
Web Release Date: December 14,
Chitosan Nanofibers from an Easily Electrospinnable UHMWPEO-Doped Chitosan Solution System
Department of Oral & Dental Science, University of Bristol, Lower Maudlin Street, Bristol BS1 2LY, United Kingdom, and NUSNNI, Division of Bioengineering and Department of Mechanical Engineering, National University of Singapore, 9 Engineering Drive 1, Singapore 117576, Singapore
Received October 11, 2007
Revised November 5, 2007
Abstract:
Conversion of natural biopolymer chitosan into nanofibers through electrospinning has significant usefulness in various biomedical applications, in particular, for constructing a biomimetic and bioactive nanofibrous artificial extracellular matrix for engineering various tissues. Here, we show that introduction of an ultrahigh-molecular-weight poly(ethylene oxide) (UHMWPEO) into aqueous chitosan solutions remarkably enhances the formation of chitosan nanofibrous structure and leads to much lower loading of the water soluble fiber-forming aiding agent of PEO down to 5 wt % as compared to previous high PEO loadings in the electrospun chitosan nanofibers. The excellent electrospinnability of the current formulation renders electrospinning of natural biopolymer chitosan a robust process for large-scale production of practically applicable nanofibrous structures.
Fabrication of chitosan (CS) nanofibers through an ultrafine fiber manufacturing technique called electrospinning1–4 has recently been extensively investigated due to potential biomedical applications such as engineering tissues and delivery of pharmaceutical agents.5–12 In these studies, homogeneous chitosan nanofibers had been directly electrospun from a solution of pure chitosan dissolved in trifluoroacetic acid (TFA)6 or concentrated acetic acid (HAc).12 Many attempts have also been made to electrospin chitosan into nanofibers from the traditional dilute aqueous acetic acid solvent,5–8, 10, 11 which can reduce toxicity concerns and enhance biocompatibility of the nanofibrous membranes for biomedical applications.13, 14 For the latter, because of the inherent poor processability of the natural biopolymer chitosan itself, fiber-forming facilitating additives such as PEO and poly(vinyl alcohol) (PVA) were typically employed to improve the electrospinnability of chitosan and, quite often, the PEO or PVA loading amounts ranging from the lower 20 wt % to higher than 50 wt % were added to make a proper electrospinnable chitosan solution system,5, 7–10 which gave rise to varied electrospun fiber morphologies. While this approach enables generation of chitosan-containing composite nanofibers and likely contributes to improvement in other physical properties of chitosan (e.g., hydrophilicity and mechanical properties) as well, inclusion of a large quantity of the water-soluble PEO or PVA will eventually affect the structural integrity/stability,11, 15, 16 handling property (tacky), and mechanical performance of the chitosan nanofibrous structure. Lowering the amount of PEO inclusion in the electrospun chitosan nanofibers is therefore desired. Recently, Bhattarai et al.11 fabricated electrospun chitosan nanofibers with a PEO minimum loading of 10 wt %, which is the lowest PEO loading content so far reported in the literature. But much remains to be explored and improved in terms of having better fiber morphology (e.g., without splash) and further decreasing PEO ratio in the chitosan nanofibers.
In this paper, we report on the feasibility and efficacy of using UHMWPEO as the fiber-forming facilitating additive to electrospin chitosan nanofibers containing low amounts of PEO without sacrificing electrospinnability. We show that such a UHMWPEO-doped chitosan solution system surprisingly possessed superb electrospinnability that allowed us to produce not only nanoscale fibers of smaller than 100 nm in diameter but also unusually large diameter microfibers of a few tens of micrometers, which is not readily achievable in electrospinning of a polymer solution system in general.
Preparation of Spinning Solutions. Chitosan from crab shells with a degree of deacetylation of >85%, acetic acid (HAc, 99.8%), and dimethyl sulfoxide (DMSO, 99+%) were purchased from Sigma-Aldrich Co. Ltd. (Gillingham UK). Poly(ethylene oxide) with molecular weights of 850 000 and >5 000 000 Da (UHMWPEO) were supplied from Sumitomo Seika Chemicals Co. Ltd. (Japan) and Avocado Research Chemicals Ltd (UK), respectively. CS/UHMWPEO blends with different weight ratios of 100/0, 95/5, 90/10, 80/20, and 0/100 were dissolved in an aqueous mixed solvent system of 3 wt % acetic acid and DMSO (10:1, w/w) and were stirred with a powerful Multi-Stirrer MC301 (Scinics Co. Ltd., Tokyo) at room temperature for 24 h until transparent solutions were obtained. These spinning solutions, regardless of the CS/UHMWPEO ratios, had the same polymer concentration of 3 wt %. The shear viscosities of the spinning solutions were measured with a Gemini Advanced Rheometer (Bohlin Instruments), with a shear rate from 0 to 1000 per s.
Electrospinning. Electrospinning was performed in a custom-made chamber where the spinneret made from a B-D 22G11/2 tip-ground-to-flat needle was mounted on an electrically insulated stand. The spinneret needle was maintained at a high electric potential by a high voltage power supply (PS/EL30R01.5–22, Glassman High Voltage, Inc.) for electrospinning, and an aluminum plate with a dimension of 14 × 14 cm2 was used as the collector. The capillary needle spinneret was connected through PTFE tubing to a plastic syringe filled with a spinning solution. A constant volume flow rate was maintained using a KD Scientific syringe pump. The electric potential, solution flow rate, and the distance between the capillary tip and the collection plate were adjusted so that a stable jet was obtained.
Characterization. Images of electrospun fibers were obtained with a JEOL 6330F field emission scanning electron microscope operated at an accelerating voltage of 10 kV and 12 µA. Prior to imaging, the samples mounted on aluminum stubs were platinum coated with an AGAR high-resolution sputter-coater to 20 nm thickness for better conductivity during imaging. Average fiber diameters were determined by measuring fibers selected randomly from the SEM images. TEM analysis was performed on a JEOL 1200EX electron microscope operated at 120 kV. Samples for TEM were prepared by directly depositing nanofibers onto carbon-coated 3 mm diameter copper grids. The X-ray diffraction (XRD) patterns of the electrospun chitosan nanofibers were recorded at ambient temperature on an X’pert Pro Diffractormeter (PW3050/60, Philips, Netherlands). The samples were irradiated with monochromatized Cu Kα (1.5405 Å) X-ray source and analyzed between 5 and 40° (2θ). The operating voltage and current used were 40 kV and 30 mA, respectively, with a beam size of 20 µm.
If referring to the definition of spinnability for conventional fiber spinning,17 electrospinnability of a polymer solution can be understood as the intricate characteristic in which continuous uniform filaments can be obtained under electrospinning. In this study, incorporation of UHMWPEO significantly improved the fiber-forming ability or electrospinnability of chitosan solutions. First of all, given the same polymer concentration of 3 wt %, geometrically uniform and almost defects free chitosan nanofibrous structure were successfully generated from not only the relatively lower 20 and 10 wt % UHMWPEO-loaded chitosan solutions (Figure 1B,C) but also the lowest 5 wt % UHMWPEO loaded one (Figure 1D). The visually larger fibers as indicated by the arrows in the parts B, C, and D of Figure 1 were observed to be parallel combined twin (or multi-) fibers (high magnification images not shown). Similar observation was also reported in the previous study.18 We speculate that the initialization and ejecting of multiple jets, charges carried by the jets during electrospinning could be responsible for such occurrences. For UHMWPEO loadings of 20, 10, and 5 wt %, the resultant CS fiber diameters from randomly measurements of 60–80 fibers of the SEM images were found to be 102 ± 14, 138 ± 15, and 114 ± 19 nm, respectively, although there is no noticeable correlation between the fiber sizes and loading contents of the UHMWPEO. Taking the Pt coating thickness of 20 nm into account, the diameters of current CS nanofibers are actually very fine in the range of 60–100 nm. In contrast, for the PEO molecular weight of 850 K loaded chitosan solution, even at the higher PEO loading (i.e., 20 wt %), the produced fibers are nonuniform and accompanied with obvious defects appearing in the forms of large splashing spots, beads, and bonded fibers (Figure 1A), suggesting that even higher PEO loadings in CS are necessary in order to improve its electrospinnability and remove the unwanted defects.
It has been known that the bending instability19 or whipping20, 21 of the charged jet in electrospinning accounts for the formation of nanoscale fibers. Interestingly, in this study, it was noted that presence of the bending instability or whipping is related to the gap distance D between the spinneret and the collector. For the UHMWPEO-doped chitosan solutions, there always existed a critical distance Dc below which the dynamic bending instability or whipping failed to take place. As schematically shown in Figure 2a, when the gap distance D was smaller than Dc, electrospinning proceeded through forming a pretty long and stable single jet and producing larger sized microfibers (Figure 2c). When D > Dc, the chaotic whipping phenomenon occurred (Figure 2b), allowing the generation of nanoscale fibers (Figure 2d). Specifically in this study, at operating parameters of 10 kV applied voltage, 0.02 mL/min solution feeding rate, and ambient conditions of 45% humidity and 23 °C temperature, the critical distances for the solutions with UHMWPEO loading contents of 5, 10, and 20 wt % are approximately 16, 20, and 23 cm, respectively, which all are remarkably longer than the usual electrospinning having a stable region of merely a few centimeters at large. The longer stable jets suggest good fiber-forming ability of current chitosan spinning dopes and imply proper viscoelastic properties of these solutions for sustaining the jets without break-up. Consequently, jets bypassing the bending instability or whipping resulted in very large sized microfibers, which is several times greater than the largest fiber sizes (<10 µm) made from normal electrospinnable polymer solutions as reported14, 22, 23 and comparable to the sizes made from traditional wet/dry fiber-spinning methods.
Evidently, it is the superb electrospinnability of current UHMWPEO-doped chitosan that enables fabrication of CS fibers ranging from tiny nanoscale to larger microscale compared to those made from normal electropinnable polymer solutions. Thus, manoeuvrability for tailored fabrication of CS fibrous structure through electrospinning is conceivable. For example, it may provide a very promising solution for the limited cell infiltration depth problem where electrospun nanofibers are used for building up biomimicking scaffolds for culturing cells. Electrospun nanofibers resemble the physical dimensions of the native ECM constituents, however, the small pores/interstices formed from nanofibers lacing each others will be too small for cells to pass through.22, 24–27 To encourage cell migration into a nanofibrous scaffold interior, apart from endowing nanofibers with appropriate wettability and biochemical signals,25, 28–30 physical characteristics such as pore size, pore structure, pore distribution, and the overall porosity of the nanofibrous scaffolds play important roles. At this point, the flexibility in producing different sized fibers from current easily electrospinnable chitosan solution system will enable us to design and fabricate (e.g., through concurrent multielectrospinning fibers onto a controlled movable collector) novel microscale and nanoscale mingled chitosan fibrous scaffolds, which are biomimetic (nanofibers), possessing large pore and porosity (contributed from the introducing of large microfibers), and mechanically strong (contributed from the microfibers and integration between nanofibers and microfibers by forming bonding).
The excellent electrospinnability of current UHMWPEO-doped chitosan solution system can also be evidenced from the fact that large area of unidirectional chitosan nanofibrous web across a frame gap of over 20 cm was readily obtained by means of previously reported frame collecting method.2, 31, 32 Accordingly, chitosan nanofiber yarn or bundle from these aligned nanofibers was easily prepared, as shown in Figure 3. This certainly enhances the feasibility and opens up the opportunity of using aligned nanofibers for probing cellular responsive characteristics with such arrayed nanofeatures and texture of biomimetic nanofibrous substrates and for practical reconstructing tendon and ligament tissues where the aligned nanofibrous structure are needed.
| | Figure 3. Fiber yarn/bundle made from aligned electrospun chitosan nanofibers: (A) low magnification (300×); (B) high magnification (5000×). |
Because of its excellent electrospinnability, hydrophilicity, and biocompatibility, water-soluble PEO, the most studied polymer in electrospinning process, has been frequently used to facilitate electrospinning those difficult processing polymers33, 34 into an ultrafine fibrous structure, in particular for the natural biopolymers such as collagen,13 silk,14 alginate,35, 36 and chitosan. Most commonly, the used PEO molecular weight is usually several hundreds of kilodaltons. The PEO molecular weight employed for current chitosan electrospinning is distinctively high (>5 000 000 Da), which means a very long chain of the polymer. To the best of our knowledge, using such an ultrahigh molecular weight PEO has not been attempted for chitosan electrospinning. As expected, the UHMWPEO was found to possess very high viscosity and extremely good electrospinnability due to its superlong and flexible polymer chains. Its viscosity can be as high as 20 Pa·s at low shear rate, and the Dc for the pure UHMWPEO solution with the same processing conditions and solution concentration as used above was found to be larger than 40 cm. Thus, introduction of a small amount of UHMWPEO had remarkably increased the viscosity of the chitosan solutions (Figure 4) and consequently would contribute to establishing good chain entanglements between the chitosan and PEO molecules for fiber spinning.37 Another reason might be attributed to good molecular level interactions between the two polymers.
| | Figure 4. Relationship of shear viscosity and shear rate for the current chitosan solution formulation system. |
Figure 5 shows the diffraction patterns of pure chitosan, pure UHMWPEO, and their blends in fibrous form from electrospinning. Pure chitosan shows a strong reflection at 19.8° and a relatively weak reflection centering at 10.4° (Figure 5a), which are associated with the crystalline regions of the linear chitosan molecules. Pure UHMWPEO shows very strong reflections at 19.1° and 23.2° and some weak reflections indicating its overall higher degree of crystallinity (Figure 5e). If monitoring the qualitative changes in these major peaks to observe the crystalline characteristics of each component in the blends, we may find that those weak reflections for both polymers are absent and the strong ones became diminished and emerged with broadened single peaks (Figure 5c,d), suggesting a mutual interference effect in their respective crystallization probably because of formation of hydrogen bonds between ethers in PEO and amino groups in chitosan. The good molecular level interaction would facilitate forming molecular level miscible polymer blend system and interchain entanglement. As shown in the TEM image (Figure 6), there are no visible phase separation domains within the electrospun CS/UHMWPEO nanofiber. At ultrahigh magnification observation (1 000 000×), wormlike CS biomacromolecules (or the formed CS-UHMWPEO complex) can be visualized (the inset image). Being a polyelectrolyte polymer, chitosan chains or molecules carry positive charges in the dilute acetic acid solution. Positive charges associated with a CS macromolecule or chain causes repulsion forces between the charges, which not only restrict the CS molecule to form usual random coil configuration as a normal uncharged polymer chain does but also repel other CS chains to come close to itself, reducing the formation of chains entanglement. In this sense, the introduction of the nonionogenic and flexible super-long-chain UHMWPEO macromolecules and the establishment of molecular interactions (e.g., formation of hydrogen bonds) between the CS and PEO chains certainly facilitate the chain entanglement37 and consequently the excellent electrospinnability and formation of fibrous structure.
| | Figure 5. XRD patterns of (a) pure CS powder, (b,c,d) electrospun chitosan nanofibers with CS/UHMWPEO ratios at 80:20, 90:10, and 95:5, respectively, and (e) pure UHMWPEO powder. |
In summary, we have demonstrated that by introducing an ultrahigh-molecular-weight poly(ethylene oxide) into aqueous chitosan solution, chitosan nanofibers (with the loaded water-soluble PEO amount as low as up to 5 wt %) can be elegantly produced through electrospinning. This will warrant for certain the chitosan nanofibers good structural stability and handling property during practical applications compared to previous higher PEO loadings. Because of the excellent electrospinnability of the current solution system, we were able to electrospin both the extremely thin nanofibers (<100 nm in diameters) and large microfibers (few tens of micrometers in diameters), which have significant implications in developing biomimetic and bioactive 3-D cell−scaffold complex for engineering tissues. The results suggest that current eco-friendly and easily electrospinnable chitosan formulation could provide great potential for robust and scale-up production of the chitosan nanofibers for efficient practical applications in wound dressings, tissue engineering, drug delivery, and other industrial uses. Moreover, our approach can be similarly extended to make other versatile natural biopolymers such as collagen and silk into structurally stable nanofibers from their dilute acidic aqueous solutions instead of using a high-polarity organic solvent (e.g., hexafluroisopropanol, HFIP), which will not only give rise to the potential problem of modifying the native structures of the natural biopolymers but also other concerns such as cost, toxicity, and environmental hazard upon application.
The financial support from the UK EPSRC is acknowledged with gratitude. We also wish to cordially thank J. A. Jones from the Department of Chemistry (University of Bristol) for his support in electron microscopy, and X. He, T. R. Hayes, and T. Sjostrom in the Biomaterials Engineering Group of the Oral & Dental Science Department (University of Bristol) for their experimental assistance.
* Authors to whom correspondence should be addressed. E-mail: y.z.zhang@bristol.ac.uk (Y.Z.Z.); b.su@bristol.ac.uk (B.S.). Telephone: +44-117-928 4180/4361 . Fax: +44-117-928 4780.
† Department of Oral & Dental Science, University of Bristol.
‡ NUSNNI, Division of Bioengineering and Department of Mechanical Engineering, National University of Singapore.
1. Reneker, D.; Chun, I. Nanotechnology 1996, 7, 216–223.
2. Huang, Z.-M.; Zhang, Y.-Z.; Kotaki, M.; Ramakrishna, S. Compos. Sci. Technol. 2003, 63, 2223–2253.
3. Li, D.; Xia, Y. Adv. Mater. 2004, 16, 1151–1170.
4. Greiner, A.; Wendorff, J. H. Angew. Chem., Int. Ed. 2007, 46, 5670–5703.
5. Duan, B.; Dong, C.; Yuan, X.; Yao, K. J. Biomater. Sci., Polym. Ed. 2004, 15, 797–811.
6. Ohkawa, K.; Cha, D.; Kim, H.; Nishida, A.; Yamamoto, H. Macromol. Rapid Commun. 2004, 25, 1600–1605.
7. Li, L.; Hsieh, Y.-L. Carbohydr. Res. 2006, 341, 374–381.
8. Zhang, Y.; Huang, X.; Duan, B.; Wu, L.; Li, S.; Yuan, X. Colloid Polym. Sci. 2007, 285, 855–863.
9. Zhou, Y.; Yang, D.; Nie, J. J. Appl. Polym. Sci. 2006, 102, 5692–5697.
10. Jia, Y.-T.; Gong, J.; Gu, X.-H.; Kim, H.-Y.; Dong, J.; Shen, X.-Y. Carbohydr. Polym. 2007, 67, 403–409.
11. Bhattarai, N.; Edmondson, D.; Veiseh, O.; Matsen, F. A.; Zhang, M. Biomaterials 2005, 26, 6176–6184.
12. Geng, X.; Kwon, O.-H.; Jang, J. Biomaterials 2005, 26, 5427–5432.
13. Huang, L.; Nagapudi, K.; Apkarian, R. P.; Chaikof, E. L. J. Biomater. Sci., Polym. Ed. 2001, 12, 979–993.
14. Jin, H.-J.; Fridrikh, S. V.; Rutledge, G. C.; Kaplan, D. L. Biomacromolecules 2002, 3, 1233–1239.
15. Duan, B.; Yuan, X.; Zhu, Y.; Zhang, Y.; Li, X.; Zhang, Y.; Yao, K. Eur. Polym. J. 2006, 42, 2013–2022.
16. Ignatova, M.; Starbova, K.; Markova, N.; Manolova, N.; Rashkov, I. Carbohydr. Res. 2006, 341, 2098–2107.
17. Ziabicki, A. Fundamentals of Fiber Formation: The Science of Fiber Spinning and Drawing; New York: Wiley, 1976.
18. Zeng, J.; Xu, X.; Chen, X.; Liang, Q.; Bian, X.; Yang, L.; Jing, X. J. Controlled Release 2003, 92, 227–231.
19. Reneker, D. H.; Yarin, A. L.; Fong, H.; Koombhongse, S. J. Appl. Phys. 2000, 87, 4531–4547.
20. Shin, Y. M.; Hohman, M. M.; Brenner, M. P.; Rutledge, G. C. Polymer 2001, 42, 09955–09967.
21. Shin, Y. M.; Hohman, M. M.; Brenner, M. P.; Rutledge, G. C. Appl. Phys. Lett. 2001, 78, 1149–1151.
22. Pham, Q. P.; Sharma, U.; Mikos, A. G. Biomacromolecules 2006, 7, 2796–2805.
23. Stitzel, J.; Pawlowski, K.; Wnek, G.; Simpson, D.; Bowlin, G. J. Biomater. Appl. 2001, 16, 22–33.
24. Kidoaki, S.; Kwon, I. K.; Matsuda, T. Biomaterials 2005, 26, 37–46.
25. Badami, A. S.; Kreke, M. R.; Thompson, M. S.; Riffle, J. S.; Goldstein, A. S. Biomaterials 2006, 27, 596–606.
26. Stankus, J. J.; Guan, J.; Fujimoto, K.; Wagner, W. R. Biomaterials 2006, 27, 735–744.
27. Eichhorn, S. J.; Sampson, W. W. J. R. Soc. Interface 2005, 2, 309–318.
28. Telemeco, T. A.; Ayres, C.; Bowlin, G. L.; Wnek, G. E.; Boland, E. D.; Cohen, N.; Baumgarten, C. M.; Mathews, J.; Simpson, D. G. Acta Biomater. 2005, 1, 377–385.
29. Zhang, Y. Z.; Ouyang, H. W.; Lim, C. T.; Ramakrishna, S.; Huang, Z. M. J. Biomed. Mater. Res., Part B 2005, 72, 156–165.
30. Stankus, J. J.; Guan, J. J.; Wagner, W. R. J. Biomed. Mater. Res., Part A 2004, 70, 603–614.
31. Dersch, R.; Liu, T.; Schaper, A. K.; Greiner, A.; Wendorff, J. H. J. Polym. Sci., Part A: Polym. Chem. 2003, 41, 545–553.
32. Li, D.; Wang, Y.; Xia, Y. Adv. Mater. 2004, 16, 361–366.
33. Norris, I. D.; Shaker, M. M.; Ko, F. K.; MacDiarmid, A. G. Synth. Met. 2000, 114, 109–114.
34. Chronakis, I. S.; Grapenson, S.; Jakob, A. Polymer 2006, 47, 1597–1603.
35. Lu, J.-W.; Zhu, Y.-L.; Guo, Z.-X.; Hu, P.; Yu, J. Polymer 2006, 47, 8026–8031.
36. Bhattarai, N.; Li, Z.; Edmondson, D.; Zhang, M. Adv. Mater. 2006, 18, 1463–1467.
37. Shenoy, S. L.; Bates, W. D.; Frisch, H. L.; Wnek, G. E. Polymer 2005, 46, 3372–3384.