- Email: [email protected]
Morphology and structure of electrospun mats from regenerated silk fibroin aqueous solutions with adjusting pH
International Journal of Biological Macromolecules 41 (2007) 469–474
Morphology and structure of electrospun mats from regenerated silk fibroin aqueous solutions with adjusting pH Jingxin Zhu, Huili Shao ∗ , Xuechao Hu State Key Laboratory for modification of Chemical Fibers and Polymer Materials, College of Material Science and Engineering, Donghua University, Shanghai 200051, PR China Received 30 January 2007; received in revised form 7 June 2007; accepted 18 June 2007 Available online 22 June 2007
Abstract In this paper, regenerated silk fibroin (SF) aqueous solutions were adjusted to a pH of 6.9 by mimicing the condition in the posterior division of silkworm’s gland and rheological behavior of solutions was investigated. The electrospinning technique was used to prepare fibers, and non-woven mats of regenerated B. mori silk fibroin were successfully obtained. The effects of electrospinning parameters on the morphology and diameter of regenerated silk fibers were investigated by orthogonal design. Statistical analysis showed that voltage, the concentration of regenerated SF solutions and the distance between tip and collection plate were the most dominant parameters to fiber morphology, diameter and diameter distribution, respectively. An optimal electrospinning condition was obtained in producing uniform cylindrical fibers with an average diameter of 1300 nm. It was as follows: the concentration 30%, voltage 40 kV, distance 20 cm. The structure of electrospun mats was characterized by Raman spectroscopy (RS), wide-angle X-ray diffraction (WAXD) and modulated differential scanning calorimetry (MDSC). It was found that electrospun mats were predominantly random coil/silk I structure, and the transition to silk II (-sheet) rich structure should be further explored. © 2007 Elsevier B.V. All rights reserved. Keywords: Electrospinning; Silk fibroin; Aqueous solutions; Morphology; Structure
1. Introduction Silks have attracted the interest of scientists of various disciplines for a long time [1,2]. In nature, fiber spinning for silkworm and spider, is based on the formation of concentrated fibroin and spidroin aqueous solutions that are then forced through small spinnerets into air. The fiber diameters produced in these natural spinning processes range from tens of micrometers to submicrometers. The production of fibers from protein solutions has typically relied upon the use of wet or dry spinning processes [3,4]. A wet spinning technique was developed by some groups to spin artificial silk protein fibers [5–8]. Recently, electrospinning has gained much attention because it is a rapid and relatively simple method to produce fibers with diameters ranging from micrometers down to tens of nanometers [9–11]. The electrospun silk fibroin (SF) fibers, with high specific surface area, high porosity and good biocompatibility, have extensive appli-
∗
Corresponding author. Tel.: +86 21 62373512; fax: +86 21 62378944. E-mail address: [email protected] (H. Shao).
0141-8130/$ – see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.ijbiomac.2007.06.006
cations in the field of biomaterials, such as tissue scaffolds and drug delivery [12]. Zarkoob et al. firstly reported that silkworm silk from Bombyx mori cocoons and spider dragline silk from Nephila claVipes silk can be electrospun into nanometer diameter fibers if solubilized in hexafluoro-2-propanol (HFIP) [13]. After that, many researchers tried to employ the electrospinning technique to prepare nanosized fibers with hexafluoroacetone-hydrate (HFAhydrate) and 98% formic acid as a spinning solvent (Ohgo et al., and Sukigara et al) [14–17]. Jin et al. firstly reported that SF fibers with a diameter range from 800 to 1000 nm were prepared by electrospinning from B. mori fibroin blending with PEO in aqueous solutions [18]. However, the use of caustic and organic solvents potentially compromises the biocompatibility and mechanical properties of silk fibers. Wang et al. tried to prepare electrospun SF fibers from pure B. mori fibroin aqueous solutions [19]. In this study, we desired to prepare the electrospun silk fibers by mimicing the silkworm’s spinning conditions, that is, the concentrated and weakly acidic SF aqueous solution was used [20]. The process was divided into three steps. First, regenerated SF solution was prepared
470
J. Zhu et al. / International Journal of Biological Macromolecules 41 (2007) 469–474
at a high concentration with water as solvent. Second, regenerated SF aqueous solution was adjusted to the same pH in posterior division of the silkworm’s gland. Third, SF fibers were produced by electrospinning technique at ambient conditions. Finally, the effects of electrospinning parameters on the morphology and diameter of regenerated SF fibers were investigated. 2. Experimental 2.1. Preparation of regenerated SF aqueous solution Cocoons of the Bombyx mori silkworm were degummed twice with 0.5 wt% Na2 CO3 solution at 100 ◦ C for 30 min and washed with warm distilled water to remove the sericin, then dried at room temperature. Degummed silk fibers were dissolved in a 9.0 M LiBr aqueous solution at 40 ◦ C for 2 h yielding a 10% (w/v) solution. After being diluted and centrifugal filtered, this solution was dialyzed in deioned water for 3 days with a cellulose semi-permeable membrane (MWCO 14,000). The dilute SF aqueous solution (about 2 wt%) gently condensed by forced airflow to a 20 wt% concentration at 10 ◦ C for about 5 days, the concentration of SF aqueous solution was controlled by velocity of airflow and monitored by weighing the remaining solid after drying.
2.4. Characterization The rheological measurements were performed on a HAAKE RS150LRheometer (Thermo Electron Co., Germany) with a (Ti, 35/1◦ ) cone plate. The shear rate was linearly increased from 0.01 to 200 s−1 , and the temperature was controlled at 25 ± 0.1 ◦ C. The morphology of the electrospun SF fibers was examined with a JSM-5600LV (JEOL Co., Japan) scanning electron microscope (SEM) at 10 kV and 1000 times magnification. The average diameter and diameter distribution were obtained by measuring 100 counts selected randomly in the SEM images with an image analysis software. Raman spectra were obtained with a LabRam-1B microscopy Raman spectrometer (Dilor, France). A He–Ne laser was used to give 6 mW of energy at 632.81 nm. The spectra were recorded from 900 to 1800 cm−1 . The wide-angle X-ray diffractograms were obtained on a D/Max-BR diffractometer (RigaKu, Japan) with Cu K␣ radiation in the 2θ range of 5–50◦ at 40 kV and 300 mA. The DSC experiments were carried out using a MDSC 2910 differential scanning calorimeter (TA Instruments Co., USA). The measurements were carried out in the range of 25–330 ◦ C under nitrogen gas with a scanning rate of 3 ◦ C/min. 3. Results and discussions
2.2. Preparation of spinning solution It was found that the pH of the SF solution was affected by the temperature and the concentration of SF aqueous solution. And regenerated SF aqueous solution prepared by this method at 10 ◦ C and a 20% concentration was weak alkalescence. It was well know that pH in the silkworm’s gland was varied and less than 7. Therefore, regenerated SF aqueous solution at a pH of 6.9, which is the same pH in the posterior division of silkworm’s gland [20], was prepared by adding 0.1 M citric acid–sodium hydroxide (NaOH)–hydrochloric acid (HCl) buffer regent into 20% concentrated SF aqueous solution in the ratio of about 1:3 (v:v), then SF aqueous solution was condensed to different concentrations at 10 ◦ C. 2.3. Electrospinning process In the electrospinning process, a high electric potential was applied to the wire on the syringe through a high voltage power supply, while the collection plate was grounded. When the electric force from the applied field became higher than the surface tension of the droplet, a charged jet of the solution was formed and ejected in the direction of the applied field. The non-woven mats were collected on a collecting plate (aluminum foil). The internal diameter of syringe tip was 0.6 mm. The flow rate of regenerated SF aqueous solutions was 2.0 ml/h. The electrospinning process was carried out at room temperature. The voltage and the distance between tip and collection plate could be adjusted.
3.1. Preparation and rheological behavior of regenerated SF aqueous solutions Properties of regenerated SF aqueous solutions are sensitive to solution temperature, fibroin concentration and shear condition [21]. If the silk solution is stored at room temperature, the viscosity of solution will rapidly increase and the solution subsequently converts to gel state. Therefore, all regenerated SF aqueous solutions were condensed and adjusted at a low temperature (10 ◦ C). Spinning solutions should be homogeneous and without any precipitates and air bubbles before electrospinning. Rheological behavior of solutions is very important for electrospinning. In this paper, the rheological behavior of regenerated SF aqueous solutions at pH 6.9 with different concentrations was investigated. Fig. 1 shows that the viscosity of lower concentrated solution (<30%) is independent of shear rate, that is, the solution is Newtonion. It is found that the viscosity of such regenerated SF aqueous solutions (<30%) is too low to electrospin. With the increase of concentration, shear thinning behavior can be observed even at lower shear rates, and when further increasing the shear rates, the viscosity tends to be constant. These show that there are intermolecular interactions at lower shear rates for concentrated regenerated SF aqueous solutions and constant viscosity is helpful for spinning. 3.2. Effects of processing parameters on morphology of electrospun silk fibers There are many important processing parameters that affect fiber morphology and diameter of regenerated SF fibers in elec-
J. Zhu et al. / International Journal of Biological Macromolecules 41 (2007) 469–474
Fig. 1. Rheological behaviors of regenerated SF aqueous solutions at pH 6.9 with different concentrations.
471
trospinning processing, such as solution concentration, voltage, the distance between tip and collection plate and the kind of solvent [22]. It is found that regenerated SF aqueous solutions with a pH of 6.9 cannot be electrospun, when concentration, voltage and the distance are less than 30%, 20 kV, 10 cm, respectively. In order to obtain a more systematic understanding of these process conditions and establish a quantitative basis for the relationship between electrospinning parameters and fiber diameter, L9 (34 ) orthogonal design was used to optimize the electrospinning parameters. Nine electrospinning conditions are shown in Table 1, and nine samples of regenerated SF fibers were obtained. SEM images, diameter distribution histograms, the average diameters and standard deviations of samples from 1 to 9 are shown in Table 2. Average diameter analysis of various levels are shown in Table 3. Range of three factors is 713, 512 and 424, respec-
Table 1 L9 (34 ) orthogonal design of electrospinning of regenerated SF aqueous solutions Samples
1 2 3 4 5 6 7 8 9
Experimental conditions
Results
Concentration (%)
Voltage (kV)
Distance (cm)
Average diameter (nm)
Standard deviation (nm)
30 30 30 33 33 33 38 38 38
20 30 40 20 30 40 20 30 40
10 15 20 15 20 10 20 10 15
1749 1507 1328 2775 1742 1602 2113 2439 2171
813 609 357 646 478 565 418 538 495
Table 2 SEM images, diameter distribution histograms, average diameters and standard deviation of samples from 1 to 9
472
J. Zhu et al. / International Journal of Biological Macromolecules 41 (2007) 469–474
Table 2 (Continued )
tively, it is found that the sequence of factors is concentration, voltage, distance according to range from high to low. Therefore, the concentration of regenerated SF solution is the most dominant factor to fiber average diameter. Standard deviation analysis of various levels are shown in Table 4. Range of three factors is 109, 154 and 221, respectively, it is found that the sequence of factors is dis-
tance, voltage, concentration according to range from high to low. Therefore, the distance between tip and collection plate is the most dominant factor to fiber the standard deviation. With the increasing of voltage, the shape of fiber changed from belt-like to uniform cylindrical, which showed that the voltage was the most dominant factor to fiber morphology. An
J. Zhu et al. / International Journal of Biological Macromolecules 41 (2007) 469–474
473
Table 3 Average diameter analysis of various levels (nm)
T1j T2j T3j M1j M2j M3j Rj
Concentration
Voltage
Distance
4584 6119 6723 1528 2040 2241 713
6637 5688 5101 2212 1896 1700 512
5790 6453 5183 1930 2151 1727 424
Table 4 Standard deviation analysis of various levels (nm)
T1j T2j T3j M1j M2j M3j Rj
Concentration
Voltage
Distance
1779 1689 1451 593 563 484 109
1877 1625 1417 626 542 472 154
1916 1714 1253 639 571 418 221
optimal electrospinning condition was obtained as follows: concentration 30%, voltage 40 kV, distance 20 cm. 3.3. Structural analysis of electrospun mats Raman spectra of degummed silk fibers and electrospun SF mats are shown in Fig. 2. The amide I, amide III and C C stretch bands of degummed silk fibers obviously appeared at 1666, 1230 and 1085 cm−1 , which are the characteristic of silk II (-sheet) conformation [23–25]. The amide I and amide III bands of electrospun SF mats appeared at 1660, 1276, 1242 cm−1 , which suggests that electrospun SF mats are in random-coiled or silk I conformation. In addition, Raman spectra of electrospun SF mats in C C stretch range also had the peaks at 1107, 950 and 930, which is attributed to the silk I conformation [24]. That is to say, random-coiled and silk I conformation coexist in electrospun SF mats.
Fig. 2. Raman spectra of (a) degummed silk fibers and (b) electrospun SF mats.
Fig. 3. X-ray diffracograms of (a) degummed silk fibers and (b) electrospun SF mats.
Fig. 3 compares the wide-angle X-ray diffractograms of degummed silk fibers with that of electrospun SF mats. Three peaks at 2θ of 9.1◦ , 20.6◦ , and 24.6◦ , which are characteristic peaks of silk II (-sheet) crystallite, are observed in degummed silk fibers. Electrospun SF mats showed a weak peak at 27.6◦ , which is the characteristic peak of silk I [26,27]. It means that there is a little silk I crystalline structure in electrospun mats. However, the main structure of electrospun mats is still randomcoiled (amorphous) under the present experiment condition. That is quite different from the semicrystalline structure (silk II) in degummed silk fibers. Fig. 4 shows the DSC curves of degummed silk fibers and electrospun SF mats. The DSC curve of degummed silk fibers displayed two endothermic peaks, one at around 75 ◦ C is due to loss of water, and another at 297 ◦ C is attributed to the thermal decomposition. However, the DSC curve of electrospun SF mats is different from that of degummed silk fibers. One endothermic peak at around 110 ◦ C is also due to loss of water and another peak at 267 ◦ C is attributed to the thermal decomposition of electrospun SF mats, which is lower than degummed silk fibers. The weak endothermic shift at 178 ◦ C indicates the glass transi-
Fig. 4. DSC curves of (a) degummed silk fiber and (b) electrospun SF mats.
474
J. Zhu et al. / International Journal of Biological Macromolecules 41 (2007) 469–474
tion (Tg) and the endo-exothermic peak at 258–264 ◦ C indicates conformational transitions from random coil/silk I to -sheet [28–30]. That further proved that electrospun SF mats has mainly amorphous and silk I structure.
[2] [3] [4] [5] [6]
4. Conclusions The electrospinning of concentrated regenerated SF solutions was successfully performed with all aqueous processing. Regenerated SF aqueous solutions were adjusted to a pH of 6.9 and rheological behavior of the solutions with different concentrations had been investigated, the result shows that when the concentration of regenerated SF aqueous solutions is less than 30%, the solution is Newtonion, and the viscosity is too low to electrospin. While for the more concentrated regenerated SF aqueous solutions, shear thinning behavior can be observed even at lower shear rates. The regenerated SF fibers were obtained by electrospinning with an average diameter of 1700 nm, which is much smaller than natural silk fiber. The effects of electrospinning parameters on the morphology and diameter of regenerated silk fibers were investigated by orthogonal design. An optimal electrospinning condition was obtained for producing uniform cylindrical fibers. Comparing with degummed silk fibers, the structure of electrospun mats was predominantly amorphous structure even though the chains were elongated during electrospinning. However, all aqueous electrospining processing of regenerated silk fibrous opens new directions in the fabrication of silk-based biomaterial for biomedical applications.
[7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20]
[21] [22] [23]
Acknowledgement
[24]
The authors thank for the financial support by Hi-Tech Research and Development Program of China (863 project), No.2002AA336060.
[25]
References [1] D.L. Kaplan, W.W. Adams, B. Farmer, C. Viney, Silk polymers: Materials Science and Biotechnology, in: ACS Symp Ser 544, American Chemical Society, Ed, Washington, DC, 1994, pp. 2–16.
[26] [27] [28] [29] [30]
G.H. Altman, F. Diaz, et al., Biomaterials 24 (2003) 401–416. F. Vollrath, Rev. Mol. Biotechnol. 74 (2000) 67–83. F. Vollrath, D.P. Knight, Nature 410 (2001) 541–548. O. Liivak, A. Blye, N. Shah, L.W. Jelinski, Macromolecules 31 (1998) 2947–2951. A. Seidel, O. Liivak, L.W. Jelinski, Macromolecules 31 (1998) 6733– 6736. A. Seidel, O. Liivak, S. Calve, et al., Macromolecules 33 (2000) 775–780. J. Yao, H. Masuda, C.H. Zhao, T. Asakura, Macromolecules 35 (2002) 6–9. D. Li, Y.N. Xia, Adv. Mater. 16 (2004) 1151–1170. W.K. Son, J.H. Youk, W.H. Park, Biomacromolecules 5 (2004) 197–201. X.Y. Geng, O.H. Kwon, J.H. Jang, Biomaterials 26 (2005) 5427–5432. B.M. Min, G. Lee, S.H. Kim, et al., Biomaterials 25 (2004) 1289–1297. S. Zarkoob, D.H. Reneker, R.K. Eby, S.D. Hudson, et al., Polymer Preprints 39 (1998) 244–245. K. Ohgo, C.H. Zhao, M. Kobayashi, T. Asakura, Polymer 44 (2003) 841–846. S. Sukigara, M. Gandhi, J. Ayutsede, M. Micklus, F. Ko, Polymer 44 (2003) 5721–5727. S. Sukigara, M. Gandhi, J. Ayutsede, M. Micklus, F. Ko, Polymer 45 (2004) 3701–3708. J. Ayutsede, M. Gandhi, S. Sukigara, M. Micklus, H.E. Chen, F. Ko, Polymer 46 (2005) 1625–1634. H.J. Jin, S.V. Fridrikh, G.C. Rutledge, D.L. Kaplan, Biomaromolecules 3 (2002) 1233–1239. H. Wang, Y.P. Zhang, H.L. Shao, X.C. Hu, J. Appl. Polym. Sci. 101 (2006) 961–968. J. Magoshi, Y. Magoshi, M.A. Becker, S. Nakamura, in: J.C. Salamone (Ed.), Polymeric Materials Encyclopedia, CRC Press, Florid, 1996, pp. 667–679. K.A. Trabbic, P. Yager, Macromolecules 31 (1998) 462–471. T. Subbiah, G.S. Bhat, R.W. Tock, S. Parameswaran, S.S. Ramkumar, J. Appl. Polym. Sci. 96 (2005) 557–569. P. Monti, G. Freddi, A. Bertoluzza, N. Kasai, M. Tsukada, J. Raman Spectrosc. 29 (1998) 297–304. P. Monti, P. Taddei, G. Freddi, T. Asakura, M. Tsukada, J. Raman Spectrosc. 32 (2001) 103–107. M.E. Rousseau, T. Lefevre, L. Beaulieu, T. Asakura, et al., Biomacromolecules 5 (2004) 2247–2257. I.C. Um, H.Y. Kweon, Y.H. Park, S. Hudson, Int. J. Biol. Macromol. 29 (2001) 91–97. S.W. Ha, Y.H. Park, S.M. Hudson, Biomacromolecules 4 (2003) 488–496. J. Magoshi, S. Nakamura, J. Appl. Polym. Sci. 19 (1975) 1013–1015. J. Magoshi, Y. Magoshi, S. Nakamura, N. Kasai, M. Kakudo, J. Polym. Sci., Part B: Polym. Phys. 15 (1977) 1675–1683. M. Tsukada, G. Freddi, N. Kasai, J. Polym. Sci., Part B: Polym. Phys. 32 (1994) 1175–1182.
Morphology and structure of electrospun mats from regenerated silk fibroin aqueous solutions with adjusting pH Jingxin Zhu, Huili Shao ∗ , Xuechao Hu State Key Laboratory for modification of Chemical Fibers and Polymer Materials, College of Material Science and Engineering, Donghua University, Shanghai 200051, PR China Received 30 January 2007; received in revised form 7 June 2007; accepted 18 June 2007 Available online 22 June 2007
Abstract In this paper, regenerated silk fibroin (SF) aqueous solutions were adjusted to a pH of 6.9 by mimicing the condition in the posterior division of silkworm’s gland and rheological behavior of solutions was investigated. The electrospinning technique was used to prepare fibers, and non-woven mats of regenerated B. mori silk fibroin were successfully obtained. The effects of electrospinning parameters on the morphology and diameter of regenerated silk fibers were investigated by orthogonal design. Statistical analysis showed that voltage, the concentration of regenerated SF solutions and the distance between tip and collection plate were the most dominant parameters to fiber morphology, diameter and diameter distribution, respectively. An optimal electrospinning condition was obtained in producing uniform cylindrical fibers with an average diameter of 1300 nm. It was as follows: the concentration 30%, voltage 40 kV, distance 20 cm. The structure of electrospun mats was characterized by Raman spectroscopy (RS), wide-angle X-ray diffraction (WAXD) and modulated differential scanning calorimetry (MDSC). It was found that electrospun mats were predominantly random coil/silk I structure, and the transition to silk II (-sheet) rich structure should be further explored. © 2007 Elsevier B.V. All rights reserved. Keywords: Electrospinning; Silk fibroin; Aqueous solutions; Morphology; Structure
1. Introduction Silks have attracted the interest of scientists of various disciplines for a long time [1,2]. In nature, fiber spinning for silkworm and spider, is based on the formation of concentrated fibroin and spidroin aqueous solutions that are then forced through small spinnerets into air. The fiber diameters produced in these natural spinning processes range from tens of micrometers to submicrometers. The production of fibers from protein solutions has typically relied upon the use of wet or dry spinning processes [3,4]. A wet spinning technique was developed by some groups to spin artificial silk protein fibers [5–8]. Recently, electrospinning has gained much attention because it is a rapid and relatively simple method to produce fibers with diameters ranging from micrometers down to tens of nanometers [9–11]. The electrospun silk fibroin (SF) fibers, with high specific surface area, high porosity and good biocompatibility, have extensive appli-
∗
Corresponding author. Tel.: +86 21 62373512; fax: +86 21 62378944. E-mail address: [email protected] (H. Shao).
0141-8130/$ – see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.ijbiomac.2007.06.006
cations in the field of biomaterials, such as tissue scaffolds and drug delivery [12]. Zarkoob et al. firstly reported that silkworm silk from Bombyx mori cocoons and spider dragline silk from Nephila claVipes silk can be electrospun into nanometer diameter fibers if solubilized in hexafluoro-2-propanol (HFIP) [13]. After that, many researchers tried to employ the electrospinning technique to prepare nanosized fibers with hexafluoroacetone-hydrate (HFAhydrate) and 98% formic acid as a spinning solvent (Ohgo et al., and Sukigara et al) [14–17]. Jin et al. firstly reported that SF fibers with a diameter range from 800 to 1000 nm were prepared by electrospinning from B. mori fibroin blending with PEO in aqueous solutions [18]. However, the use of caustic and organic solvents potentially compromises the biocompatibility and mechanical properties of silk fibers. Wang et al. tried to prepare electrospun SF fibers from pure B. mori fibroin aqueous solutions [19]. In this study, we desired to prepare the electrospun silk fibers by mimicing the silkworm’s spinning conditions, that is, the concentrated and weakly acidic SF aqueous solution was used [20]. The process was divided into three steps. First, regenerated SF solution was prepared
470
J. Zhu et al. / International Journal of Biological Macromolecules 41 (2007) 469–474
at a high concentration with water as solvent. Second, regenerated SF aqueous solution was adjusted to the same pH in posterior division of the silkworm’s gland. Third, SF fibers were produced by electrospinning technique at ambient conditions. Finally, the effects of electrospinning parameters on the morphology and diameter of regenerated SF fibers were investigated. 2. Experimental 2.1. Preparation of regenerated SF aqueous solution Cocoons of the Bombyx mori silkworm were degummed twice with 0.5 wt% Na2 CO3 solution at 100 ◦ C for 30 min and washed with warm distilled water to remove the sericin, then dried at room temperature. Degummed silk fibers were dissolved in a 9.0 M LiBr aqueous solution at 40 ◦ C for 2 h yielding a 10% (w/v) solution. After being diluted and centrifugal filtered, this solution was dialyzed in deioned water for 3 days with a cellulose semi-permeable membrane (MWCO 14,000). The dilute SF aqueous solution (about 2 wt%) gently condensed by forced airflow to a 20 wt% concentration at 10 ◦ C for about 5 days, the concentration of SF aqueous solution was controlled by velocity of airflow and monitored by weighing the remaining solid after drying.
2.4. Characterization The rheological measurements were performed on a HAAKE RS150LRheometer (Thermo Electron Co., Germany) with a (Ti, 35/1◦ ) cone plate. The shear rate was linearly increased from 0.01 to 200 s−1 , and the temperature was controlled at 25 ± 0.1 ◦ C. The morphology of the electrospun SF fibers was examined with a JSM-5600LV (JEOL Co., Japan) scanning electron microscope (SEM) at 10 kV and 1000 times magnification. The average diameter and diameter distribution were obtained by measuring 100 counts selected randomly in the SEM images with an image analysis software. Raman spectra were obtained with a LabRam-1B microscopy Raman spectrometer (Dilor, France). A He–Ne laser was used to give 6 mW of energy at 632.81 nm. The spectra were recorded from 900 to 1800 cm−1 . The wide-angle X-ray diffractograms were obtained on a D/Max-BR diffractometer (RigaKu, Japan) with Cu K␣ radiation in the 2θ range of 5–50◦ at 40 kV and 300 mA. The DSC experiments were carried out using a MDSC 2910 differential scanning calorimeter (TA Instruments Co., USA). The measurements were carried out in the range of 25–330 ◦ C under nitrogen gas with a scanning rate of 3 ◦ C/min. 3. Results and discussions
2.2. Preparation of spinning solution It was found that the pH of the SF solution was affected by the temperature and the concentration of SF aqueous solution. And regenerated SF aqueous solution prepared by this method at 10 ◦ C and a 20% concentration was weak alkalescence. It was well know that pH in the silkworm’s gland was varied and less than 7. Therefore, regenerated SF aqueous solution at a pH of 6.9, which is the same pH in the posterior division of silkworm’s gland [20], was prepared by adding 0.1 M citric acid–sodium hydroxide (NaOH)–hydrochloric acid (HCl) buffer regent into 20% concentrated SF aqueous solution in the ratio of about 1:3 (v:v), then SF aqueous solution was condensed to different concentrations at 10 ◦ C. 2.3. Electrospinning process In the electrospinning process, a high electric potential was applied to the wire on the syringe through a high voltage power supply, while the collection plate was grounded. When the electric force from the applied field became higher than the surface tension of the droplet, a charged jet of the solution was formed and ejected in the direction of the applied field. The non-woven mats were collected on a collecting plate (aluminum foil). The internal diameter of syringe tip was 0.6 mm. The flow rate of regenerated SF aqueous solutions was 2.0 ml/h. The electrospinning process was carried out at room temperature. The voltage and the distance between tip and collection plate could be adjusted.
3.1. Preparation and rheological behavior of regenerated SF aqueous solutions Properties of regenerated SF aqueous solutions are sensitive to solution temperature, fibroin concentration and shear condition [21]. If the silk solution is stored at room temperature, the viscosity of solution will rapidly increase and the solution subsequently converts to gel state. Therefore, all regenerated SF aqueous solutions were condensed and adjusted at a low temperature (10 ◦ C). Spinning solutions should be homogeneous and without any precipitates and air bubbles before electrospinning. Rheological behavior of solutions is very important for electrospinning. In this paper, the rheological behavior of regenerated SF aqueous solutions at pH 6.9 with different concentrations was investigated. Fig. 1 shows that the viscosity of lower concentrated solution (<30%) is independent of shear rate, that is, the solution is Newtonion. It is found that the viscosity of such regenerated SF aqueous solutions (<30%) is too low to electrospin. With the increase of concentration, shear thinning behavior can be observed even at lower shear rates, and when further increasing the shear rates, the viscosity tends to be constant. These show that there are intermolecular interactions at lower shear rates for concentrated regenerated SF aqueous solutions and constant viscosity is helpful for spinning. 3.2. Effects of processing parameters on morphology of electrospun silk fibers There are many important processing parameters that affect fiber morphology and diameter of regenerated SF fibers in elec-
J. Zhu et al. / International Journal of Biological Macromolecules 41 (2007) 469–474
Fig. 1. Rheological behaviors of regenerated SF aqueous solutions at pH 6.9 with different concentrations.
471
trospinning processing, such as solution concentration, voltage, the distance between tip and collection plate and the kind of solvent [22]. It is found that regenerated SF aqueous solutions with a pH of 6.9 cannot be electrospun, when concentration, voltage and the distance are less than 30%, 20 kV, 10 cm, respectively. In order to obtain a more systematic understanding of these process conditions and establish a quantitative basis for the relationship between electrospinning parameters and fiber diameter, L9 (34 ) orthogonal design was used to optimize the electrospinning parameters. Nine electrospinning conditions are shown in Table 1, and nine samples of regenerated SF fibers were obtained. SEM images, diameter distribution histograms, the average diameters and standard deviations of samples from 1 to 9 are shown in Table 2. Average diameter analysis of various levels are shown in Table 3. Range of three factors is 713, 512 and 424, respec-
Table 1 L9 (34 ) orthogonal design of electrospinning of regenerated SF aqueous solutions Samples
1 2 3 4 5 6 7 8 9
Experimental conditions
Results
Concentration (%)
Voltage (kV)
Distance (cm)
Average diameter (nm)
Standard deviation (nm)
30 30 30 33 33 33 38 38 38
20 30 40 20 30 40 20 30 40
10 15 20 15 20 10 20 10 15
1749 1507 1328 2775 1742 1602 2113 2439 2171
813 609 357 646 478 565 418 538 495
Table 2 SEM images, diameter distribution histograms, average diameters and standard deviation of samples from 1 to 9
472
J. Zhu et al. / International Journal of Biological Macromolecules 41 (2007) 469–474
Table 2 (Continued )
tively, it is found that the sequence of factors is concentration, voltage, distance according to range from high to low. Therefore, the concentration of regenerated SF solution is the most dominant factor to fiber average diameter. Standard deviation analysis of various levels are shown in Table 4. Range of three factors is 109, 154 and 221, respectively, it is found that the sequence of factors is dis-
tance, voltage, concentration according to range from high to low. Therefore, the distance between tip and collection plate is the most dominant factor to fiber the standard deviation. With the increasing of voltage, the shape of fiber changed from belt-like to uniform cylindrical, which showed that the voltage was the most dominant factor to fiber morphology. An
J. Zhu et al. / International Journal of Biological Macromolecules 41 (2007) 469–474
473
Table 3 Average diameter analysis of various levels (nm)
T1j T2j T3j M1j M2j M3j Rj
Concentration
Voltage
Distance
4584 6119 6723 1528 2040 2241 713
6637 5688 5101 2212 1896 1700 512
5790 6453 5183 1930 2151 1727 424
Table 4 Standard deviation analysis of various levels (nm)
T1j T2j T3j M1j M2j M3j Rj
Concentration
Voltage
Distance
1779 1689 1451 593 563 484 109
1877 1625 1417 626 542 472 154
1916 1714 1253 639 571 418 221
optimal electrospinning condition was obtained as follows: concentration 30%, voltage 40 kV, distance 20 cm. 3.3. Structural analysis of electrospun mats Raman spectra of degummed silk fibers and electrospun SF mats are shown in Fig. 2. The amide I, amide III and C C stretch bands of degummed silk fibers obviously appeared at 1666, 1230 and 1085 cm−1 , which are the characteristic of silk II (-sheet) conformation [23–25]. The amide I and amide III bands of electrospun SF mats appeared at 1660, 1276, 1242 cm−1 , which suggests that electrospun SF mats are in random-coiled or silk I conformation. In addition, Raman spectra of electrospun SF mats in C C stretch range also had the peaks at 1107, 950 and 930, which is attributed to the silk I conformation [24]. That is to say, random-coiled and silk I conformation coexist in electrospun SF mats.
Fig. 2. Raman spectra of (a) degummed silk fibers and (b) electrospun SF mats.
Fig. 3. X-ray diffracograms of (a) degummed silk fibers and (b) electrospun SF mats.
Fig. 3 compares the wide-angle X-ray diffractograms of degummed silk fibers with that of electrospun SF mats. Three peaks at 2θ of 9.1◦ , 20.6◦ , and 24.6◦ , which are characteristic peaks of silk II (-sheet) crystallite, are observed in degummed silk fibers. Electrospun SF mats showed a weak peak at 27.6◦ , which is the characteristic peak of silk I [26,27]. It means that there is a little silk I crystalline structure in electrospun mats. However, the main structure of electrospun mats is still randomcoiled (amorphous) under the present experiment condition. That is quite different from the semicrystalline structure (silk II) in degummed silk fibers. Fig. 4 shows the DSC curves of degummed silk fibers and electrospun SF mats. The DSC curve of degummed silk fibers displayed two endothermic peaks, one at around 75 ◦ C is due to loss of water, and another at 297 ◦ C is attributed to the thermal decomposition. However, the DSC curve of electrospun SF mats is different from that of degummed silk fibers. One endothermic peak at around 110 ◦ C is also due to loss of water and another peak at 267 ◦ C is attributed to the thermal decomposition of electrospun SF mats, which is lower than degummed silk fibers. The weak endothermic shift at 178 ◦ C indicates the glass transi-
Fig. 4. DSC curves of (a) degummed silk fiber and (b) electrospun SF mats.
474
J. Zhu et al. / International Journal of Biological Macromolecules 41 (2007) 469–474
tion (Tg) and the endo-exothermic peak at 258–264 ◦ C indicates conformational transitions from random coil/silk I to -sheet [28–30]. That further proved that electrospun SF mats has mainly amorphous and silk I structure.
[2] [3] [4] [5] [6]
4. Conclusions The electrospinning of concentrated regenerated SF solutions was successfully performed with all aqueous processing. Regenerated SF aqueous solutions were adjusted to a pH of 6.9 and rheological behavior of the solutions with different concentrations had been investigated, the result shows that when the concentration of regenerated SF aqueous solutions is less than 30%, the solution is Newtonion, and the viscosity is too low to electrospin. While for the more concentrated regenerated SF aqueous solutions, shear thinning behavior can be observed even at lower shear rates. The regenerated SF fibers were obtained by electrospinning with an average diameter of 1700 nm, which is much smaller than natural silk fiber. The effects of electrospinning parameters on the morphology and diameter of regenerated silk fibers were investigated by orthogonal design. An optimal electrospinning condition was obtained for producing uniform cylindrical fibers. Comparing with degummed silk fibers, the structure of electrospun mats was predominantly amorphous structure even though the chains were elongated during electrospinning. However, all aqueous electrospining processing of regenerated silk fibrous opens new directions in the fabrication of silk-based biomaterial for biomedical applications.
[7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20]
[21] [22] [23]
Acknowledgement
[24]
The authors thank for the financial support by Hi-Tech Research and Development Program of China (863 project), No.2002AA336060.
[25]
References [1] D.L. Kaplan, W.W. Adams, B. Farmer, C. Viney, Silk polymers: Materials Science and Biotechnology, in: ACS Symp Ser 544, American Chemical Society, Ed, Washington, DC, 1994, pp. 2–16.
[26] [27] [28] [29] [30]
G.H. Altman, F. Diaz, et al., Biomaterials 24 (2003) 401–416. F. Vollrath, Rev. Mol. Biotechnol. 74 (2000) 67–83. F. Vollrath, D.P. Knight, Nature 410 (2001) 541–548. O. Liivak, A. Blye, N. Shah, L.W. Jelinski, Macromolecules 31 (1998) 2947–2951. A. Seidel, O. Liivak, L.W. Jelinski, Macromolecules 31 (1998) 6733– 6736. A. Seidel, O. Liivak, S. Calve, et al., Macromolecules 33 (2000) 775–780. J. Yao, H. Masuda, C.H. Zhao, T. Asakura, Macromolecules 35 (2002) 6–9. D. Li, Y.N. Xia, Adv. Mater. 16 (2004) 1151–1170. W.K. Son, J.H. Youk, W.H. Park, Biomacromolecules 5 (2004) 197–201. X.Y. Geng, O.H. Kwon, J.H. Jang, Biomaterials 26 (2005) 5427–5432. B.M. Min, G. Lee, S.H. Kim, et al., Biomaterials 25 (2004) 1289–1297. S. Zarkoob, D.H. Reneker, R.K. Eby, S.D. Hudson, et al., Polymer Preprints 39 (1998) 244–245. K. Ohgo, C.H. Zhao, M. Kobayashi, T. Asakura, Polymer 44 (2003) 841–846. S. Sukigara, M. Gandhi, J. Ayutsede, M. Micklus, F. Ko, Polymer 44 (2003) 5721–5727. S. Sukigara, M. Gandhi, J. Ayutsede, M. Micklus, F. Ko, Polymer 45 (2004) 3701–3708. J. Ayutsede, M. Gandhi, S. Sukigara, M. Micklus, H.E. Chen, F. Ko, Polymer 46 (2005) 1625–1634. H.J. Jin, S.V. Fridrikh, G.C. Rutledge, D.L. Kaplan, Biomaromolecules 3 (2002) 1233–1239. H. Wang, Y.P. Zhang, H.L. Shao, X.C. Hu, J. Appl. Polym. Sci. 101 (2006) 961–968. J. Magoshi, Y. Magoshi, M.A. Becker, S. Nakamura, in: J.C. Salamone (Ed.), Polymeric Materials Encyclopedia, CRC Press, Florid, 1996, pp. 667–679. K.A. Trabbic, P. Yager, Macromolecules 31 (1998) 462–471. T. Subbiah, G.S. Bhat, R.W. Tock, S. Parameswaran, S.S. Ramkumar, J. Appl. Polym. Sci. 96 (2005) 557–569. P. Monti, G. Freddi, A. Bertoluzza, N. Kasai, M. Tsukada, J. Raman Spectrosc. 29 (1998) 297–304. P. Monti, P. Taddei, G. Freddi, T. Asakura, M. Tsukada, J. Raman Spectrosc. 32 (2001) 103–107. M.E. Rousseau, T. Lefevre, L. Beaulieu, T. Asakura, et al., Biomacromolecules 5 (2004) 2247–2257. I.C. Um, H.Y. Kweon, Y.H. Park, S. Hudson, Int. J. Biol. Macromol. 29 (2001) 91–97. S.W. Ha, Y.H. Park, S.M. Hudson, Biomacromolecules 4 (2003) 488–496. J. Magoshi, S. Nakamura, J. Appl. Polym. Sci. 19 (1975) 1013–1015. J. Magoshi, Y. Magoshi, S. Nakamura, N. Kasai, M. Kakudo, J. Polym. Sci., Part B: Polym. Phys. 15 (1977) 1675–1683. M. Tsukada, G. Freddi, N. Kasai, J. Polym. Sci., Part B: Polym. Phys. 32 (1994) 1175–1182.