Synthesis and shape memory property of segmented poly(ester urethane) with poly(butylene 1,4-cyclohexanedicarboxylate) as the soft segment

Abstract

A series of thermoplastic poly(ester urethane)s (PUs) containing poly(butylene 1,4-cyclohexanedicarboxylate) (PBC) as a soft segment were prepared for shape memory materials via a two-step synthesis, with isophorone diisocyanate and 1,4-butanediol as its hard segments. The isomerization of the 1,4-cyclohexylene ring moiety (CHRM) happened during the preparation of PBC oligomers with molecular weight of 1700, 2300 and 3300 g mol⁻¹, but such phenomenon was not observed during the reaction for the synthesis of PU. The three PBC oligomers could all crystallize, but their crystallization ability was depressed after incorporation of PUs. Thermo, thermo-mechanical and mechanical properties of the corresponding PUs were studied to understand their structural information. It was found that the PU containing PBC1700 was an elastomer at elevated temperatures, and its shape memory properties were evaluated by DMA procedures. Regardless of its low degree of micro-phase separation, it was very interesting to find that the shape recovery ability of such PU was excellent, which was in contrast to traditional findings.

---

Introduction

Segmented polyurethanes are of importance due to their good mechanical properties, such as high tensile and tear strength, chemical and abrasion resistance, good processibility, and protective barrier properties. They are widely used in synthetic leather, fiber, adhesive and biomedical areas, and their various usages arise from the changeable segment chemistry, adjustable molecular weights, and the hard/soft segmental ratio.^1–5 The block structures of polyurethane chains, with alternating hard and soft segments, offers the control of unique morphology and properties. The hard segments are usually formed by the reaction of short-chain diols (chain extenders) with diisocyanates, whereas the soft segments are composed of the long-chain diols, such as polyester or polyether polyols.^6–10

Soft segments, which generally have glass transition temperatures in the range of −120 to −50 °C, provide low temperature flexibility, toughness and elasticity to the segmented polyurethane. The effect of the soft segments on morphology and properties of polyurethanes is prominent and has been studied extensively. The vast majority of polyurethanes containing poly(tetramethylene glycol) (PTMG) as the soft segment have good toughness and processability, are usually used as the thermoplastic elastomer.^11–15 Polycaprolactone (PCL) can also be used as the soft segment of polyurethane, which is biocompatible and biodegradable and can be used in medical field.^16–19 Furthermore, some non-polar soft segments such as polydimethylsiloxane (PDMS), polyisobutylene (PIB), poly(ethylene butylene) (PEB) and fluorinated polyether (FPE) oligomers can influence the surface properties of polyurethanes, which possess low surface energy or hydrophobic and water repellent surfaces.^20–25 With the design of soft segments, high performance polyurethane or polyurethane with specific application, such as shape memory property, could be prepared. In this work, we will focus on improving the mechanical properties of the polyurethane via introducing new kind of soft segment.

Polymers with aliphatic ring structures, such as poly(butylene 1,4-cyclohexanedicarboxylate) (PBC), have comparable rigidity to their aromatic counterparts but show much better toughness. Due to its unique structure, PBC exhibited excellent mechanical properties, like high modulus and good elasticity.^26–32 And it is reasonable to presume that the incorporation of PBC into polyurethane can help to prepare a new material with good properties. In this work, a new poly(ester urethane) containing PBC as the soft segment is prepared for the first time. The effects of the molecular weight of PBC on the structure and properties of polyurethane are also studied. By tuning the molecular weight of PBC glycols from high to low, the mechanical behavior of polyurethanes will evolve from stiff plastics to soft elastomers. Accordingly, an effective strategy to synthesize new polyurethane material with excellent shape memory property is established.

Results and discussions

Synthesis and compositional analysis of PBC and PUs

PBC and PUs were synthesized as presented in experimental part. However, due to the isomerization between cis- and trans-1,4-cyclohexylenering moiety (CHRMs) at high temperature, the final cis-CHRMs content in PBC need to be determined by ¹HNMR, which was displayed in Fig. 1. As shown, the peaks with chemical shifts at 2.48 (a_cis) and 2.29 ppm (a_trans), were signals from –CH– in cis- and trans-CHDA unit respectively. On the other hand, the peaks located at 4.09 (c) and 3.67 ppm (c_terminal) were assigned to the protons in –OCH₂– and terminal –OCH₂– which was connected with the hydroxyl group. At last, the composition and the molecular weight could also be calculated according to the following equation.

M_PBC = M × DP + X (1)

DP = 1 + I_c/I_cterminal (2)

w_cis = I_acis/(I_acis + I_atrans) (3)

where M is the molecular weight of monomer PBC. DP is the degree of polymerization determined by NMR. X is the molecular weight of the terminal butanediol. I_c, I_cterminal, I_acis and I_atrans are the integration of the chemical shift for c, c_terminal, a_cis and a_trans respectively. M_PBC and w_cis are the molecular weight of PBC and the molar ratio of the cis-CHRMS. The detail chemical structure of PBC glycol was summarized in Table 1, and the molar fraction of cis-CHRMs of PBC was 40%. Such result means that approximately 20% of cis-CHDA experienced isomerization and changed from cis to trans-CHDA. We synthesized three PBC glycols with different molecular weights for the subsequent preparation of polyurethane, and their molecular weight increased nonlinearly with the polycondensation time which was also shown in Table 1.

Fig. 1 ¹H NMR spectra (CDCl₃, 400 MHz) of PBC glycols.

Table 1 The composition and molecular weight of PBC glycols

Sample	Polycondensation time (min)	cis-CHRMs%		M (kg mol^−1)
		Feed	Final
PBC1700	120	50	40	1.7
PBC2300	180	50	40	2.3
PBC3300	240	50	39	3.3

---

Similarly, the true composition of PUs was also calculated according to the ¹HNMR spectrum, which was illustrated in Fig. 2. The two peaks with chemical shifts at 2.48 (a_cis) and 2.29 ppm (a_trans), corresponding to –CH– in cis- and trans-CHDA unit, were the characteristic peaks of soft segment. At the same time, the signal at 2.93 ppm was assigned to the two protons on the methylenes next to the amide group, which represented the characteristic peaks of hard segment. Furthermore, the peak at 4.11 ppm was correlated to the protons in –OCH₂– located at soft segment and chain extender. The detail chemical structures of PUs were summarized in Table 2. It showed that the ratio of cis-CHRMs in PU was around 40%, which was in accordance with the oligomer PBC, and indicated that no isomerization occurred during the synthesis of PUs from PBC and other chemicals. Specifically, the weight fraction of the soft segment content was controlled between 65–68% in PU chain and all of the PU samples had comparable molecular weights (M_n) and molecular weight distributions (M_w/M_n).

Fig. 2 ¹H NMR spectra (CDCl₃, 400 MHz) of PU.

Table 2 The composition and molecular weight of PUs

Sample	cis-CHRM%	PBC/IPDI/BDO (molar)	PBC (wt%)	Mn^a (kg mol^−1)	Mw/Mn^a
a The molecular weight and molecular weight distribution were measured by GPC.
PU1700	40	1/3/1.6	67	32	3.1
PU2300	40	1/3.5/2.3	68	34	3.2
PU3300	39	1/6.1/5	65	25	2.8

---

The chemical structure of PUs was confirmed by FTIR. The FTIR spectra (Fig. 3) displayed strong absorptions at 1670–1730 cm⁻¹, which were attributed to the stretching vibration of carbonyls. The characteristic bands of the urethane group at 3370 cm⁻¹ and 1540 cm⁻¹ were obtained, which corresponded to the N–H and the amide II band, respectively. They reflected the essential stretching vibration of carbonyl and amino groups in PU, indicating PU was successfully synthesized.

Fig. 3 The FTIR spectra of PUswith different PBC soft segments.

Hydrogen bonding in PUs

PU is a random copolymer composed of soft and hard domains by micro-phase separation. In the hard domains, the carbonyls and amino groups could form hydrogen bonds, which played an important role in the mechanical behavior and shape memory property. As described in Fig. 4, the FTIR spectra in the carbonyl stretching region from 1670–1760 cm⁻¹ could be divided into two spectral regions. The peak at 1700 cm⁻¹ was corresponded to the hydrogen bonding between the hard segments and that at 1730 cm⁻¹ was assigned to the stretching vibration of the free carbonyl group in the hard and soft segments. Consequently, the molar fraction of the hydrogen bonded carbonyl group (f) could be calculated by using the integrated intensity of the two individual vibrations by the following equation.where I₁₇₀₀ is the peak area at 1700 cm⁻¹ and I₁₇₃₀ is the peak area at 1730 cm⁻¹. Fig. 4(d) showed that the fraction of the hydrogen bonded carbonyl group were 33.3%, 30.6% and 28.5% for PU1700, PU2300 and PU3300 respectively. These percentages were significantly lower than that of most PUs with PTMG as the soft segment, which was usually more than 50%.³³ Like many previous research, we treat the fraction of the hydrogen bonded carbonyl group as the degree of the micro-phase separation of PU in some extent.^2,33 Consequently, the low hydrogen bonding meant the improved compatibility between hard and soft domains. In this work, it was probably because of the cyclohexane ring structure existing in both phases. Besides, the fraction of the hydrogen bonded carbonyl group decreased a little with the increase of PBC molecular weight, which implied better solubility between hard and soft phases.

Fig. 4 (a–c) The absorbance FTIR spectra in the C=O stretching region of different PU samples: (a) PU1700; (b) PU2300; (c) PU3300. (d) The fraction of the hydrogen bonded carbonyl group of PUs.

Thermal and thermo-mechanical property of PUs

The DSC analysis was conducted to characterize the thermal behaviors of PBC glycols and corresponding PUs, and the results were presented in Fig. 5. It showed that all PBC oligomers could crystallize and the melting peaks of PBC2300 and PBC3300 were around 83 and 92 °C, respectively. Nevertheless, their melting regions were broad, especially for PU1700, indicating that the crystallization was weak and incomplete. With the molecular weight increased, the melting temperature and enthalpy were both enhanced, meaning that the crystallinity of PBC increased as the molecular weight increased. Another thermal transition was observed at about −15 °C, which was attributed to the glass transition of PBC glycols. For PU1700, PU2300 and PU3300 samples, all of them had a thermal transition at around 40 °C, which, according to the previous works, was the glass transition temperature of the IPDI hard segment.³³ Interestingly, the glass transition of the soft segment PBC could not be observed by DSC anymore, which is probably depressed by the partially mixed phases. DSC results also demonstrated that the random block PU copolymer had well mixed hard and soft segments, just as analyzed by FTIR. Exceptionally, PU3300 had a melting peak at 95 °C, which could be correspond to the fusing of PBC crystals. However, the melting enthalpy of PU3300 was lower than that of PBC3300, indicating that the hard segment IPDI could disrupt the formation of PBC crystals. Again, it could be confirmed that the crystallization ability of PBC increased with its molecular weight.

Fig. 5 The DSC second heating scans of (a) PBC oligomers and (b) PUs at 10 °C min⁻¹.

In order to further investigate the crystallization of PBC and PU, XRD experiment was carried out and the data was shown in Fig. 6. It indicated that PBC3300 was significantly crystallized. The strongest diffraction index was at 2θ = 15.6°, 18.7°, 21.1°, 23.0° and 29.1°, (d = 0.56, 0.47, 0.41, 0.37 and 0.29 nm respectively), which was in accordance with Gigli's work.³² It is reasonable to find that the intensity of diffraction peaks decreased with the decrease of the molecular weight of PBC. As to PBC1700, the diffraction peaks at 2θ = 21.1° and 29.1° were not found, implied the incompletion of crystals. While for PUs, the PU1700 had a wide and smooth peak from 15 to 30°, suggesting an amorphous state without ordered crystal structure. As the molecular weight of PBC increased, the diffraction peak of PU became sharper and their shoulder peaks at 23° appeared, demonstrated that PUs with longer segment, such as PU2300 and PU3300, could crystallize. Especially for PU3300, the diffraction peaks at 2θ = 15.6°, 18.7°, 21.1°, 23.0° and 29.1° could all be observed and the intensity of the diffraction peaks was much stronger than that of PU2300 and PU1700. It can be summarized that the crystal in PU3300 grew more complete than that in PU2300. Again, we confirmed that both the crystallinity and perfection of crystals increased in PUs as the molecular weight of PBC increased.

Fig. 6 The XRD patterns of (a) PBC oligomers and (b) PUs.

Fig. 7 displayed DMA results of the storage moduli and mechanical tan δ as a function of temperature with the test frequency of 1 Hz. In the temperature range from −110 to 100 °C, two relaxation regions of tan δ at approximately −70/−40 °C and 0/60 °C were observed, which were denoted as β and α relaxation, respectively. The β peak was generally considered as the chair–boat–chair conformational transition of 1,4-cyclohexylene ring.^34,35 On the other hand, the α peak was assigned to the glass transition of PU, which was in accordance with the T_g value determined by DSC. It need to be emphasized that the DMA analysis, exhibited only one glass transition too and assured the relatively good compatibility between two segments, as inferred from DSC and FTIR experiments. Interestingly, both the glass transition temperature and tan δ of PU1700 were highest among the three sample, and it could be correlated with the highest degree of phase separation of PU1700 among the three samples. The hard domains would experience ‘melt’ process in the glass transition region, the motion inside the hard domains promote the value of tan δ. In addition, the storage modulus (E′) and loss modulus (E′′) of PU showed a sharp decrease of about three orders of magnitude with temperature above the glass transition, because the PU material changed from glass state to rubber state. On the contrary, E′ and E′′ decreased a little at β transition. What's more, E′ and E′′ of PU3300 were higher than the other two samples, suggesting the high soft molecular weight could lead to the higher thermal-mechanical values.

Fig. 7 (a) DMA storage modulus and (b) loss modulus vs. temperature at 1 Hz, (c) the corresponding tan δ plot of PUs. The heating rate was 2 °C min⁻¹.

Mechanical properties of the PUs

Mechanical properties of polymers are important with respect to the practical applications. Fig. 8 showed the representative tensile strain–stress curves of PUs. It was noted that curves of PU2300 and PU3300 were consisted with four regions: (1) linear and non-linear viscoelasticity, (2) neck region, (3) plastic flow and (4) strain hardening. Obviously, they fit typical semi-crystalline thermoplastic polymer features. Yielding and neck-forming phenomena were attributed to the stretching and re-orientation of the PBC crystals. Although the necking process was not obvious for PU1700 in single strain–stress curve, the cyclic tensile experiment showed that the strain induced crystallization happened in PU 1700. Since the molecular weights of all the PU samples were similar and the degree of micro-phase separation was close, the crystallinity of PU became the most prominent factor that led to the difference in mechanical properties.

Fig. 8 (a) The representative tensile strain–stress curves of PUs. (b) The tensile strain–stress curves of PUs when the strain was smaller than 75%.

Young's modulus (E), tensile strength (σ_t), elongation at break (ε_b), stress at yield (σ_y) as well as strain at yield (ε_y), averaged from more than 5 specimens for each PU, were summarized in Table 3. The Young's moduli were higher than 75 MPa for these PUs. With the increase of the PBC molecular weight, the Young's modulus and tensile strength were all increased, but the elongation at break of all the samples kept around 500%. The crystallized structure of soft segment had great effect on E and σ_t, and the amorphous structure could account for ε_b. In our work, the crystallinity of PU was low, which resulted in the similar elongation at break. From the Table 3, we could conclude that the PUs simultaneously possessed excellent tensile properties with high Young's modulus and large elongation at break.

Table 3 The tensile properties of PU samples

Samples	E (MPa)	σt (MPa)	εb (%)	σy (MPa)	εy (%)
PU1700	75 ± 4	15 ± 0.5	500 ± 12	None	None
PU2300	180 ± 11	20 ± 1.4	510 ± 18	8 ± 0.6	13 ± 1
PU3300	210 ± 15	42 ± 1.2	520 ± 20	17 ± 1	12 ± 1

---

Shape memory property of the PU1700

According to the tensile properties of PU1700, it showed satisfied mechanical property as an elastomer, and its shape memory property was evaluated by DMA experiment through widely adopt procedure, as shown in Fig. 9(a). PU1700 was firstly heated to the stretching temperature (T_s1) for 5 min under a low tensile stress of 0.01 N. A strain-controlled uniaxial stretching was then applied with engineering strain of 100%. The strain was held constantly during the subsequent quenching to T_s2 and then the stress was removed. A free strain recovery was immediately performed under continuous heating condition of 3 °C min⁻¹, and the evolution of the strain was recorded during the recovery. At the last step, the sample was kept on strain recovery process for 10 min after the temperature reached at T_s1. The stretching and recovering temperature selected for shape memory testing (programming and release) was 65 °C, a temperature at which the material showed elasticity of the amorphous soft segment. And the fixing temperature was −25 °C, a temperature at which PU was in glass state. Recalling that the glass transition temperature of PU was 35 °C, it could be concluded that the hard domain was responsible for shape recovery and amorphous soft segments for shape fixity.

Fig. 9 (a) Schematic description of a triple-shape memory test at continuous heating condition, and (b) evolution of stress, strain, and temperature during shape memory experiment for PU1700.

Fig. 9(b) presented the evolution of stress, strain, and temperature during shape memory process of PU1700. The shape memory property could be evaluated by the recovery ratio (R_r) and the fixity ratio (R_f). The former parameter represented the ability of recovering to the original shape, and the latter one stood for the ability of storing the temporary shape. The two parameters could be calculated by the following equations, respectively.

with ε_s2, the maximum strain achieved during stretching, ε₂, the strain of the temporary shape after unloading, and ε_r1, the strain of the recovered original shape. For PU1700, R_f values were more than 95% and were nearly constant in the three shape memory test cycles, suggesting the glass transition temperature of the soft segments had no dependence on thermo-mechanical history, and was important for retaining its functionality as the switching segment for three experimental cycles.^36–38 The recovery ratio at the first cycle was 90%, which was very close to that of the polyurethanes with high degree of micro-phase separation.^11,12 In contrary, our experiment proved that good compatibility between hard and soft phases in this type of PU did not deteriorate the shape memory property. However, R_r decreased gradually to 69% at the third cycle, which could be attributed to the irreversible deformation due to partial modifications of hard domains by chain slippage and disentanglement during stretching.^39–42 However, for PU2300 and PU3300, the shape memory test could not be completed by DMA due to the loss of the shapes at 65 °C. For these semi-crystallized PU materials, if the stretching temperature was lower than the melting point but higher than the glass transition, the crystals would be destroyed and its shape could not be restored after the stretching was removed. If the stretching temperature was higher than the melting point, chains would flow along the stress direction and the material could not even maintain its shape anymore. Judging from these experimental facts, it was crucial that the molecular weight of PBC was controlled properly to obtain a satisfied shape memory property of PU. Further work will be done for attaining a better shape memory PU.

Conclusions

In this work, three PBC oligomers, with molecular weight of 1700, 2300 and 3300 g mol⁻¹ respectively, have been prepared, and their detailed structures were characterized by NMR and GPC. It has been found that 20% of cis-CHDA experienced isomerization from cis to trans during the polymerization process, and the final cis to trans ratio in PBC was 4 : 6. DSC results showed that all PBC oligomers could crystallize. After that, a new polyurethane copolymer with PBC as the soft segment was synthesized and the structure–property of PU was also studied.

The molecular weight of PBC had decisive effect on the structure and mechanical properties of the PUs. When its molecular weight was lower than 2300, PU was amorphous and behaved as an elastomer at static conditions. If its molecular weight was higher than 2300, they could crystallize and exhibited as a plastic with higher modulus. Nevertheless, the crystallization ability of PBC is depressed after polymerization with isophorone diisocyanate. Consequently, the crystallization of PBC segment affected the shape memory behavior of the corresponding PUs. PU with PBC1700 had good shape memory property with the recovery ratio of 90% and fixity ratio of 95%, while PU2300 and PU3300 had no shape memory property at all. It is interesting to notice that the good compatibility between PBC and IPDI segments, as implied by FTIR, DSC and DMA results, did not deteriorate the shape memory property, which is contradictory to traditional conclusion. The present work demonstrates that designing the structure of the soft segment is an effective method to improve the mechanical and shape memory property of polyurethane, which is very informational for the design of even better shape memory PUs.

Experimental

Materials

1,4-Butanediol (BDO), titanium(IV) butoxide, isophorone diisocyanate (IPDI), dibutylamine and dibutyltindilaurate (DBTL) were obtained from the Aladdin Reagents (Shanghai) Co., Ltd. Chloroform, methanol, acetone and hydrogen chloride were purchased from Sinopharm Chemical Reagent Co., Ltd. 1,4-Cyclohexanedicarboxylic acid (CHDA) with a cis/trans ratio of 50% was obtained from Nanjing Chemlin Chemical Industry Co., Ltd. All chemicals were used as received.

Synthesis of PBC glycol

The PBC glycol was synthesized as shown in Scheme 1. CHDA (0.2 mol), BDO (0.3 mol) together with titanium(IV) butoxide (0.1 mmol) as the esterification catalyst were firstly mixed in a three flask and reacted at 130 °C for 60 min, 145 °C for 60 min and 160 °C for 180 min in sequence under a dry nitrogen atmosphere. After that, titanium(IV) butoxide (0.1 mmol) as the polycondensation catalyst was added and the reaction was processed under vacuum at 300 Pa for 30 min. The system was then heated to 180 °C and evacuated below 100 Pa for at least 120 min. The molecular weight of PBC was determined by the reaction time of this procedure, which were 120, 180 and 240 min in this work. The PBC product was dissolved in CHCl₃ and precipitated in methanol. At last, the PBC sample was dried under vacuum at 40 °C for 24 h. The PBC glycol was labeled as PBCx, where x represented the molecular weight of PBC.

Scheme 1 Synthesis of PBC glycol from CHDA and BDO.

Synthesis of poly(ester)urethane

Scheme 2 illustrated the synthetic pathway of the poly(ester)urethanes (PU), which was a two-step polymerization process. The feeding ratio of the PU was designed by the following rule: the weight percentage of the soft segment PBC was 65% (wt%). Firstly, the IPDI and PBC together with DBTL as the catalyst were dissolved in CHCl₃ and reacted at 45 °C for 180 min under a dry nitrogen atmosphere. The residual isocyanate (NCO) content was determined by titration according to DIN EN ISO 14896-2009 (ESI S1†). Then an appropriate amount of BDO as the chain extender was added and the system was heated to 55 °C for 180 min. Finally the mixture was concentrated by rotary evaporation at room temperature and purified by dissolving–precipitating in a chloroform and methanol solution (20 : 80 vol%). The product PU was labeled as PUx, where x represented the molecular weight of PBC.

Scheme 2 Two-step synthesis of PU synthesis and monomer structures.

Characterization

The structure of PBC and PU were verified by proton nuclear magnetic resonance (¹HNMR) in CDCl₃ solvent using a BrukerAVIII400 NMR spectrometer at room temperature. Besides, the Fourier transform infrared (FTIR) characterization was performed on a Thermo Nicolet 6700 Fourier transform infrared spectrometer from Thermo-Fisher Scientific, scanning from 500 to 4000 cm⁻¹ and 32 scans were collected for each sample to confirm their structures. And the molecular weight of PU was measured using a HLC8320 GPC with CHCl₃ as the solvent.

The thermal behavior of PBC and PU were determined by a differential scanning calorimeter (DSC Q2000, TA Instruments) with the following procedure: heating from 25 °C to 150 °C at 10 °C min⁻¹, cooling to −50 °C at 10 °C min⁻¹, and then heating to 150 °C at 10 °C min⁻¹. Notably, the first heating scan was used to erase the thermal history of the samples. The dynamic mechanical behavior and shape memory property were examined by using a dynamic mechanical analysis (DMA Q800, TA Instruments). The samples were cut into rectangular specimens of 15 mm × 4 mm × 0.5 mm. The tests were performed in tension mode from −100 to 100 °C at 3 °C min⁻¹ with the frequency of 1 Hz. The tensile testing was carried out using an Instron5567 tensile testing machine with a 500 N load cell at 25 °C, while the sample size was 35 mm × 2 mm × 0.5 mm.

Acknowledgements

The authors thank the generous financial support by Zhejiang Provincial Natural Science Foundation of China (Grant No. LY15B040006), Open-end Funds of Beijing National Laboratory for Molecular Science (20140147, 20150115) and the Open Fund of Zhejiang Provincial Top Key Discipline of Aquaculture (Grant No. XKZSC05).

Notes and references

1. G. Oectel, Polyurethane Handbook, Hanser Publishers, 1985.
2. L. L. Harrell, Macromolecules, 1969, 2, 607–612.
3. N. S. Schneider, C. S. P. Sung, R. W. Matton and J. L. Illinger, Macromolecules, 1975, 8, 62–67.
4. C. B. Wang and S. L. Cooper, Macromolecules, 1983, 16, 775–786.
5. T. G. Mackay, D. J. Wheatley, G. M. Bernacca, A. C. Fisher and C. S. Hindle, Biomaterials, 1996, 17, 1857–1863.
6. Y. Li, Z. Ren, M. Zhao, H. Yang and B. Chu, Macromolecules, 1993, 26, 612–622.
7. Y. Li, W. Kang, J. O. Stoffer and B. Chu, Macromolecules, 1994, 27, 612–614.
8. C. Li, S. L. Goodman, R. M. Albrecht and S. L. Cooper, Macromolecules, 1988, 21, 2367–2375.
9. S. Leng, X. Lan, Y. J. Liu and S. Y. Du, Prog. Mater. Sci., 2011, 56, 1077–1135.
10. M. Ahmad, J. K. Luo, H. Purnawali, W. M. Huang, P. J. King, P. R. Chalker, M. Mireftab and J. Geng, J. Mater. Chem., 2012, 22, 8192–8194.
11. Y. C. Chung, D. K. Nguyen, J. W. Choi and B. C. Chun, J. Appl. Polym. Sci., 2011, 120, 2063–2073.
12. H. H. Wang and U. E. Yuen, J. Appl. Polym. Sci., 2006, 102, 607–615.
13. Z. M. Chang, M. Q. Zhang, A. G. Hudson, E. B. Orler, R. B. Moore, G. L. Wilkes and S. R. Turner, Polymer, 2013, 54, 6910–6917.
14. J. R. Lin and L. W. Chen, J. Appl. Polym. Sci., 1998, 69, 1575–1586.
15. D. B. Klinedinst, E. Yilgor, I. Yilgor, F. L. Beyer, J. P. Sheth and G. L. Wilkes, Rubber Chem. Technol., 2005, 78, 737–753.
16. M. A. Woodruff and D. W. Hutmacher, Prog. Polym. Sci., 2010, 35, 1217–1256.
17. M. Barikani and C. Hepburn, Cell. Polym., 1987, 6, 29–36.
18. M. Momtaz, M. R. Nouri and M. Barikani, J. Mater. Sci., 2014, 49, 7575–7584.
19. L. Xue, S. Y. Dai and Z. Li, Macromolecules, 2009, 42, 964–972.
20. I. Yilgor and J. E. McGrath, Adv. Polym. Sci., 1988, 86, 1–86.
21. S. K. Jewrajka, E. Yilgor, I. Yilgor and J. P. Kennedy, J. Polym. Sci., Part A: Polym. Chem., 2009, 47, 38–48.
22. D. B. Klinedinst, E. Yilgor, I. Yilgor, F. L. Beyer and G. L. Wilkes, Polymer, 2005, 46, 10191–10201.
23. T. Choi, K. A. Masser, E. Moore, J. Weksler, A. Padsalgikar and J. Runt, J. Polym. Sci., Part A: Polym. Chem., 2011, 49, 865–872.
24. U. Ojha, P. Kulkarni and R. Faust, Polymer, 2009, 50, 3448–3457.
25. T. Choi, J. Weksler, A. Padsalgikar and J. Runt, Polymer, 2009, 50, 2320–2327.
26. D. J. Brunelle and T. Jang, Polymer, 2006, 47, 4094.
27. C. Berti, A. Celli, P. Marchese, E. Marianucci, G. Barbiroli and F. D. Credico, Macromol. Chem. Phys., 2008, 209, 1333–1344.
28. F. Liu, J. Zhang, J. Wang, X. Liu, R. Zhang, G. Hu, H. Na and J. Zhu, J. Mater. Chem. A, 2015, 3, 13637–13641.
29. F. Liu, J. Zhang, J. Wang, H. Na and J. Zhu, RSC Adv., 2015, 5, 94091–94098.
30. M. Fabbri, M. Soccio, M. Gigli, G. Guidotti, R. Gamberini, M. Gazzano, V. Siracusa, B. Rimini and N. Lotti, Polymer, 2016, 83, 154–161.
31. L. Genovese, M. Soccio, M. Gigli, N. Lotti, M. Gazzano, V. Siracusa and A. Munari, RSC Adv., 2016, 6, 55331–55342.
32. M. Gigli, N. Lotti, V. Siracusa, M. Gazzano, A. Munari and M. D. Rosa, Eur. Polym. J., 2016, 78, 314–325.
33. L. S. Zhang, S. S. Shams, Y. P. Wei, X. Q. Liu, S. Q. Ma, R. Y. Zhang and J. Zhu, J. Mater. Chem. A, 2014, 2, 20010–20016.
34. L. P. Chen, A. F. Yee, J. M. Goetz and J. Schaefer, Macromolecules, 1998, 31, 5371–5382.
35. J. Liu and A. F. Yee, Macromolecules, 1998, 31, 7865–7870.
36. M. L. Auad, V. S. Contos, S. Nutt, M. I. Aranguren and N. E. Marcovich, Polym. Int., 2008, 57, 651–659.
37. A. Saralegi, L. Rueda, B. Fernandez-d'Arlas, I. Mondragon, A. Eceiza and M. A. Corcuera, Polym. Int., 2013, 62, 106–115.
38. A. Saralegi, S. C. M. Fernandes, A. Alonso-Varona, T. Palomares, E. J. Foster, C. Weder, A. Eceiza and M. A. Corcuera, Biomacromolecules, 2013, 14, 4475–4482.
39. D. Ratna and J. Karger-Kocsis, J. Mater. Sci., 2008, 43, 254–269.
40. Q. Zhang and Q. S. Yang, J. Appl. Polym. Sci., 2012, 123, 1502–1508.
41. M. M. Huang, X. Dong, L. L. Wang, Y. Y. Gao and D. J. Wang, RSC Adv., 2015, 5, 50628–50637.
42. M. M. Huang, X. Dong, Y. Y. Gao, Q. Xing, W. L. Li and D. J. Wang, Polymer, 2014, 55, 4289–4298.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra16325d

---