

(10) Miquelard-Garnier, G.; Demoures, S.; Creton, C.; Hourdet, D. Macromolecules 2006, 39, 8128. (11) Yu, Q. M.; Tanaka, Y.; Furukawa, H.; Kurokawa, T.; Gong, J. P. Macromolecules 2009, 42, 3852. (12) Tanaka, Y.; Kawauchi, Y.; Kurokawa, T.; Furukawa, H.; Okajima, T.; Gong, J. P. Macromol. Rapid Commun. 2008, 29, 1514. (13) Liang, S.; Wu, Z. L.; Hu, J.; Kurokawa, T.; Yu, Q. M.; Gong, J. P. Macromolecules 2011, 44 (8), 3016. (14) Liang, S.; Yu, Q. M.; Yin, H.; Kurokawa, T.; Gong, J. P. Chem. Commun. 2009, 48, 7518. (15) Stacer, R. G.; Von Meerwall, E. D.; Kelley, F. N. Rubber Chem. Technol. 1985, 58, 913. (16) Greensmith, H. W.; Thomas, A. G. J. Polym. Sci. 1955, 18, 189. (17) Stacer, R. G.; Yanyo, L. C.; Kelley, F. N. Rubber Chem. Technol. 1985, 58, 421. (18) Wang, C.; Chang, C. I. J. Appl. Polym. Sci. 2000, 75, 1033. (19) Tanaka, Y.; Abe, H.; Kurokawa, T.; Furukawa, H.; Gong, J. P. Macromolecules 2009, 42, 5425. (20) Lu, T.; Shan, G. R. Acta Polym. Sin. 2010, 2, 156. (21) Kwok, A. Y.; Prime, E. L.; Qiao, G. G.; Solomon, D. H. Polymer 2003, 44, 7335. (22) Caykara, T.; Bulut, M.; Dilsiz, N.; Akyuz, Y. J. Macromol. Sci., Part A: Pure Appl. Chem. 2006, 43, 889. (23) Nakajima, T.; Furukawa, H.; Tanaka, Y.; Kurokawa, T.; Osada, Y.; Gong, J. P. Macromolecules 2009, 42, 2184. (24) Baumberger, T.; Caroli, C.; Martina, D. Nat. Mater. 2006, 5, 552.

comb-like structure is due to the break of the PAAm network during the strong stretching process. The difference in the spicule length and distribution between the DN-W and DN-P gels could be attributed to the presence of PEG. The enhanced intra/interchain interaction of PAAm due to the presence of nonsolvent PEG leads to an increased viscoelastic effect of the ductile component in the DN gels and, accordingly, an enhanced pull-out resistance of PAAm during fracture.24



CONCLUSION The strength, toughness, and crack dynamics of the DN gels swelled in PEG-400 aqueous solution have been studied by using the ultrathin film technique and phase contrast microscope. PEG not only enhances the strength and toughness of the DN gels but also changes the crack dynamics and the fracture structure of the DN gels. Striking transition of the tearing behavior from the stable and smooth tearing to the stick−slip like tearing mode of the DN gels is observed with the introduction of PEG. A possible explanation for this transition is the interaction between PEG and PAAm, which forms some heterogeneous structure in the DN-P gels. Compared with the wrinkle-like fracture structure of the DN-W gels, the characteristic structure in the damage zone of the DN-P gels was more diverse with highly expanded thickness and shape. Both the multitriangular structure and the wrinkle-like structure were visualized in the crack front vicinity regardless of the similar tearing mode. Moreover, spicule-like structure, for the first time, is observed on the fractured surface of the DN gels, either swelled in water or in PEG solution. This spicule-like structure shows PEG dependence in the length and distribution and is formed by the stretching break of the PAAm network.



AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Present Address ⊥

Vontron Membrane Technology Co., LTD, Guiyang, Guizhou 550018, China.

Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS This research was financially supported by a Grant-in-Aid for Specially Promoted Research (No. 18002002) from the Ministry of Education, Science, Sports and Culture of Japan.



REFERENCES

(1) Gong, J. P.; Katsuyama, Y.; Kurokawa, T.; Osada, Y. Adv. Mater. 2003, 15, 1155. (2) Tanaka, Y.; Kuwabara, R.; Na, Y. H.; Kurokawa, T.; Gong, J. P.; Osada, Y. J. Phys. Chem. B 2005, 109, 11559. (3) Tsukeshiba, H.; Huang, M.; Na, Y.-H.; Kurokawa, T.; Kuwabara, R.; Tanaka, Y.; Furukawa, H.; Osada, Y.; Gong, J. P. J. Phys.Chem. B 2005, 109, 16304. (4) Huang, M.; Furukawa, H.; Tanaka, Y.; Nakajima, T.; Osada, Y.; Gong, J. P. Macromolecules 2007, 40, 6658. (5) Na, Y. H.; Tanaka, Y.; Kawauchi, Y.; Furukawa, H.; Sumiyoshi, T.; Gong, J. P.; Osada, Y. Macromolecules 2006, 39, 4641. (6) Brown, H. R. Macromolecules 2007, 40, 3815. (7) Shull, K. R. J. Polym. Sci., Part B: Polym. Phys. 2006, 44, 3436. (8) Tanaka, Y. Europhys. Lett. 2007, 78, 56005. (9) Webber, R. E.; Creton, C.; Brown, H. R.; Gong, J. P. Macromolecules 2007, 40, 2919. 4763

dx.doi.org/10.1021/ma300357f | Macromolecules 2012, 45, 4758−4763