Calcium sulfate hemihydrate whisker reinforced polyvinyl alcohol with improved shape memory effect

Abstract

A new shape memory polymer is synthesized by introducing calcium sulfate hemihydrate whiskers (HHW) to reinforce polyvinyl alcohol (PVA). The tensile strength of the composite is increased by 57%, and the storage modulus reaches 22.24 GPa.

---

Shape-memory polymers (SMPS) which can undergo a large range recoverable deformation on exposure to external stimuli such as solvent, heat, electricity, light, and magnetic field,^1–7 have been extensively studied since 1980s. Compared with shape memory ceramics and shape memory alloys (SMAs), SMPs have a great many advantages, for instance, low density, high shape deformability, low cost, high shape recoverability, good processability, and tunable transition temperatures. Therefore, SMPs have broad potential applications in intelligent biomedical devices,^8–10 smart textiles,¹¹ sensors and actuators.^12,13 However, the mechanical properties of SMPs are much weaker than those of SMAs which severely limits their practical applications.

To overcome this disadvantage, SMP composites have been developed. More specifically, there is a growing concern to integrate SMPs with inorganic materials to develop new SMPs with improved mechanical properties for practical applications. For instance, Tan et al. reported that a graphene-based polyurethane block copolymer exhibited higher modulus and breaking stress than pristine polyurethene.¹⁴ Sahoo prepared multi-walled carbon nanotubes (MWNTs)/polyurethane block copolymer (PU). The addition of 2.5% MWNTs improved the mechanical properties of MWNT/PU composite significantly where the modulus and tensile strength increased 200% and 37%, respectively.¹⁵ Amirian used MWCNTs enhanced poly(L-lactide-co-ε-caprolactone) (PLACL), and the SMPs composite exhibited higher tensile strength and better shape memory effect.¹⁶ Kim incorporated silica particles into amorphous polyurethanes which significantly improved the mechanical properties of the SMPs composite.¹⁷ Fu investigated different inorganic agents (alumina, silica, and clay) reinforced cross-linked polystyrene (PS) copolymer. These SMPs composites exhibited higher mechanical properties and shape memory properties. Furthermore, the rod-shaped clay exhibited better reinforcement than spherical particles, due to their property to enhance in multiple directions and high aspect ratio.¹⁸

HHW with high aspect ratio is single-crystal fibers, and well known for its prominent properties, such as chemical resistance, excellent thermal stability, fine compatibility with polymer matrix, low cost, almost perfect structure, and high mechanical properties (Young's modulus = 178 GPa and tensile strength = 20.5 GPa).¹⁹ Therefore, HHW is widely used in reinforcing polymer, rubber, ceramic and so on.²⁰ Moreover, calcium sulfate hemihydrate (HH) is an excellent biocompatible material with the longest clinical history which is more than one hundred years. Based on this, we decide to reinforce SMPs by introducing HHW with high aspect ratio in order to improve the mechanical properties and shape memory effect of SMPs.

According to our preliminary work, hydroxyl groups may form hydrogen bonds with HHW.²¹ Therefore, we decided to combine PVA with HHW to develop new SMPs composite in order to improve the mechanical properties and shape memory effect of PVA-based SMPs. PVA-based SMPs have drawn more and more attention in various fields,²² because of their excellent properties, such as biocompatibility, nontoxic nature, high hydrophilicity, and good mechanical properties. Furthermore, PVA has plenty of hydroxyl groups in polymer chains which is easier to be modified and cross-linked. However, until now, only a few works focus on PVA-based SMPs. For instance, Hirai crosslinked PVA gel prepared by cyclic freezing/thawing method with glutaraldehyde.^23,24 Du prepared SMPs based on PVA chemically cross-linked with glutaraldehyde.^25,26 Zhou prepared acidic multi-wall carbon nanotubes (AMWNTs)/PVA composites by solution casting method.²⁷ Fu prepared graphene oxide (GO)/PVA composites through solution casting method.²⁸ Bai prepared Al₃O₂/PVA composites using cyclic freezing/thawing method. However, HH whisker/PVA composites have not been reported.²⁹

As stated previously, shape recovery of SMPs can be actuated by various strategies. Among these strategies, thermo-induced SMPs have extensively studied which have good memory effects, controllable thermal transition temperature, and simple transition process. However, in terms of the biomedical applications, it would be better to prepare SMPs with water-induced shape memory effects since it does not damage the tissue and has no interactive response to the blood while implanted into human body. Furthermore, water-induced shape recovery is considered much easier and more applicable. Since Huang and his co-workers successfully prepared water-induced polyurethane SMP, the water-induced SMPs were extensively studied.³⁰ However, these works were almost focus on water-induced shape memory effect around ambient temperature. The shape memory effect driven by 37 °C water has rarely been investigated in previous works.

Herein, we described a novel water responsive shape memory PVA/HHW composite by solvent casting method. HH is a metastable compound which is easy to hydrate and form dihydrate calcium sulphate (DH).³¹ Thus, HHW is usually transition to anhydrous dead roasted calcium sulfate whisker (stable phase) through high-temperature calcination before used as reinforcing agent. However, this pretreatment would increase the cost, moreover, lower the aspect ratio of the whisker. Thus, two problems are interesting of in this study. One is whether HHW without pretreatment will maintain its metastable structure during the synthesis process and improve the mechanical properties of PVA/HHW composite. The other is whether the adding of HHW will improve the water-induced shape memory effect of PVA. To illustrate the two problems, PVA/HHW composite was prepared by incorporating HHW into PVA matrix using 98 °C water as the processing medium, and then the recovery process of the PVA/HHW composite was recorded and the corresponding mechanism was systematically studied. We summarize two highlights of the work. First, we introduce HHW as additional physical cross-linked points in PVA matrix to improve the mechanical properties and shape memory effect of PVA without any pre-treatment. Second, the shape memory effect is induced by 37 °C water, which will help to prevent the unexpected effects caused by external heating. We expect that this work will provide a framework for developing new PVA based shape memory polymers (SMPs) applied in biomedicine.

The detailed preparation procedures for the SMPs composite are shown in experiment section. Briefly, a certain amount of HHW aqueous dispersion gradually poured into PVA solution at 98 °C, and stirred for 2 h. Then, the homogeneous solution was poured into a Teflon plate and kept for 4 h at 90 °C to cast the film. As shown in Fig. S1,† HHW are well-dispersed in PVA matrix. Fig. 1 demonstrates the synthesis process and supposed formation mechanism of PVA/HHW composite. In the fabrication process of the PVA/HHW films, the cross-linking in the composite is probably ascribed to crystallites formed in PVA matrix and hydrogen bonds between oxygen contained in SO₄²⁻ which with dense distribution on the side facets of HHW and hydroxyl groups in PVA chains which both acted as the additional physically cross-linking points. The additional physically cross-linking points formed a dense three dimensional network which improved the mechanical properties of the composite. Moreover, the rod-like HHW may bridge many PVA chains and afford effective load transfer which would also improve the mechanical properties of the PVA/HHW composite.

Fig. 1 The preparation procedures of PVA/HHW composite and the formation mechanism is demonstrated.

To investigate the possible interaction between both components, and confirm the whiskers present in PVA matrix are HHW rather than DHW. X-ray diffraction (XRD) was carried out. The XRD patterns of HHW (curve a), pure PVA (curve b), PVA/HHW with 5 wt% loading (curve c), and PVA/HHW with 7 wt% loading (curve d) shown in Fig. 2. The typical diffraction peaks of HH were observed in both curve c and d indicated that the filler was HHW. The diffraction peak of PVA was occurrence in all cases, but the diffraction peak intensity was decreased with the HHW loading increasing, namely, the crystallinity of PVA matrix was decline as the HHW loading increasing. The increase in the amount of HHW with decline in crystallinity indicated that the strong interfacial interactions between the two components and homogeneous dispersion of HHW in the polymer matrix. Similar findings have been reported in other works.^28,32

Fig. 2 XRD patterns (a) HHW, (b) pure PVA, (c) PVA/HHW with 5 wt% HHW loading, (d) PVA/HHW with 7 wt% HHW loading.

Because of interfacial interactions would affect the mobility of polymer chains. Therefore, differential scanning calorimetry (DSC) was used to further investigate the interfacial interactions between them were due to hydrogen-bond interactions. The glass transition temperature (T_g) of pure PVA and PVA/HHW composite with different HHW loading shown in Fig. 3. The T_g of PVA/HHW composite with 5 wt% and 7 wt% loading increased from 75.4 to 76.9 and 77.7 °C. The increase in T_g indicates that the PVA chains were truly restricted by hydrogen bond interactions, and similar effect also reported in other works.^33,34 As shown in Fig. S2,† the thermal stability of PVA/HHW with different HHW loading are also investigated by TGA. The increase in thermal stability may be also ascribed to interactions between PVA and HHW.

Fig. 3 DSC curves of pure PVA and PVA/HHW composite.

To further confirm the hydrogen bonds between PVA and HHW, the FTIR was carried out. As shown in Fig. 4, curve a shows a representative strong and large adsorption band at 3000–3600 cm⁻¹ locating at 3410 cm⁻¹, causing by the stretching vibration of –OH because of extensively intermolecular/intramolecular hydrogen bond interactions for that of free –OH of secondary alcohols is usual occurrence at 3620 cm⁻¹.³⁵ The peak located at 2943 and 2897 cm⁻¹ can be ascribed to C–H stretching vibration, and the deformation vibration of CH₂ group occurred at 1431 and 1326 cm⁻¹. The symmetric C–C stretching vibrations occurring at 1150 cm⁻¹ suggests that the existence of crystalline region in PVA matrix.³⁶ Compared with curve a, the stretching vibration of hydroxyl group at curve b and c shift from 3410 cm⁻¹ to 3332 and 3327 cm⁻¹ respectively. This further confirms the existence of hydrogen bonding interactions between PVA matrix and HHW.

Fig. 4 FTIR spectras. (a) Pure PVA, (b) PVA/HHW film with 5 wt% loading of HHW, (c) PVA/HHW film with 7 wt% loading of HHW, (d) HHW.

As a result of homogeneous dispersion of HHW and the additional physical cross-linking points (crystallite region and the hydrogen bond interactions), the PVA/HHW composite exhibits fine mechanical properties. The typical stress–strain curves for the SMPs composite with different HHW loading are shown in Fig. 5. The mechanical behavior is obviously influenced by the loading weight of HHW. For instance, with 5 wt% HHW loading, the breaking strength is increased by 57% from 45.8 to 71.9 MPa, but the breaking elongation of the composite is shorter than pure PVA. As the HHW content increasing to 7 wt%, the breaking strength of 67.9 MPa is obtained, and the breaking elongation of the composite is lower than 5 wt% loading. Similar findings have been reported in other work.¹⁸ Moreover, Balazs and his co-workers have investigated the supramolecular networks formed from nanoscale rods in binary mixture, and indicated that there was a transition density, ρ*, in which the rod-like fillers were needed to availably form a percolating network with lamellar morphology. If the number density of the rod-like filler is significantly beyond ρ*, this percolating network will be destroyed, which would reduce the mechanical behavior of the composite.³⁷ This may explain the mechanical properties with 5 wt% HHW loading is better than 7 wt% loading.

Fig. 5 The typical stress–strain curves for PVA/HHW composite.

The water-induced shape memory behaviours of pure PVA and PVA/HHW composites shown in Fig. 6. The straight samples were deformed to “U” like shape at 90 °C and kept this shape under constant stress during cooling bank to room temperature. After that, the deformed specimens were immersed in 37 °C water and the experimental phenomena were recorded by digital camera during the recovery process. According to the recovery process, pure PVA specimen could not fully recover its original shape, while PVA/HHW with 5 wt% loading regain its original shape within 80 s after immersed in 37 °C water. As the HHW loading up to 7 wt%, the recovery time decreased to 70 s. The result clearly suggests that the addition of HHW improves the shape memory effect of PVA. This can be ascribed to the enhanced physically cross-linking by introducing HHW. It has been proved previous that hydrogen bonds were formed between HHW and PVA. In addition, the rod like whisk would cause firm twisting with the PVA chains. Both of these will enhanced the hard domains, by decreasing chain slippage during deformation.²⁸ Therefore, HHW can act as additional physically cross-linking points, which improves the shape memory effect of PVA. Furthermore, the shape memory effect of the composite was evaluated in a quantitative way using bending test which described by Lendlein.³⁸ The shape recovery ratios of pure PVA and PVA/5 wt% HHW in 37 °C water were shown in Fig. 7. The pure PVA sample cannot recover its original shape while the PVA with 5 wt% HHW loading recovers its original shape quickly and the recovery ratio is up to 96%.

Fig. 6 The apparent shape recovery of PVA and PVA/HHW films (fix to “U” like shape) performed at 37 °C water.

Fig. 7 Shape recovery ratio of pure PVA and PVA/5 wt% HHW in 37 °C water.

It is obvious that the weight of the composites increase sharply after being immersed in water (Fig. S3†). Huang and his co-workers have investigated the mechanism of water-induced shape memory effect of polyurethane. They found that the hydrogen bonding played a key role in the process.³⁰ The water molecular, entered into the polymer matrix, would weakened the hydrogen bonding between C=O and N–H groups, which ascribed to the sharply reducing of the glass transition temperature.

To confirm the glass transition temperature decreasing after immersion in water, the dynamic mechanical analysis of composite with 5 wt% HHW loading after different immersion times were carried out. Fig. 8a shows tan δ of PVA/HHW with 5 wt% HHW loading after different immersion times in 37 °C water. The temperature at maximum peak of tan δ is defined as glass transition temperature (T_g). T_g of the composite is decreasing with the immersion time increasing. For example, with an immersion time of 0, 20, and 90 s, T_g is 61.1, 58.5, and 21.3 °C, respectively. Furthermore, the composite immersed in water for 40 s and 60 s reveals two or three transitions. Similar phenomenons have been reported by other researchers, which indicated that the water works in a gradual manner.^28,39 Fig. 8b shows the storage modulus of PVA/HHW with 5 wt% HHW loading after different immersion times in 37 °C water. It is obvious that the storage modulus of PVA/HHW composite is decreasing with the immersion time increasing. Both the T_g and storage modulus decreasing result in the shape recovery of the deformed composite.²⁸ However, the storage modulus of PVA/HHW are extremely high, the storage modulus of PVA/HHW with 5 wt% HHW loading is 22.24 GPa. Even immersion in water for 90 s, the storage modulus is still up to 11.09 GPa which much higher than other PVA composites. For instance, the storage modulus of PVA/0.5 wt% GO, PVA/0.5 wt% GO after immersion in water for 60 s, and PVA/4 wt% CNTs is 7.81 GPa, 4.14 GPa and 5.49 GPa, respectively.^27,28

Fig. 8 tan δ and storage modulus of PVA/HHW composite with 5 wt% HHW loading after being immersed in 37 °C water with different times.

To confirm T_g decreasing is induced by water weakened the hydrogen bonding between PVA and HHW, the ATR-FTIR spectrums of the PVA/HHW with 5 wt% HHW loading after different immersion times were carried out. Fig. 9 shows the ATR-FTIR spectrums of the PVA/HHW with 5 wt% HHW loading with different immersion times. The peak at 1640 cm⁻¹ is relevant to adsorbed water which increasing in the intensity of the band with the immersion time increase. According to Buslov and his co-worker, the band at 1640 cm⁻¹ is ascribed to bending vibration δ(O–H) of adsorbed water.⁴⁰ They also point out that one to three water molecules directly bind with one hydroxyl group of PVA in the hydrated film. Therefore, the ATR-FTIR results indicate that hydrogen boning is formed between adsorbed water and PVA, which weakens hydrogen bonding in the PVA/HHW composite.

Fig. 9 ATR-FTIR spectrums of PVA/HHW composite with 5 wt% HHW loading after being immersed in 37 °C water for different times.

In summary, a novel shape memory PVA/HHW composite is successfully synthesised using solvent casting method. HHW works as additional physical cross-linking points resulting in the improvement of mechanical properties of the composite. Simultaneously, PVA/HHW composite exhibits excellent water induced shape memory effect. It is worth noting that storage modulus of PVA/HHW composite is much higher than other PVA composites. Moreover, HH and PVA perform excellent biocompatibility,^41,42 therefore, this study provides a framework for developing new PVA based shape memory polymers applied in biomedicine.

Acknowledgements

This work is supported by the National Natural Science Foundation of China (21106075 and 21306094), Promotive research fund for excellent young and middle-aged scientists of Shandong Province (BS2012CL016), and Natural Science Foundation of Shandong Province (ZR2011BM006).

Notes and references

1. F. Ji, Y. Zhu, J. Hu, Y. Liu, L. Yeung and G. Ye, Smart Mater. Struct., 2006, 15, 1547–1554.
2. Y. Liu, K. Gall, M. L. Dunn and P. McCluskey, Mech. Mater., 2004, 36, 929.
3. R. Mohr, K. Kratz, T. Weigel, M. Lucka-Gabor, M. Moneke and A. Lendlein, Proc. Natl. Acad. Sci. U. S. A., 2006, 103, 3540.
4. A. M. Schmidt, Macromol. Rapid Commun., 2006, 27, 1168.
5. J. W. Cho, J. W. Kim, Y. C. Jung and N. S. Goo, Macromol. Rapid Commun., 2005, 26, 412.
6. N. G. Sahoo, Y. C. Jung and J. W. Cho, Mater. Manuf. Processes, 2007, 22, 419.
7. A. Lendlein, H. Y. Jiang, O. Junger and R. Langer, Nature, 2005, 434, 879.
8. A. Lendlein and R. Langer, Science, 2002, 96, 1673.
9. A. Metcalfe, A.-C. Desfaits, I. Salazkin, L. Yahiab, W. M. Sokolowskic and J. Raymonda, Biomaterials, 2003, 24, 491.
10. H. M. Wache, D. J. Tartakowska, A. Hentrich and M. H. Wagner, J. Mater. Sci.: Mater. Med., 2004, 14, 109.
11. Q. H. Meng, J. L. Hu, Y. Zhu, J. Lu and Y. Liu, Smart Mater. Struct., 2007, 16, 1192.
12. J. Leng, H. Lu and Y. Liu, Proc. SPIE, 2008, 693109.
13. X. Lan, W. M. Huang, N. Liu and J. S. Leng, Proc. SPIE-Int. Soc. Opt. Eng., 2008, 6927.
14. S. Rana, J. W. Cho and L. P. Tan, RSC Adv., 2013, 3, 13796.
15. N. G. Sahoo, Y. C. Jung, H. J. Yoo and J. W. Cho, Compos. Sci. Technol., 2007, 67, 1920.
16. M. Amirian, A. N. Chakoli, J. Sui and W. Cai, J. Polym. Res., 2012, 19, 9777.
17. C. Y. Bae, J. H. Park, E. Y. Kim, Y. S. Kang and B. K. Kim, J. Mater. Chem., 2011, 21, 11288.
18. B. Xu, Y. Q. Fu, M. Ahmad, J. K. Luo, W. M. Huang, A. Kraft, R. Reuben, Y. T. Pei, Z. G. Chen and J. Th. M. De Hosson, J. Mater. Chem., 2010, 20, 3442.
19. X. Song, L. Zhang, J. Zhao, Y. Xu, Z. Sun, P. Li and J. Yu, Cryst. Res. Technol., 2011, 46(2), 166–172.
20. W. Zhao, C. Gao, F. Guo and Y. Wu, Res. Chem. Intermed., 2016, 42, 2953.
21. W. Zhao, Y. Wu, J. Xu and C. Gao, RSC Adv., 2015, 5, 50544.
22. C. M. Hassan and N. A. Peppas, Macromolecules, 2000, 33, 2472.
23. T. Hirai, H. Maruyama, T. Suzuki and S. Hayashi, J. Appl. Polym. Sci., 1992, 45, 1849.
24. T. Hirai, H. Maruyama, T. Suzuki and S. Hayashi, J. Appl. Polym. Sci., 1992, 46, 1449.
25. H. Du and J. Zhang, Colloid Polym. Sci., 2010, 288, 15.
26. H. Du and J. Zhang, Soft Matter, 2010, 6, 3370.
27. F. Du, E. Ye, C. Tang, S. Ng, X. Zhou and X. Xie, J. Appl. Polym. Sci., 2013, 129(3), 1299.
28. X. Qi, X. Yao, S. Deng, T. Zhou and Q. Fu, J. Mater. Chem. A, 2014, 2, 2240.
29. Q. Bai, G. Zhang, B. Xu, X. Feng, H. Jiang and H. Li, RSC Adv., 2015, 5, 91213.
30. W. M. Huang and B. Yang, Appl. Phys. Lett., 2005, 86, 114105.
31. J. Mao, G. Jiang, Q. Chen and B. Guan, Colloids Surf., A, 2014, 443, 265.
32. S. Pandey, S. K. Pandey, V. Parashar, G. K. Mehrotrab and A. C. Pandey, J. Mater. Chem., 2011, 21, 17154.
33. J. Liang, Y. Huang, L. Zhang, Y. Wang, Y. Ma, T. Guo and Y. Chen, Adv. Funct. Mater., 2009, 19, 2297.
34. L. Liu, A. H. Barber, S. Nuriel and H. D. Wagner, Adv. Funct. Mater., 2005, 15, 975.
35. P. Song, Z. Xu and Q. Guo, ACS Macro Lett., 2013, 2, 1100.
36. A. Singhal, M. Kaur, K. A. Dubey, Y. K. Bhardwaj, D. Jain, C. G. S. Pillai and A. K. Tyagi, RSC Adv., 2012, 2, 7180.
37. G. Peng, F. Qiu, V. V. Ginzburg, D. Jasnow and A. C. Balazs, Science, 2000, 288, 1802.
38. A. Lendlein and S. Kelch, Angew. Chem., Int. Ed., 2002, 41, 2034.
39. B. Yang, W. M. Huang, C. Li and L. Li, Polymer, 2006, 47, 1348.
40. D. K. Buslov, N. I. Sushko and O. N. Tretinnikov, Polym. Sci., Ser. A, 2011, 53(12), 1121.
41. Z. Pan, Y. Lou, G. Yang, X. Ni, M. Chen, H. Xu, X. Miao, J. Liu, C. Hu and Q. Huang, Ceram. Int., 2013, 39, 5495.
42. Y. Shi, D. S. Xiong, Y. Peng and N. Wang, eXPRESS Polym. Lett., 2016, 10(2), 125.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra03717h

---