3.2. Rheological Behavior of Solutions
Rheological curves of SF and S2E8Y solutions with different concentrations are shown in Figure 2
. Flow curves of SF and S2E8Y solutions exhibited the same trends in the range of concentrations investigated. Viscosity-dependent shear rate showed greater shear thinning at lower shear rate, Newtonian flow at intermediate shear rate, and lower shear thinning at higher shear rates. In the lower shear rate region, flow curves of SF solution showed greater shear thinning due to the destruction of weak intermolecular interactions in SF solution in the static state [34
]. In the intermediate shear rate range, flow curves of SF solution showed a plateau viscosity region, termed the zero shear viscosity (η0
), which suggested the formation of replacement interaction in a dynamic equilibrium [35
]. As the concentration of SF increased, a shear thinning region emerged at higher shear rate, which was similar to pseudoplastic flow behavior of polymer solutions. These results implied that the shear-sensitivity of the solutions was enhanced by the increase in SF concentration, and shear thinning behavior was induced by the rearrangement and orientation of macromolecular chains along the shear direction [36
]. Compared to the zero shear viscosity (about 0.2 Pa·s) of S2E8Y solutions, the zero shear viscosity of SF solutions (about 2–20 Pa·s) was higher than S2E8Y by one to two orders of magnitude. In general, the viscosity of SF and S2E8Y solutions increased with concentration, and compared to the viscosity of S2E8Y solutions, the viscosity of SF solution increased more significantly when the solution concentration was increased.
3.3. Effect of Processing on Electrospun SF-SELP Fibers
In most coaxial electrospinning, the shell polymer solution has sufficient viscoelasticity and good electrospinnability, and the core solution can have either poor or good electrospinnability [37
]. In the present study in contrast to these prior studies, the SELP solutions with a relatively low viscosity were chosen as the shell, and the SF solutions with a relative high viscosity were chosen as the core. The process parameters such as concentration, flow rate, distance between the needle tip and collector, and applied voltage were studied in order to ensure a stable state cone-jet for the formation of homogeneous core-shell fibers.
The effect of SF concentrations on nanofiber morphology is shown in Figure 3
. Other process parameters and fiber diameters are listed in Table 1
. During electrospinning, the concentration of SF was varied from 26% to 32%, while the concentration of SELP solution was kept constant at 15 wt %. The 26% to 32% range was selected because the viscosity of SF solutions with concentrations lower than 26 wt % were too low to be electrospun and SF solutions with concentrations higher than 32% were not easy to electrospun due to the high viscosity. Other process parameters were kept the same for direct comparisons: the inner flow rate was 0.4 mL/h; the outer flow rate was 0.17 mL/h; the applied voltage was 20 kV; the tip to collector distance was 12 cm. Core solution at concentrations of 26 wt % and 32 wt % resulted in fiber mats with tiny droplets and beads on strings (Figure 3
a,c). When the concentration of SF was 29 wt %, consistent fibers without beads were obtained, and the average diameter was about 301 ± 108 nm (Figure 3
b). These results indicated that the concentration of the core solution, SF solution, is critical for the coaxial electrospinning. If the concentration was too low, e.g., 26 wt %, the viscosity of the solution was low (about 2.0 Pa·s), so that the core solution was not viscose enough to serve as a template for the shell solution to form the steady state of cone-jet. However, if the concentration was too high, e.g., 32 wt %, the applied electricity was not adequate to overcome the surface tension of Taylor cone and split the cone-jet flow into thin fibers due to the high viscosity (about 20 Pa·s) of the core SF solution [39
The effect of SELP concentration on the nanofiber morphology is shown in Figure 4
. Other operating parameters are listed in Table 1
. During electrospinning, the concentration of the SELP solution was increased from 12 to 15 wt %, while the concentration of the SF solution was kept constant at 29 wt %. When the SELP concentration was below 12 wt %, no continuous core-shell fiber was obtained, so the SELP concentration was selected between 12% and 15% to explore the lower concentration limits. For 12 wt % SELP solution, tiny droplets and beaded fibers formed in the fibers and the fiber average diameter was about 373 ± 129 nm (Figure 4
a). For the 13.5 wt % SELP solution, beads were still visible but fewer in numbers and the fibers average diameter was about 406 ± 151 nm. When the concentration of SELP was increased to 15 wt %, uniform nanofibers with an average diameter of 361 ± 98 nm without beads were obtained. It indicated that increasing the concentration of SELP in solution promoted the formation of core-shell structured fibers, due to the improved viscosity and electrospinnability of SELP solution.
In summary, the SF concentration had a major impact on fiber morphology in the coaxial electrospining process. The SF solution with appropriate viscosity was able to serve as a spinning aid to successfully prepare nanofibers from low viscosity SELP solutions.
In order to generate a good “compound Taylor cone” with a core-shell structure, controlled and balanced injection speeds of the inner and outer fluids are critical to keep the “compound Taylor cone” in dynamic stabilization [40
]. In the present study, 29 wt % electrospinnable SF solution was chosen as the core solution, which had shown good electrospinnable features based on Table 1
, and 12 wt % nonelectrospinnable S2E8Y solution was chosen as the shell solution to demonstrate that the electrospinnable core solution played a critical role as a spinning aid to electrospin the nonelectrospinnable shell solution. The influence of the flow rate and applied voltage on the morphology and diameter of the core-shell fibers were also investigated. The process parameters and resultant fiber diameters are listed in Table 2
The influence of the core SF flow rate on the morphology and diameter of the fibers is shown on the top-line of Figure 5
. For the 0.4 mL/h (Figure 5
(A1)) and the 0.8 mL/h (Figure 5
(A3)) conditions, the collected fiber mats had droplets and breads-on-strings. With the core SF flow rate was increased from 0.4 to 0.6 then to 0.8 mL/h, the average diameter of the nanofibers increased from 370 ± 121, to 408 ± 150 then to 474 ± 162 nm, respectively. These data show that the effect of the core solution flow rate in coaxial electrospinning is similar to that for single electrospinning. An increase in solution flow rate generally resulted in larger fiber diameters [41
The middle line of Figure 5
shows the influence of the shell SELP flow rate on the morphology and diameter of fibers. When the shell SELP flow rate increased from 0.07 (Figure 5
(B1)) to 0.17 mL/h (Figure 5
(B2)), the average diameter of the nanofibers decreased from 712 ± 215 nm to 408 ± 121 nm because the shell solution provided solvent to “wet” the high viscosity core solution that resulted in thinner fibers. However, the morphology of the fiber mats was changed to larger droplets and beaded fibers when the flow rate of the SELP solution was 0.34 mL/h (Figure 5
(B3)). This phenomenon can be attributed to the oversupply of the SELP shell solution.
The last line of Figure 5
shows the effect of applied voltage on the morphology and diameter of fibers. There were no breads on fibers at 20 kV (Figure 5
C1). With the applied voltage increased from 25 (Figure 5
(C2)) to 30 kV (Figure 5
(C3)), the amount of beads and droplets increased. This is because the strength of the electric field exceeded that required for the material, so that the lower viscosity shell solution could not follow the core fluid path and disassociated into separate paths [22
The results suggested that in this coaxial electrospinning process, the SF solution served as a spinning aid to successfully prepare nanofibers from nearly nonelectrospinnable SELP solution. The core SF solution carried the shell SELP solution through the formation of a stable Taylor cone and continuous jet ejection during the process. The optimum conditions to electrospin uniform SF-SELP nanofibers with different processes parameters are shown phase diagram in Figure 6
. In addition, a combination of the following parameters was also critical. (a) Using the same solvent, water, led to low interfacial tension between the two solutions and maintained the steady and continuous jet by “wetting” the high viscosity core solutions [42
]; (b) The higher conductivity of the shell layer helped the high shear stress to be applied on the core material, stabilized the coelectrospinning process and subsequent elongational force, and resulted in a thinner core; (c) Low vapor pressure of the solvent (boiling point of water = 100 °C) facilitated the spinning process. High vapor pressure solvents can produce unstable Taylor cones in coaxial electrospinning [43
3.6. Mechanical Testing of SF and SF-SELP Fibers Mats Before and After Methanol Treatment
The tensile stress-strain curves of SF fibers mats and SF-SELP core-shell fiber mats before and after methanol treatment are shown in Figure 9
. The average thickness, Young’s modulus, tensile strength and elongation at break are summarized in Table 3
. Before methanol treatment, SF mats (Figure 9
curve a) had lower Young’s modulus and tensile strength which were 1.02 ± 0.12 MPa and 1.00 ± 0.22 MPa, respectively, meanwhile, the mats were brittle and easy to fracture with an elongation at break at only 1.05% ± 0.15%. In contrast, the SF-SELP mats (Figure 9
curve c) exhibited slightly higher Young’s modulus and tensile strength 1.30 ± 0.35 MPa and 1.72 ± 0.24 MPa, and good flexibility with elongation at break of 4.70% ± 0.31%. This difference may be attributed to the smaller diameters of SF-SELP core-shell fibers compared with the SF fibers (data not shown), and the elasticity of the SELP shell phase. After methanol treatment, the thickness of both specimens decreased and the elongation at break slightly increased due to the β-sheet crystal structure in SF and the silk domain in the SELPs, which caused contraction and higher packing density of the fiber mats. On the other hand, although tensile strength of SF and SF-SELP mats improved to 2.15 ± 0.32 MPa and 4.81 ± 0.35 MPa, respectively, the Young’s modulus of the SF-SELP core-shell mats was almost the same, and the elongation at break was 5.20% ± 0.57%. These results suggested that the SELPs reduce the brittleness of SF mats whose elongation at break was only1.38% ± 0.22%.