Hydrophobic modification of polypropylene/starch blend foams through tailoring cell diameter for oil-spill cleanup

Abstract

Frequent oil spillages and industrial discard of many organic solvents have created severe environmental and ecological problems. Therefore, it is imperative to find effective absorbent materials with high performance. Simultaneously, a green process of preparing such absorbent materials should be developed. Herein we present a facile approach to prepare open-cell polypropylene (PP)/starch blend foams with low density by twin-screw extrusion using water as a physical blowing agent and starch as an effective water carrier. The cell diameter of the prepared PP/starch blend foams was controlled by using different nucleating agents and changing the die geometry. Foams with mean cell diameter in the range of 0.4–4.5 mm and open-cell content larger than 90% were successfully obtained. Moreover, a remarkable improvement of hydrophobicity of the foams was obtained when decreasing the cell diameters. Consequently, the water contact angle and oil recovery efficiency were increased up to 142.2° and 98.4%, respectively, when the mean cell diameter was reduced to 0.4 mm. These characteristics make this foam a promising candidate absorbent material for use in oil-spill cleanup.

---

Introduction

Oil spillages and industrial discharge of organic solvents have created environmental and ecological problems in the sea and rivers.^1–4 The use of absorbents to remove oil/organic solvents from water has been considered as the most promising cleaning method in the remediation process due to its simple, efficient and fast countermeasures for the cleanup of oil spills.^5–7

The commonly used oil absorbents include porous inorganic particles,^8,9 fibers derived from both natural and synthetic polymers^10,11 and sponges.^12–15 However, these materials often suffer from problems of poor buoyancy, low oil sorption capacity and unsatisfactory selectivity. Recently, various advanced absorbent materials with superhydrophobicity and excellent oil sorption capacity have been developed to solve the above mentioned problems.^16–19 For example, superhydrophobic graphene-based sponges were successfully fabricated and the obtained absorbent demonstrated high selectivity, good recyclability and excellent oil absorption capacity.¹⁶

Polymer foams with low density and high open-cell content are attractive candidate oil absorbent materials since they are usually cheap and the production process is easy to scale up. The blowing agents are usually used for large-scale production of polymer foams in industry including both chemical and physical blowing agents. Traditional physical blowing agents, mainly including volatile organic hydrocarbons and their derivatives, are usually harmful to the environment and/or flammable.^20–22 Recently, we reported a facile and green extrusion foaming method for the production of polypropylene (PP)/starch blend foams using water as a blowing agent.²³ The results showed that PP/starch blend foams with low density and high open-cell content could be severed as a potential oil absorbent material. However, the hydrophobicity of the above mentioned foams needs to be further improved to better satisfy the oil-spill cleanup application.

Many previous studies showed that surfaces with good hydrophobicity could be obtained by the combination of low surface energy and proper surface roughness.^24–27 In the case of polymer foams, their cellular structure could provide a natural rough surface. Therefore, many researchers have tried to improve polymer foams' hydrophobicity by surface chemical modification.^28–30 For example, Zhou et al. prepared superhydrophobic sponge by a vapor-phase deposition process.²⁸ Surprisingly, very few literatures investigated the relationship between the cellular structure and hydrophobicity of polymer foams, although this may offer us a facile and green method of preparing polymer foams with good hydrophobicity without any chemical modifications.

Therefore, in this study, open-cell PP/starch blend foams with low density were first prepared by facile twin-screw extrusion using water as a physical blowing agent. Then we presented a simple and inexpensive method for the fabrication of hydrophobic open-cell foams without involving any chemical reaction by only tailoring the cell morphology. The cell diameter of the prepared PP/starch blend foams was tailored by adding different nucleating agents and altering the die geometry. Then, the relationship between cell diameter and hydrophobicity of the obtained foams was studied. Finally, the oil absorbing behavior of the prepared foams was further estimated in detail.

Experimental

Materials

A linear PP (T30S) with melt flow rate of 20 g/10 min was purchased from Jihua Petrochemical Corporation, China. Corn starch containing about 28% amylose was purchased from Changchun Dacheng Company, China. Talc with a mean particle size of 14 μm was supplied by Beitie Stone Powder Factory, Haicheng. Calcium carbonate (CaCO₃) with an average diameter of 13 μm was purchased from Dalian Jinsheng Fine Chemicals Corporation, China. Montmorillonite (MMT, DK4) was supplied by Zhejiang Fenghong New Material Corporation, China. All the raw materials were used as received. The chemical or crystal structure of the compounds used to prepare the foams was provided in Fig. S1.†

Extrusion foaming

Before extrusion foaming, starch samples were mixed with distilled water and talc/CaCO₃/MMT in a high-speed stand mixer for 2 min. After that the PP resin was added and then mixed for another 3 min. In this study, the weight blend ratio of PP and starch was kept at 70/30 and the water content of starch was fixed at 20 wt%. Finally, the mixtures were moved to a sealed plastic bag before extrusion.

A co-rotating twin-screw extruder (ZSK-35) was used in this study for extrusion foaming process. And a filamentary die with varying diameter was equipped at the end of the extrusion foaming system. The first three extruder segments were used to gelatinize the starch in which the temperatures were set at 60/80/100 °C. Then the temperature was increased to 190 °C for melting and mixing zone of the extruder. Finally, the temperature was progressively decreased to 170 °C during the cooling zone. The die temperature was kept at 150 °C. The screw speed was fixed at 300 rpm for all experiments and the feeding rate was estimated to be in the range of 16.2–16.8 kg h⁻¹. The foaming procedure was described in detail in our previous work.²³

Characterizations

The water displacement method was applied to measure the apparent density of the prepared foams. Among these tests, over 5 samples were used and then averaged for each formulation. The volume expansion ratio (ER) was determined by following equation:where ρ_f and ρ_m are the densities of the foam and the polymer matrix, respectively.

The foam samples were cut by a sharp blade or cryogenically fractured and then coated with gold before the observation by a scanning electron microscope (SEM). Image J software was used to analyze the obtained micrographs in order to gain cell parameters. The mean cell diameter (D) was calculated as the average diameter of over 100 cells while the cell density (N_c) was determined with:

where n is the number of cells in the SEM micrograph, M is the magnification factor and A is the area of the micrograph.

To test the open-cell content of the obtained foams, Ultrapyc 1200e was used and over 3 samples were tested for each formulation. In the equilibrium stage, a low gas (N₂) pressure (0.1 MPa) was set to minimize the rupture of the cells. Then measurements were carried out after the pressure reached an equilibrium value for 5 min.

Contact angle measurements were carried out using a contact angle goniometer (DSA 100, Kruss GmbH, Hamburg, Germany) at 25 °C. Water droplets were dropped carefully to the surface of the foams and the water contact angle values were determined by at least five separate measurements at different positions of a single sample.

During the oil sorption capacity tests, foam samples were immersed into 150 mL of oil in a glass beaker of 500 mL for 30 min. The oil sorption capacity was calculated according to following equation:

where S₀ is the original weight of the dry sorbents and S_t is the total weight of sorbents with absorbed oil.

To test the foams' selectivity of water and oil, a certain amount of pump oil (4.0 g, 8.0 g and 16 g) was poured into a 100 mL beaker which contained 50 mL deionized water and then placed on a shaker apparatus. The foam samples was first weighed and then added to the oil–water mixture, which was shaken for 30 min at three different frequencies: 0, 90 or 150 cycles per min. After that, the wetted sorbent was removed from the mixture and weighed. The water sorption content of the sorbent was determined according to ASTM D95. Then the net oil sorption was determined by subtracting the water content from the total amount of the sorption. Finally, the foams' selectivity could be estimated by oil recovery efficiency (R) which is calculated by following equation according to previous work:³¹

where M_o is the net oil sorption and M_t is the total amount of the sorption.

Results and discussion

Effect of different nucleating agents on cell diameter

PP/starch blend foams with low density and high open-cell content were first prepared by extrusion foaming using starch as a water carrier and water as a blowing agent. As seen in Fig. 1a, without addition of any nucleating agent, the obtained PP/starch blend foams demonstrated very poor cellular structure with quite large cells (D = 4.5 mm) and non-uniform cell size distribution. The low nucleating efficiency of PP/starch blend system using water as a blowing agent might be attributed to the high affinity of starch to water molecule due to the large amount of hydrogen bond between them. Therefore, the thermodynamic instability needed for nucleation was low for starch/water solution at high temperature and pressure, which resulted in low nucleation rate.

Fig. 1 Cellular morphology of PP/starch blend foams without or with different kinds of nucleating agents: (a) without nucleating agent; (b) 2 wt% CaCO₃; (c) 2 wt% MMT; (d) 2 wt% talc.

It has been proved that the addition of nucleating agents was an effective way to increase nucleating rate and improve the cellular structure.^32,33 In this study, three typical inorganic particles (talc, CaCO₃ and MMT) were selected to test their effectiveness when severing as nucleating agents. As seen in Fig. 1b–d, with the addition of nucleating agents, the cell size was significantly reduced and the cell size distribution became much more uniform. With the same loading, however, different nucleating agent showed quite different efficiency in increasing the nucleation rate of PP/starch blend system. The results showed that MMT was the least effective nucleating agent (D = 2.0 mm). This could be attributed to the high hydrophilicity nature of MMT particles that made it difficult for water molecules to form gas nuclei on their surface.³⁴ Considering the particle size of CaCO₃ and talc was comparable, their surface geometries might be the key factor that determined their efficiency in enhancing nucleation. It turned out that scaly talc demonstrated higher efficiency (D = 0.8 mm) when compared with spherical CaCO₃ (D = 1.8 mm). This result was consistent with one previous study reported by Leung et al.³⁵ According to the above discussion, talc was the best nucleating agent among the three studied ones and was selected as the nucleating agent in following sections.

Effect of talc contents on cell diameter

Fig. 2a demonstrated the dependence of mean cell diameter and cell density on talc contents. As expected, with increasing talc loading the potential nucleation sites within unit volume increased progressively. Therefore, more water was used for nucleation rather than cell growth that resulted in reduced mean cell diameter and increased cell density. For instance, the mean cell diameter decreased from 4.5 to 0.8 mm and the cell density increased from 6.0 × 10² to 2.5 × 10⁵ #/cm³, respectively, when talc content increased from 0 to 2 wt%. On the other hand, as the talc content turned high, the aggregation of talc particles also became severe. As a result, when talc content exceeded 2 wt%, the mean cell diameter remained almost unchanged as seen from Fig. 2a since the effective nucleation sites didn't increase much due to the aggregation of talc at high loadings.

Fig. 2 (a) Dependence of mean cell diameter and cell density on talc loading; (b) dependence of volume expansion ratio on talc loading.

It's noteworthy that if the talc loading was too high (5 wt%), the volume expansion ratio of PP/starch blend foams decreased significantly, as seen from Fig. 2b. It was found that the foams shrank a lot during extrusion foaming process when the talc content reached 5 wt% that led to the decease of volume expansion ratio. Apparently, the existence of inorganic particles within cell walls, especially at high loadings, would accelerate the cell wall rupture process since they could serve as potential defects.³⁶ Therefore, more cell walls would rupture at early stage of cell growth process leading to the shrinkage of extruded foams. According to the above discussion, other strategy should also be applied so as to further reduce the cell size and increase cell density without apparent reduction of volume expansion ratio.

Effect of die geometry on cell diameter

It has been proved that high nucleation rates could also be achieved by rapid pressure drop as the melts containing blowing agents pass the die since it will induce high thermal instability.³⁷ According to the power-law model, for a given polymer/gas solution the die pressure drop (ΔP) is determined by die geometry and flow rate:³⁸where m is a consistency constant, n is power law index, l and r is the length and radius of the die and Q is the volume flow rate.

Then the pressure drop rate (dP/dt) can be estimated by:³⁹

Based on theoretical calculation, high pressure drop rate can be achieved by the reduction of die diameter for a given polymer/gas solution passing a filamentary die. Therefore, a series of filamentary dies with varying die diameters and fixed length–diameter ratio (l/2r = 2.5) were used to study the effect of die geometry on the cellular structure of PP/starch blend foams using water as a blowing agent in this study. In this series of experiments the talc content was fixed at 2 wt%. As seen from Fig. 3, SEM observation of the cellular structure clearly showed that with decreasing die diameters, the mean cell size decreased and the cell density increased progressively due to the increased nucleation rate induced by rapid pressure drop. When the die diameter was reduced to 1.0 mm, the mean cell diameter of PP/starch blend foams could be further reduced to 0.4 mm and the cell density was increased up to 6.0 × 10⁵ #/cm³ (Fig. 4). In addition, the volume expansion ratio of the foams remained almost unchanged with varying die diameters.

Fig. 3 Cellular morphology of PP/starch blend foams extruded with different dies: (a) 2r = 4.0 mm; (b) 2r = 3.0 mm; (c) 2r = 2.0 mm; (d) 2r = 1.0 mm.

Fig. 4 Dependence of PP/starch blend foams' mean cell diameter and cell density on die diameter.

Open-cell content of PP/starch blend foams

Polymer foams with high open-cell content are preferred for oil absorption application since they will allow more oil to fill the space within cells easily. As seen in Fig. 5, regardless of varying talc contents or die diameters, the open-cell content of PP/starch blend foams remained at very high levels (>90%). The cell walls fracture mechanism during extrusion foaming of PP/starch blends have been discussed in detail in our previous work.²³ It was believed that at later stage of foaming, starch lost most water and transferred into glassy state while the PP resin was still in molten state. The deformability under stretching of glassy starch is quite lower than that of PP melt. Therefore, cell walls tended to rupture at the interface between starch and PP, leading to the high open-cell content of PP/starch blend foams. Apparently, the addition of talc or changing die diameters mainly altered the cell nucleation rate and had little effect on cell opening process. As a result, the open-cell contents of PP/starch blend foams remained at high levels when the talc contents and die diameters were varied.

Fig. 5 Open-cell content of PP/starch blend foams prepared with varying talc contents and die geometries.

Effect of cell diameter on hydrophobicity of PP/starch blend foams

As seen in Fig. 6, four typical PP/starch blend foams with different mean cell diameters and comparable apparent densities (20–23 kg m⁻³) were used to test their hydrophobicity. For comparison, water contact angle of flat PP films was also tested and the obtained value is 95.6°. And the water contact angle of starch films was reported to be 85.9° previously.⁴⁰ Interestingly, the foamed PP/starch blends demonstrated much better hydrophobicity. Moreover, it was found that the hydrophobicity improved obviously as the mean cell diameters of the foams decreased. When the mean cell diameter decreased to 0.4 mm, the water contact angle increased up to 142.2° that is close to the critical value of superhydrophobic surfaces (150°).²⁴

Fig. 6 Water contact angle of PP/starch blend foams with different mean cell diameters and comparable apparent densities (20–23 kg m⁻³).

The Cassie and Baxter model was often applied to explain the improved hydrophobicity of the rough surfaces with micro- or nanostructure when water droplets were much larger than the structures presented on substrate surface.⁴¹ Considering the cell size (0.4–4.5 mm) was comparable with water droplets, the Cassie and Baxter model was no longer suitable to correlate the improved hydrophobicity of PP/starch blend foams with reduced mean cell diameters. Herein, the cells were simplified into capillary tubes. When a water droplet was put on the top surface of the capillary tube, it would enter the tube due to its own gravity. Then, a meniscus with the shape depicted in Fig. 7 formed at the interface between the water and air since the inner surface of the capillary tube (i.e. cell) was mainly composed of hydrophobic PP resin. Meanwhile, a capillary pressure (ΔP) would generate, which could hinder the entering of water droplet into tube. When water droplet's gravity was larger than a single capillary pressure, it would go down and contact with more cells generating larger capillary pressure until the equilibrium was reached. According to Laplace equation:

where σ is the surface tension of water, R₁ and R₂ are the radius of the meniscus. Assuming that the meniscus is spherical (i.e. R₁ = R₂ = R_m), the Laplace equation becomes:

Fig. 7 Illustration of a spherical meniscus with water contact angle θ larger than 90°.

R_m can be calculated by following equation:

R_m = a cos(180 − θ) (9)

where a is the radius of capillary tube and θ is the water contact angle of the inner surface of cells. Apparently, as the mean cell diameter decreases a becomes smaller, which leads to the decreased R_m and increased ΔP. Therefore, water droplet will contact with fewer cells and maintain its shape more close to sphere demonstrating a larger water contact angle and improved hydrophobicity as seen in Fig. 6.

The water-repellent behavior of the obtained PP/starch blend foams was also tested to further confirm the hydrophobicity of them. As seen in Fig. 8a, foam sample with D = 1.8 mm only could support an approximately hemispherical dyed water droplet on its surface. And the water droplet only could partially be absorbed by a piece of filter paper. On the contrary, foam sample with D = 0.4 mm could support a nearly spherical water droplet on its surface and the droplet could easily be absorbed by the filter paper almost totally (Fig. 8b).

Fig. 8 Water-repellent behavior of PP/starch blend foams with different mean cell diameters: (a) D = 1.8 mm; (b) D = 0.4 mm.

Oil sorption behavior of PP/starch blend foams

Based on the above discussion, PP/starch blend foams with low density (20–23 kg m⁻³) and high open-cell content (>90%) could be prepared by a facile and green extrusion foaming method using water as a blowing agent. And then the hydrophobicity of the foams could be improved by tailoring their cell diameter through several strategies. It is believed that PP/starch blend foams occupied with low density, high open-cell content and good hydrophobicity could act as a promising absorbent material for oil/solvent–water separation. As illustrated in Fig. 9a–d, foam sample (D = 0.4 mm, ρ_f = 22 kg m⁻³) could selectively absorb corn oil (dyed with Sudan III) that floated on the surface of water in a short period of time. Besides, it could also absorb dyed chloroform droplet under water as it contact with the solvent (Fig. 9e–h). Then the absorbed oil/solvent could easily collected just by squeezing the foam samples.

Fig. 9 (a–d) Photographs of removing corn oil (dyed with Sudan III) from water surface with a PP/starch blend foam sample and (e–h) photographs of absorbing chloroform (dyed with Sudan III) under water with the foam sample.

To test the PP/starch blend foams' selectivity of water and oil quantitatively, the oil recovery efficiency of them was determined and the obtained results were demonstrated in Fig. 10. One could see that the selectivity of foams with large cell size (D = 4.5 mm) were very poor and they would absorb a certain high amount of water (24.4%) when they were immersed into oil–water mixtures. On the contrary, foams with small cell size (D = 0.4 mm) only absorb a little of water and the oil recovery efficiency was up to 98.4%. The foams' selectivity was positively correlated with their hydrophobicity which improved progressively as their mean cell diameter decreased, thus leading to the increasing oil recovery efficiency.

Fig. 10 Oil recovery efficiency of PP/starch blend foams with different mean cell diameters (testing conditions: pump oil films with a thickness of 2–3 mm; at fixed shaking frequency of 90 cycles per min).

It is noteworthy that the selectivity of the PP/starch blend foams was also affected by some other factors such as the viscosity of the oils, the thickness of the oil films and the shaking frequency. It was found that the oil film was more easily broken into pieces by turbulence of water when the viscosity of the tested oil was lower and/or the shaking frequency was higher. Therefore, the foam samples would have more opportunity to contact with water and absorb more water, which led to the decreased oil recovery efficiency (Fig. S2 and S3†). On the other hand, with increasing thickness of oil films, the foam sample were more likely surrounded by oils and absorb more oil leading to the increased oil recovery efficiency (Fig. S4†). Besides, it was also found that the selectivity of the foams with larger cell size was more sensitive to testing conditions due to their poor hydrophobicity. On the contrary, foams with small cell size (D = 0.4 mm) and good hydrophobicity remained a relatively high oil recovery efficiency (>95%) under varying testing conditions.

Fig. 11 described the sorption capacity of the foam samples (D = 0.4 mm, ρ_f = 22 kg m⁻³) quantitatively. It was found that PP/starch blend foam demonstrated good sorption capacity toward a wide range of oils and organic solvents. It could absorb oil and solvents up to 32–60 times of its own weight depending on the density of the liquids. It is noteworthy that since the volume expansion ratio and the open-cell content of PP/starch foams with different cell size were comparable, their sorption capacities were very similar.

Fig. 11 Sorption capacities of PP/starch blend foams toward several common oils and organic solvents.

The reusability of the PP/starch blend foams (D = 0.4 mm, ρ_f = 22 kg m⁻³) was also estimated using pump oil as an example and the obtained results were shown in Fig. 12. It was found that the foam samples could sustain cyclic compression and maintained an acceptable oil sorption capacity (31.8 g g⁻¹) after 10 cycles. The reduction of the foams' sorption capacity during cycles was attributed to the irreversible deformation of the foams caused by cyclic compression. In addition, the water contact angle of the foam sample decreased slightly to 139.6° after 10 cycles since a small part of the cells were destroyed during cyclic compression.

Fig. 12 Reusability of PP/starch blend foams.

Conclusions

A facile and green extrusion foaming method was applied in this study to prepare PP/starch blend foams with low density and high open-cell content. Then several strategies were used to tailor the cell diameter of the foam in order to improve its hydrophobicity. The results showed that open-cell PP/starch blend foams with mean cell diameter of 0.4 mm and cell density of 6.0 × 10⁵ #/cm³ were successfully obtained by adding a suitable nucleating agent and using a proper filamentary die. The water contact angle and oil recovery efficiency of this foam were up to 142.2° and 98.4%, which illustrated its good hydrophobicity and selectivity. Moreover, this foam could absorb a wide range of solvents/oils up to 32–60 times of its own weight. These characteristics make such foam a promising candidate absorbent material for oil-spill cleanup.

Acknowledgements

This work is supported by the National Natural Science Foundation of China (51273201, 51021003, 50703042). Part of this work is supported by Jilin Province Science and Technology Agency and Changchun Municipal Science and Technology Bureau.

Notes and references

1. C. P. Ruan, K. L. Ai, X. B. Li and L. H. Lu, Angew. Chem., Int. Ed., 2014, 53, 5556–5560.
2. P. Calcagnile, D. Fragouli, I. S. Bayer, G. C. Anyfantis, L. Martiradonna, P. D. Cozzoli, R. Cingolani and A. Athanassiou, ACS Nano, 2012, 6, 5413–5419.
3. Y. Liu, J. K. Ma, T. Wu, X. R. Wang, G. B. Huang, Y. Liu, H. X. Qiu, Y. Li, W. Wang and J. P. Gao, ACS Appl. Mater. Interfaces, 2013, 5, 10018–10026.
4. J. Lahann, Nat. Nanotechnol., 2008, 3, 320–321.
5. M. O. Adebajo, R. L. Frost, J. T. Kloprogge, O. Carmody and S. Kokot, J. Porous Mater., 2003, 10, 159–170.
6. D. Ceylan, S. Dogu, B. Karacik, S. D. Yakan, O. S. Okay and O. Okay, Environ. Sci. Technol., 2009, 43, 3846–3852.
7. X. Y. Zhou, Z. Z. Zhang, X. H. Xu, G. B. Huang, Y. Liu, X. H. Men and X. T. Zhu, Ind. Eng. Chem. Res., 2013, 52, 9411–9416.
8. A. Bayat, S. F. Aghamiri, A. Moheb and G. R. Vakili-Nezhaad, Chem. Eng. Technol., 2005, 28, 1525–1528.
9. C.-K. Tsai, C.-Y. Liao, H. P. Wang, Y.-C. Chien and C.-J. G. Jou, Mar. Pollut. Bull., 2008, 57, 895–898.
10. M. M. Radetic, D. M. Jocic, P. M. Jovancic, Z. L. Petrovic and H. F. Thomas, Environ. Sci. Technol., 2003, 37, 1008–1012.
11. Q. F. Wei, R. R. Mather, A. F. Fotheringham and R. D. Yang, Mar. Pollut. Bull., 2003, 46, 780–783.
12. J. Zhao, C. F. Xiao and N. K. Xu, Environ. Sci. Pollut. Res., 2013, 20, 4137–4145.
13. W. X. Liang and Z. G. Guo, RSC Adv., 2013, 3, 16469–16474.
14. C. F. Wang and S. J. Lin, ACS Appl. Mater. Interfaces, 2013, 5, 8861–8864.
15. L. Zhu, C. Li, J. Wang, H. Zhang, J. Zhang, Y. Shen, C. Li, C. Wang and A. Xie, Appl. Surf. Sci., 2012, 258, 6326–6330.
16. D. D. Nguyen, N. H. Tai, S. B. Lee and W. S. Kuo, Energy Environ. Sci., 2012, 5, 7908–7912.
17. Z. Y. Wu, C. Li, H. W. Liang, J. F. Chen and S. H. Yu, Angew. Chem., Int. Ed., 2013, 52, 2925–2929.
18. M. Cheng, Y. Gao, X. Guo, Z. Shi, J. Chen and F. Shi, Langmuir, 2011, 27, 7371–7375.
19. J. Yuan, X. Liu, O. Akbulut, J. Hu, S. L. Suib, J. Kong and F. Stellacci, Nat. Nanotechnol., 2008, 3, 332–336.
20. J. Crevecoeur, L. Nelissen and P. J. Lemstra, Polymer, 1999, 40, 3685–3689.
21. J. Pallay, P. Kelemen, H. Berghmans and D. V. Dommelen, Macromol. Mater. Eng., 2000, 275, 15–25.
22. L. J. Lee, C. C. Zeng, X. Cao, X. M. Han, J. Shen and G. J. Xu, Compos. Sci. Technol., 2005, 65, 2344–2363.
23. M. Z. Xu, J. J. Bian, C. Y. Han and L. S. Dong, Macromol. Mater. Eng., 2016, 301, 149–159.
24. A. Tuteja, W. Choi, M. Ma, J. M. Mabry, S. A. Mazzella, G. C. Rutledge, G. H. McKinley and R. E. Cohen, Science, 2007, 318, 1618–1622.
25. H. X. Wang, Y. H. Xue and T. Lin, Soft Matter, 2011, 7, 8158–8162.
26. M. Callies and D. Quere, Soft Matter, 2005, 1, 55–61.
27. M. Nosonovsky, Langmuir, 2007, 23, 3157–3161.
28. X. Y. Zhou, Z. Z. Zhang, X. H. Xu, X. H. Men and X. T. Zhu, Ind. Eng. Chem. Res., 2013, 52, 941–953.
29. H. Li, L. F. Liu and F. L. Yang, Mar. Pollut. Bull., 2012, 64, 1648–1653.
30. L. Peng, S. A. Yuan, G. Yan, P. Yu and Y. B. Luo, J. Appl. Polym. Sci., 2014, 131, 1366–1373.
31. A. Bayat, S. F. Aghamiri, A. Moheb and G. R. Vakili-Nezhaad, Chem. Eng. Technol., 2005, 28, 1525–1528.
32. R. H. Hansen and W. M. Martin, J. Polym. Sci Part B: Polym. Lett., 1965, 3, 325–331.
33. J. S. Colton and N. P. Suh, Polym. Eng. Sci., 1987, 27, 493–499.
34. C. L. Zhang, B. Zhu, D. C. Li and L. J. Lee, Polymer, 2012, 53, 2435–2442.
35. S. N. Leung, A. Wong, C. B. Park and J. H. Zong, J. Appl. Polym. Sci., 2008, 108, 3997–4003.
36. M. Mihai, M. A. Huneault, B. D. Favis and H. B. Li, Macromol. Biosci., 2007, 7, 907–920.
37. C. B. Park, D. F. Baldwin and N. P. Suh, Polym. Eng. Sci., 1995, 35, 432–440.
38. R. B. Bird, R. C. Armstrong and O. Hassager, Dynamics of Polymeric Liquids, Wiley, 1977.
39. H. E. Naguib, C. B. Park and P. C. Lee, J. Cellul. Plast., 2003, 39, 499–511.
40. J. W. Lawton, Starch/Staerke, 1995, 47, 62–67.
41. X. H. Wu and S. F. Wang, ACS Appl. Mater. Interfaces, 2012, 4, 4966–4975.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra19642j

---