Strain hardening and highly resilient hydrogels crosslinked by chain-extended reactive pseudo-polyrotaxane

Abstract

Strain hardening and high resilience are two unique mechanical characteristics of many soft biological hydrogels. However, these properties, especially the strain hardening behaviour, are generally not seen in synthetic polymer hydrogels. Here, hydrogels are prepared by free-radical copolymerization of acrylamide and chain-extended vinyl-modified pseudo-polyrotaxane, which acts as a multifunctional crosslinker. The reactive pseudo-polyrotaxane is based on a β-cyclodextrin monomer and amine-terminated PEG–PPG–PEG (Pluronic F127). The obtained hydrogels can be stretched from 10 to more than 26 times their original length before breaking and withstand a compression strain of 95% and even 98% without fracture. Tensile stretching tests show obvious strain hardening behaviours in high stretching and compression deformation regimes. The strain hardening behaviour in stretching deformation is considered to be the orientation and aggregation of the movable crosslinkers along the axial polymer backbone. Moreover, the formation of a second supramolecular network due to the chain-extension effect may also be responsible for it. Highly resilient behaviours with almost no hysteresis and residual strains are also observed even with a maximum strain of λ = 12 because of the inherent freely movable character of the crosslinkers.

---

Introduction

The fast-growing area of hydrogels is of the utmost importance for fields such as artificial implants, biomedical devices, tissue engineering and regenerative medicine, due to their unique properties such as similar flexibility, high water content, and molecular diffusion to natural soft tissues.^1–5 Thus, during the past years, various functional hydrogels, including stimuli-responsive gels, self-healing gels, and shape-memory gels, were developed for different bioapplications.^6–11 However, classical hydrogels are often mechanically weak and brittle, just like jelly or paste. It is these disadvantages that severely limit their great potential applications as ideal structural biomaterials, where mechanical performance is required. Therefore, from the beginning of this century, several kinds of hydrogels, such as sliding gels,¹² nanocomposite gels,¹³ double-network gels,¹⁴ macromolecular microsphere composite gels,¹⁵ tetra-PEG-based gels,¹⁶ and our polyurethane-urea supramolecular (PUUS) hydrogels,^10,11,17 were developed to meet the mechanical requirements.

However, there are still many other special mechanical characteristics of biological hydrogels that deserve material scientists' attention. For example, many kinds of biological protein/polymer-based hydrogels demonstrate unique viscoelastic properties differing from synthetic polymer hydrogels. This unique viscoelasticity of biopolymer-based hydrogels is fundamental to their biological function and the maintenance of normal physiology. For instance, actin, collagen, fibrin, vimentin, and neurofilament-based hydrogels show strain hardening behaviour, i.e. a sharp increase in material stiffness at large strains, thereby preventing large deformations that could threaten tissue integrity.^18–20 Moreover, elastin and resilin-based biological hydrogels can be reversibly deformed without energy loss. These hydrogels are also well known to have resilience, an important feature developed by natural selection that facilitates repeated movement.^21–23 However, these protein-based synthetic hydrogels usually do not show high ductility, which significantly limits the scope of their potential biomedical applications.

Therefore, synthetic hydrogels with similar viscoelastic properties to biological soft tissues/hydrogels are of high importance and interest to the principles, design and application of current hydrogel-based biomaterials and many other soft materials such as thermoplastic elastomers. Highly resilient behaviour was recently observed in several synthetic hydrogel systems.^24–28 Strain hardening was mainly observed in protein/biopolymer-based hydrogels,^{18–20,29,30} and was recently observed in nanocomposite gels for the first time. Although strain hardening behaviour has been widely observed in the compression deformation condition,^{14–17,31,32} synthetic hydrogels with both strain hardening and high-resilience behaviours in both stretching and compression processes have rarely been reported because this may require complicated chemical synthesis and rational molecular design.

As one of the famous mechanically strong hydrogels, sliding gels have been widely studied due to their unique topological characteristics and various potential applications.^12,33 However, traditional sliding hydrogels are always prepared by crosslinking of polyrotaxane (PR) based on PEG and α-cyclodextrin (α-CD) via complicated procedures. To obtain pure PR, excess α-CD has to be completely removed before crosslinking to avoid non-effective crosslinking between free α-CD and the crosslinking reagent used, and this often gives low conversion of PR. At the same time, the crosslinking reaction can often only be conducted in a very limited number of solvents because of the poor solubility of the formed PR. Thus, hydrogels can only be obtained after a water-exchange process to remove the organic solvents/inorganic compounds. On the other hand, pseudo-polyrotaxane (PPR) and PR based on PPG or PEG–PPG–PEG triblock copolymers and β-cyclodextrin (β-CD) have been studied a long time ago.^34–38 However, to our knowledge, sliding gels using β-CD as the moving crosslinker are not yet reported. This may mainly be because of the difficulty in the preparation of β-CD-based PR due to its cavity being so large that it is difficult to find suitable stopper groups.³⁷

In this work, we present a novel kind of synthetic sliding hydrogel based on β-CD and PPR via a simple modular method. The hydrogels are fabricated by UV-initiated free-radical copolymerization of acrylamide and vinyl-modified PPR constructed by β-CD monomer and amine-terminated PEG–PPG–PEG triblock copolymer (Scheme 1). Particularly, the hydrogels not only show good ductility, but also have excellent strain hardening and high-resilience properties.

Scheme 1 Synthetic procedure of the hydrogels prepared by free-radical copolymerization.

Experimental section

Materials

Pluronic F127 (PEG–PPG–PEG) was obtained from Sigma-Aldrich. Polyethylene glycol (PEG, Sinopharm Chemical Reagent Co. Ltd) with average M_n = 2000 g mol⁻¹ was dried at 80 °C under vacuum in the presence of P₂O₅ overnight before use. β-Cyclodextrin (β-CD), N,N-carbonyldiimidazole (CDI), ethylenediamine and acrylamide were purchased from Sinopharm Chemical Reagent Co. Ltd. The photoinitiator α-ketoglutaric acid (α-ka, 99%) was purchased from Fluka. Isophorone diisocyanate (IPDI) was purchased from Aladdin. 2-Isocyanatoethyl acrylate (AOI) was kindly donated by Showa Denko Company. The catalyst dibutyltin dilaurate (DBTDL) was purchased from TCI. N,N-Dimethylformamide (DMF) and tetrahydrofuran (THF) were dried by 4 Å molecular sieves before use.

Synthesis

Details of the synthesis of H₂N–PEG–PPG–PEG–NH₂ (AF-127), AOI–β-CD and OCN–PEG–NCO can be found in the ESI.†

Hydrogel preparation

As shown in Scheme 1, the hydrogels were prepared by UV-irradiated free-radical copolymerization of acrylamide and chain-extended vinyl-modified PPR, which acts as a multifunctional supramolecular crosslinker. Here, the PPR is based on AOI–β-CD monomer and amino-terminated Pluronic F127 (AF-127). A typical synthetic procedure of the hydrogel is as follows. Firstly, AOI–β-CD was added to an aqueous solution of AF-127 (0.5, 1.0 or 2.5 wt%), and the molar ratio of AOI–β-CD to AF-127 was 5 : 1, 10 : 1, or 15 : 1. The mixture was then stirred for 2 days at room temperature to let AOI–β-CD fully thread onto the PPG chains. Secondly, isocyanate end-capped PEG (OCN–PEG–NCO) was added to the solution to yield chain-extended PPR due to the quick reaction between –NH₂ and –NCO. Finally, acrylamide and α-ka (photoinitiator) were added to the chain-extended PPR solution, and the mixed solutions were then transferred into transparent plastic syringes and sealed. The sealed plastic tubes were irradiated by 365 nm UV light (40 W, the distance between the samples and the lamp was 10 cm) for 1 h. For all the samples, the concentrations of acrylamide and α-ka were fixed at 10.0 wt% and 0.2 wt%, respectively. The obtained hydrogels are named as x%–yCD, where x and y denote the weight concentration of AF-127 and the molar ratio of AOI–β-CD to AF-127. For example, 1.0%–10CD indicates the concentration of AF-127 is 1.0 wt% and the molar ratio of AOI–β-CD to AF-127 is 10.

Characterizations

¹H NMR spectra in D₂O were recorded using a 400 MHz Inova NMR spectrometer. Rheology measurements were performed on a HAAKE rheometer (RS 6000) with a parallel plate accessory (20 mm in diameter). The existence and extent of the linear viscoelastic regime were determined by measuring the dynamic shear storage modulus (G′) and loss modulus (G′′) as a function of strain (0.001 < γ < 10) at an angular frequency of 6.28 rad s⁻¹. All the measurements were carried out within the linear viscoelastic range, where G′ and G′′ are independent of strain. Tensile and compression tests were carried out on a universal tensile test machine (KJ-1065B, Kejian-tech) with a 50 N loading cell; the cross-head speeds of the tensile and compression measurements were 50 mm min⁻¹ and 3 mm min⁻¹, respectively. The normal force F and the gap l (for the tensile test) or h (for the compression test) between the two clamps or plates were recorded. The true stress σ_true was adopted as in a previous report.³¹ The nominal stress σ_nominal was estimated as σ_nominal = F/S₀, where S₀ is the original cross-section area of the gel before deformation, σ_true = σ_nominal × l/l₀ or σ_true = σ_nominal × h/h₀ (in the compression case). Morphological study of the cross-section of the freeze-dried hydrogel sample was carried out on a Hitachi S-4700 scanning electron microscope (SEM). The sample for SEM was sputter-coated with a thin layer of Au prior to the observations to prevent sample charging problems.

Results and discussion

As shown in Scheme 1, the hydrogels were prepared by simple free-radical copolymerization of acrylamide and chain-extended vinyl-modified PPR under UV irradiation. Here, α-ka (α-ketoglutaric acid) was used as the photoinitiator. AF-127 was synthesized from Pluronic F127 according to previous reports.³⁵ Details of the synthesis of AOI–β-CD and isocyanate-terminated PEG (OCN–PEG–NCO) are given in ESI.† For all the hydrogel samples, the concentration of acrylamide was fixed at 10.0 wt%. The obtained hydrogels were named as x%–yCD, where x and y denote the weight concentration of AF-127 and the molar ratio of AOI–β-CD to AF-127. For example, 1.0%–10CD indicates the concentration of AF-127 used is 1.0 wt% and the molar ratio of AOI–β-CD to AF-127 is 10. Of special note, the present strategy is totally different from Takeoka's novel sliding gels, which were prepared by copolymerization of N-isopropylacrylamide and adamantane end-capped vinyl-modified PEG/α-CD-based PR in DMSO.³⁹ Compared with traditional and Takeoka's novel sliding gels, the advantages of the present method as shown in Scheme 1 include: (1) the whole procedure was carried out in one pot with three steps using water as solvent; (2) instead of attachment of a bulky end group giving low conversion of PR, chain extension was used to obtain long-chain multi-block PPR (chain-extended PPR) such that there were still enough threaded AOI–β-CD monomers on the polymer main chain in the gelation process; (3) there was no need to remove the free AOI–β-CD monomers in the first two steps, because they can copolymerize with acrylamide in the last step without yielding crosslinking points. A SEM micrograph showing a cross-section of the freeze-dried 1.0%–15CD hydrogel sample can be found in Fig. S2 in ESI.†

Rheology experiments (Fig. 1 and S3 in ESI†) show that the storage modulus (G′) at 10 rad s⁻¹ of the resulting hydrogel ranges from 500 Pa to 1000 Pa, these are comparable to the traditional sliding hydrogels (500–2000 Pa).¹² At the same time, G′ depends both on the concentration of AF-127 and the molar ratio of AOI–β-CD to AF-127. As shown in Fig. 1a and S3a and b in ESI,† at a fixed concentration of AF-127, G′ increases with an increasing concentration of AOI–β-CD. Similar trends are also observed with an increasing concentration of AF-127 at a fixed molar ratio of AOI–β-CD to AF-127 (Fig. 1b and S3c and d in ESI†). This is all because of the increased crosslinking density of the obtained hydrogels with increasing concentration of AOI–β-CD (Fig. 1a) or AF-127 (Fig. 1b).

Fig. 1 Shear storage modulus of the obtained hydrogels: (a) the concentration of AF-127 is fixed at 1.0 wt%, and the molar ratios of AOI–β-CD to AF-127 are 5, 10 and 15; (b) the molar ratio of AOI–β-CD to AF-127 is 10, and the concentrations of AF-127 are 0.5, 1.0 and 2.5 wt%. For all the samples, the concentration of acrylamide is 10 wt%.

Tensile tests were carried out to further investigate the viscoelastic property of the resulting hydrogels. The elongation ratio λ is taken as the deformed length l related to the original length l₀, λ = l/l₀, the true stress is adopted as σ_true = σ_nominal × l/l₀ according to a previous report.³¹ As shown in Fig. 2 and S4 in ESI,† the hydrogels show excellent ductility and can be stretched from 9 to more than 26 times their original length before breaking. At the same time, at a fixed concentration of AF-127 (Fig. 2a and S4a and b in ESI†), σ_true increases with an increasing concentration of AOI–β-CD, while λ at breaking decreases. Similar trends are also observed with an increasing concentration of AF-127 at a fixed molar ratio of AOI–β-CD to AF-127 (Fig. 2b and S4c and d in ESI†). This is all because of the increased crosslinking density of the obtained hydrogels with an increasing concentration of AOI–β-CD or AF-127, and also in accordance with the results in Fig. 1. The network's strength increases with increasing crosslinking density, but its extendibility will decrease.^25,28

Fig. 2 Stretching tensile test curves of the obtained hydrogels as shown in Fig. 1. λ = l/l₀, l and l₀ are the stretched and original lengths of the sample, respectively.

Intriguingly, obvious strain hardening behaviour, a sharp increase in stress in a large strain regime is observed in Fig. 2 and S4 in ESI.† For example, in the case of sample 1%–15CD (blue curve in Fig. 2a), σ_true slowly and almost linearly increased to 120 kPa in a large strain region from λ = 1 to 9, while it abruptly increases to 2400 kPa at λ = 15, which is 20 times that at λ = 9. Furthermore, as the results shown in Fig. 2 and S4 in ESI,† the degree of the strain hardening phenomena also depends both on the concentration of AF-127 and the molar ratio of AOI–β-CD to AF-127. Although the underlying mechanism is still not fully understood, the possible mechanism of the strain hardening behaviour may be due to the following issues: (1) similar to classic sliding gels,⁴⁰ the sliding crosslinkers (β-CD) can orientate along the PPG chains and form aggregates while stretching; however, β-CD is more likely to locate on the middle PPG blocks due to stronger hydrophobic interactions rather than on the PEG chains,^35,36,38 and thus much higher stress is needed for the further sliding of the formed aggregates along the PEG chains in a larger strain regime. This is different from traditional sliding gels, where the formed aggregates of α-CD cannot be further moved due to the bulky end group leading to breaking of the gels; (2) the formation of a second supramolecular network due to the chain-extension effect may also be responsible for it. This is because the chain-extension effect not only yields strong hydrogen-bonding units (urea) due to the reaction between –NCO and –NH₂, but also greatly improves the chain length of the axial polymer backbone. Thus, a second weak supramolecular network forms due to the strong hydrogen-bonding interaction between urea groups and the enhanced chain entanglement because of increasing chain length. Similar to the function of the second network in the DN gels, the second weak network greatly improves the mechanical property of the resulting hydrogels in a large deformation regime.¹⁴ This is different from traditional sliding gels, in which the linear polymer backbone only acts as an axis for the sliding of the threaded cyclic molecules. On the other hand, control experiments (Fig. S5 in ESI†) further show that the chain-extension effect does not have an obvious influence on the elongation ratio of the hydrogels, but it greatly improves the strain hardening property in a higher strain regime. Moreover, although strain hardening behaviour has been observed in the compressing deformation condition of many other hydrogels, including sliding hydrogels,^{14–17,31,32} to our knowledge, this is the first time that such obvious strain hardening behaviour in stretching deformation of sliding hydrogels has been observed.

Another interesting and important characteristic of the obtained hydrogels is their highly resilient property. For simplicity, sample 1%–15CD with the best strain hardening behaviour is chosen for the cyclic tensile tests. As shown in Fig. 3, with a maximum strain of λ = 12, the 5-cycle loading-unloading curves perfectly overlap with each other and show negligible loops (Fig. 3a), indicating that there is almost no hysteresis or permanent network damage in the cycles. It is necessary to notice that there is also almost no decrease in the maximum stress of the immediately following 4 cycles in comparison with the first cycle (Fig. 3b), which further confirms the highly resilient behaviour. Images in Fig. 3c (details of which can be found in Movie S1 in ESI†) show that, even after being stretched to about 15 times its original length, the hydrogel can still almost recover its original shape. The high resilience should be attributed to the unique pulley effect of the freely movable sliding crosslinks, which makes it different from physical and chemical gels.³³ Note that, although sliding gels have been widely studied previously, to our knowledge this may be the first cyclic tensile test showing such excellent resilient behaviour.

Fig. 3 Cyclic tensile test curves (a and b) and photo images (c) of the 1%–15CD sample. λ = l/l₀, l and l₀ are the stretched and original lengths of the sample, respectively.

Besides the excellent stretching property, the hydrogel also shows good compression behaviour. As shown in Fig. 4, the hydrogel can withstand 95% (sample 1%–15CD) and even 98% (sample 1%–5CD) compression strain without breaking.

Fig. 4 Compression test curves of samples 1%–15CD and 1%–5CD with a maximum compression strain of 95% (the upper limit of the machine, 50 N) and 98% (the minimum gap between the two plates of the machine), respectively, without breaking. σ_nominal = F/S₀, compression strain = (Δh/h₀) × 100%.

Moreover, a 5-run cyclic compression test with a maximum compression strain of λ′ = 0.1 ((Δh/h₀) × 100% = 90%) was also conducted. Although, unlike the cyclic tensile tests, hysteresis or energy loss is observed during the loading and unloading compressing process, all immediately subsequent cycles almost follow exactly the same procedure as for the first cycle (Fig. 5a). This observation reveals that a full recovery of the initial state is reached during the time scale of the experiment. Moreover, there is also no decrease in the maximum stress in the immediately following 4 cycles in comparison with the first cycle (Fig. 5b). Fig. 5c (details of which can be found in Movie S2 in ESI†) further demonstrates the highly ductile and resilient behaviour of the hydrogel in repeated compression deformation. Moreover, as shown in Fig. 5a and S5 in ESI,† the hydrogel also shows clear strain hardening behaviour in the compression process. This is in accordance with Ito's recent results observed in sliding gels.³²

Fig. 5 Cyclic compression test curves (a and b) and photo images (c) of the 1%–15CD sample. λ′ = h/h₀, h and h₀ are the compressed and original heights of the sample, respectively.

These highly resilient behaviours demonstrate that there should be almost no permanent damage to the hydrogel's network or dethreading of β-CD during either stretching or compression deformation, which indicates that the long PEG chains can act as end-capping blocks with similar functions to the bulky end groups in PR, because β-CD is more likely to locate on the PPG chains due to stronger hydrophobic interactions rather than on the PEG chains.^35,36,38 Therefore, the chain-extension effect may further enhance this function due to the formation of strong hydrogen-bonding interactions and the increasing chain length and entanglement of the axial polymer backbone.

Conclusions

In summary, hydrogels with obvious strain hardening and high resilience properties in both stretching and compressing processes were constructed by simple free-radical copolymerization of acrylamide and chain-extended vinyl-modified PPR, which acted as a multifunctional crosslinker. Instead of crosslinking of low-yielding PR with bulky end groups in organic solvents via a complicated process, here, with the chain-extension technique the whole procedure can be carried out in one pot in water. Obvious strain hardening behaviour in stretching deformation is observed for the first time, which is considered to be the orientation and aggregation of the movable crosslinkers along the axial polymer backbone. Moreover, the chain extension-induced formation of a second supramolecular network may also be responsible for this. The freely movable character of the crosslinkers endows the hydrogels with high resilience properties. These results not only present a new type of β-CD-based sliding-ring soft material, but also provide progress toward addressing the challenges of using synthetic hydrogels to achieve biological soft tissue/hydrogel's unique viscoelastic properties. The strain hardening and highly resilient characteristics combined with the simple modular synthetic strategy present appealing avenues for design, fundamental and application studies of new biomimetic materials and other new soft materials with unique biomimicking viscoelastic behaviours.

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (NSFC, no. 21304063 and 21274102) and the Priority Academic Program Development (PAPD) of Jiangsu Higher Education Institutions.

Notes and references

1. O. Wichterle and D. Lim, Nature, 1960, 185, 117–118.
2. J. Malda, J. Visser, F. P. Melchels, T. Juengst, W. E. Hennink, W. J. A. Dhert, J. Groll and D. W. Hutmacher, Adv. Mater., 2013, 25, 5011–5028.
3. B. V. Slaughter, S. S. Khurshid, O. Z. Fisher, A. Khademhosseini and N. A. Peppas, Adv. Mater., 2009, 21, 3307–3329.
4. J. Thiele, Y. Ma, S. M. C. Bruekers, S. Ma and W. T. S. Huck, Adv. Mater., 2014, 26, 125–148.
5. A. S. Hoffman, Hydrogels for Biomedical Applications, Ann. N. Y. Acad. Sci., 2001, 944, 62–73.
6. X. Yan, F. Wang, B. Zheng and F. Huang, Chem. Soc. Rev., 2012, 41, 6042–6065.
7. I. Tomatsu, K. Peng and A. Kros, Adv. Drug Delivery Rev., 2011, 63, 1257–1266.
8. J. Kopecek and J. Yang, Angew. Chem., Int. Ed., 2012, 51, 7396–7417.
9. D. Habault, H. Zhang and Y. Zhao, Chem. Soc. Rev., 2013, 42, 7244–7256.
10. M. Guo, L. M. Pitet, H. M. Wyss, M. Vos, P. Y. W. Dankers and E. W. Meijer, J. Am. Chem. Soc., 2014, 136, 6969–6977.
11. Y. Cui, M. Tan, A. Zhu and M. Guo, J. Mater. Chem. B, 2014, 2, 2978–2982.
12. Y. Okumura and K. Ito, Adv. Mater., 2001, 13, 485–487.
13. K. Haraguchi and T. Takehisa, Adv. Mater., 2002, 14, 1120–1124.
14. J. P. Gong, Y. Katsuyama, T. Kurokawa and Y. Osada, Adv. Mater., 2003, 15, 1155–1158.
15. T. Huang, H. Xu, K. Jiao, L. Zhu, H. R. Brown and H. Wang, Adv. Mater., 2007, 19, 1622–1626.
16. T. Sakai, T. Matsunaga, Y. Yamamoto, C. Ito, R. Yoshida, S. Suzuki, N. Sasaki, M. Shibayama and U.-i. Chung, Macromolecules, 2008, 41, 5379–5384.
17. C. Deng, Y. Cui, T. Zhao, M. Tan, H. Huang and M. Guo, RSC Adv., 2014, 4, 24095–24102.
18. O. Miyashita, J. N. Onuchic and P. G. Wolynes, Proc. Natl. Acad. Sci. U. S. A., 2003, 100, 12570–12575.
19. M. L. Gardel, J. H. Shin, F. C. MacKintosh, L. Mahadevan, P. Matsudaira and D. A. Weitz, Science, 2004, 304, 1301–1305.
20. C. Storm, J. J. Pastore, F. C. MacKintosh, T. C. Lubensky and P. A. Janmey, Nature, 2005, 435, 191–194.
21. J. Gosline, M. Lillie, E. Carrington, P. Guerette, C. Ortlepp and K. Savage, Philos. Trans. R. Soc., B, 2002, 357, 121–132.
22. L. Li, S. Teller, R. J. Clifton, X. Jia and K. L. Kiick, Biomacromolecules, 2011, 12, 2302–2310.
23. C. M. Elvin, A. G. Carr, M. G. Huson, J. M. Maxwell, R. D. Pearson, T. Vuocolo, N. E. Liyou, D. C. C. Wong, D. J. Merritt and N. E. Dixon, Nature, 2005, 437, 999–1002.
24. J. Cui, M. A. Lackey, A. E. Madkour, E. M. Saffer, D. M. Griffin, S. R. Bhatia, A. J. Crosby and G. N. Tew, Biomacromolecules, 2012, 13, 584–588.
25. M. Tan, T. T. Zhao, H. Huang and M. Y. Guo, Polym. Chem., 2013, 4, 5570–5576.
26. C. He, Z. Zheng, D. Zhao, J. Liu, J. Ouyang and H. Wang, Soft Matter, 2013, 9, 2837–2844.
27. T. Sakai, Y. Akagi, T. Matsunaga, M. Kurakazu, U.-i. Chung and M. Shibayama, Macromol. Rapid Commun., 2010, 31, 1954–1959.
28. T. Zhao, M. Tan, Y. Cui, C. Deng, H. Huang and M. Guo, Polym. Chem., 2014, 5, 4965–4973.
29. P. Chen and V. B. Shenoy, Soft Matter, 2011, 7, 355–358.
30. Y. C. Lin, G. H. Koenderink, F. C. MacKintosh and D. A. Weitz, Soft Matter, 2011, 7, 902–906.
31. T. Wang, D. Liu, C. Lian, S. Zheng, X. Liu and Z. Tong, Soft Matter, 2012, 8, 774–783.
32. K. Kato, T. Yasuda and K. Ito, Polymer, 2014, 55, 2614–2619.
33. Y. Noda, Y. Hayashi and K. Ito, J. Appl. Polym. Sci., 2014 DOI:10.1002/app.40509.
34. A. Harada and M. Kamachi, J. Chem. Soc., Chem. Commun., 1990, 1322–1323.
35. H. Fujita, T. Ooya and N. Yui, Macromolecules, 1999, 32, 2534–2541.
36. H. Fujita, T. Ooya and N. Yui, Polym. J., 1999, 31, 1099–1104.
37. M. Okada and A. Harada, Org. Lett., 2004, 6, 361–364.
38. J. Wang, P. Gao, L. Ye, A.-y. Zhang and Z.-g. Peng, J. Phys. Chem. B, 2010, 114, 5342–5349.
39. A. Bin Imran, T. Seki, T. Kataoka, M. Kidowaki, K. Ito and Y. Takeoka, Chem. Commun., 2008, 5227–5229.
40. K. Ito, Polym. J., 2007, 39, 489–499.

---

Footnote

† Electronic supplementary information (ESI) available: Synthesis, Fig. S1–5, Movies S1 and S2. See DOI: 10.1039/c4ra10928g

---