An applicable electrospinning process for fabricating a mechanically improved nanofiber mat.

INTRODUCTION

One of the main challenges in tissue engineering and biomaterials is fabricating biodegradable scaffolds, in which cultured cells can reside and live. It is necessary to consider their ease of handling, good biocompatibility with cultured cells, biodegradability, easy shape-ability, and mechanical properties [1, 2]. Of these properties, the mechanical properties of scaffolds have an important role in providing mechanical support for cell growth and migration.

Polymeric scaffolds have been manufactured using various methods such as stereolithography or rapid prototyping process [3], fiber bonding (unwoven meshes), solvent casting or particular leaching, phase separation, high-pressure gas expansion, emulsion freeze-drying, and electrospinning [4, 5]. Of these methods, scaffold consisting of nanofibers fabricated using electrospinning has been a potential new biomedical material due to its large surface area and high porosity. To produce an artificial biomimetic scaffold that imitates the structure and biological functions of the natural extracellular matrix (ECM), nano-/microfibers fabricated using electrospinning should have the right size and structure for constructing scaffolds. According to Ma et al., the ECM consists of a basement membrane and a crosslinked structure of multifibril collagen, with diameters ranging from 15 [micro]m to 5 nm and interspersed with glycosaminoglycans [6].

However, the mechanical properties of electrospun nanofiber mats are still poor, and it is difficult to expose cultured cells to the correct stress environment to produce neotissues. To improve the deficiency of mechanical properties of electrospun fibers, several studies have researched various blending systems, controlling processing parameters, and postprocessing treatments [7-12].

The objective of this study was to fabricate mechanically improved PCL nanofiber mats. To achieve this goal, two mechanical systems were used: an auxiliary electrode stabilized the Taylor cone and initial spun jet solution at the nozzle tip, and a collector of the nanofibers using a computer-aided design (CAD) system with a moving x-y robotic stage. To determine the effects of the various auxiliary electrodes on the shape of the Taylor cone and the behavior of the initial spun jet solution, we calculated the electrical field concentration factor (EFCF) near the nozzle tip. Although the EFCF was well introduced from Kim's [13] article, there was still a need to study various processing parameters of the auxiliary electrode, such as geometries of conical auxiliary electrodes and positions between the nozzle tip and the electrodes. Furthermore, a process diagram showing the relationship between the flow rate and critical voltage ([V.sub.c]) needed to fabricate nanofibers was measured. The tensile properties of uniaxially aligned nanofiber mats, prepared while moving the collector system at different speeds, showed reasonable orthotropic properties resulting from the fiber orientation. Cellular behavior of human dermal fibroblasts (HDFs) within the oriented nanofiber web was investigated for future applications as scaffolds of skin tissue.

EXPERIMENTS

Materials

A biodegradable polymer solution was prepared by dissolving 2.4 g of poly([epsilon]-carprolactone) (PCL, [M.sub.w] = 80,000) in 30 g of a solvent mixture composed of methyl-enechloride (MC) and dimethylformamide (DMF). The solvents were used in MC/DMF wt% ratios of 80/20. PCL solution was prepared at a fixed concentration of 8 wt% in the solvent, and the electrical conductivity of the solution was 51 [micro]S/m. The polymer solution was placed in a 20-ml glass syringe with a G-20 needle (size = 0.9 x 20 [mm.sup.2]). The flow rate of the polymer solution was controlled using a syringe pump (KDS 230: KD Scientific, Holliston, MA).

Electrospinning Setup

The fundamental principle of electrospinning is that a Taylor cone is formed by applying an electrical field to the polymer solution hanging from a syringe tip, and jets of electrically charged solution are emitted when the applied electrostatic force exceeds the surface tension of the solution. These jets erupt from the apex of the cone at the nozzle and travel toward an electrically grounded target on which they are stacked as a nonwoven mat.

In this article, PCL nanofibers were obtained using an electrospinning technique that employed various auxiliary electrodes composed of thin copper film (thickness = 0.2 mm) connected to the nozzle tip. The conical electrodes and nozzle were connected with a copper wire, and an equal voltage of 15-20 kV could be applied to both as the fluid passed through a nozzle. A detailed setup of this electrospinning is shown in Fig. 1 and the dimensions of the auxiliary electrodes are described in Table 1. A high-voltage power supply (SHV300RD-50K; Convertech, Wharton, NJ) was used to apply a high electrical field.

[FIGURE 1 OMITTED]

The Taylor cone and initial spun solution were photographed using a digital camera (E-300; Olympus, Tokyo, Japan). The spun fibers were deposited on poly(ethylene terephthalate) (PET) film (thickness = 120 [micro]m, surface resistance = [10.sup.15][ohm]) attached to the moving stage, to which a rectangular piece of copper film was fixed as a counter electrode.

Structural Morphology and Mechanical Testing of Oriented PCL Scaffolds

The morphology of the electrospun PCL mats was observed under an optical microscope (BX FM-32; Olympus) connected to a digital camera. The tensile properties of the nanofiber mats were characterized using a universal tensile machine (MTS; Tytronics, Lowell, MA) at a cross-head speed of 10 mm/min.

Cell Culturing on the Electrospun Mats

The electrospun nanofiber webs were sterilized with 70% ethanol and exposed to UV light for 1 h. They were prewarmed with Hank's balanced salt solution (HBSS) for 2 h at 50 C. HDFs were cultured in Dulbecco's Modified Eagle Medium (DMEM) supplemented with 10% fetal bovine serum (FBS) and maintained up to passage 7. Isolated fibroblasts were seeded on each sample (1 x 1 [cm.sup.2]) at a density of 2 x [10.sup.2] cells and cultured for up to 3 days at 37[degrees]C and 5% C[O.sub.2]. The cells were fixed with 2.5% glutaraldehyde for 1 h at room temperature and dehydrated through a series of ethanol dilutions. The samples were sputter-coated with Pt. Cell morphology and growth was observed using a scanning electron microscopy.

RESULTS

Figure 2 shows the EFCF, which was defined by Kim [13]. The parameter describes the degree of convergence of spun jets to the spinning line and the factor defined as [E.sub.r]/||E||, where [E.sub.r] is the r-directional component of the electrical field in a cylindrical coordinator. The normalized r-directional value of electric field can vary from -1 to +1, and the positive sign means that an electric field at the calculated position tends toward outside areas, while negative means a convergence to the spinning axis.

The EFCF inferred from the shapes of the auxiliary electrodes used in this electrospinning process was analyzed using the commercial software ANSYS/Emag, with the finite element method for the electromagnetic analysis. The numerical models consisted of two-dimensional plane121 elements and 8-node charge-based electric elements. In Fig. 2, the EFCFs were plotted for a normal electrospinning and the modified process using various auxiliary electrodes (T-1, T-2, T-3, T-4, and T-5).

As shown in the Fig. 2a and b, the various auxiliary electrodes altered the distribution of the electrical field near the nozzle tip. In the figure, the x-axis is the normalized distance between the nozzle tip and target electrode. In Fig. 2a, the EFCF calculated using various shapes of auxiliary electrodes near the nozzle tip showed low EFCF values relative to the normal process (w/o aux. electrode). However, as shown in the table in Fig. 2, for the processes using the auxiliary electrodes, the electrical field strength (E) at the nozzle tip decreased steeply relative to the normal process.

Figure 2b shows the EFCFs for three different relative positions of the nozzle tip within the same conical shape electrode. The figure shows that the field-interference caused by the T-4 electrode was high near the nozzle, but the electric field strength near the nozzle tip using the T-4 electrode, driving electrostatic force applied in the nozzle tip used to generate nanofibers decreased markedly relative to the T-1 and T-5 electrodes. From this calculated results, we can estimate that although the T-4 electrode provides electric field distribution to enable stable initial spun jets in the electrospinning process, high critical voltage to generate nanofibres can be required, while the T-5 electrode supplies satisfactory EFCF values and relative low critical voltage compared with the T-4 electrode. The estimation will be assessed experimentally in the next section.

Figure 3 shows a shape angle effect of the auxiliary electrode. To observe the effect, we used the two different types of the auxiliary electrodes (T-1 and T-3) and two cylindrical electrodes having different diameters (d = 11 and 33 mm). As shown in Fig. 3a, the EFCF of the T-1 electrode has a different value compared with that of the cylindrical electrode with the same diameter (d = 11 mm) near the nozzle tip. However, in the reverse case (Fig. 3b), the values of EFCF of the T-3 electrode was the same as that of the cylindrical electrode (d = 33 mm). From these calculated results, we can estimate that the angle of the converging conical electrode can affect the electric field distribution near the nozzle.

[FIGURE 3 OMITTED]

To observe the effect of the auxiliary electrodes on the shape of the Taylor cone and initial solution, we photographed the nozzle tip with various auxiliary electrodes. Figure 4 shows how different auxiliary electrodes under the same applied voltage (15 kV) influenced the shape of the Taylor cone and the direction of the initial solution jet. This occurred because the electrodes affected the field strength and field distribution near the nozzle tip, as predicted in the simulations. As shown in Fig. 4a, in the normal case, the direction of the initial spun solution relative to the counter electrode was unstable because of environmental conditions, such as the airflow and proximity to scattered or charged nanofibers. In contrast, with the supplementary electrodes (Fig. 4b), the initial direction of the solution was very stable and the jet toward the counter electrode was straight, despite the influence of environmental conditions. Figure 4c described various shapes of Taylor cone under various auxiliary electrodes. As the calculated results, the size of Taylor cone was larger in the diverging conical electrode (T-3 electrode), which had low electric field strength near the nozzle tip compared to those of T-1 and T-2 electrodes.

[FIGURE 4 OMITTED]

[FIGURE 5 OMITTED]

Figure 5 shows an operating diagram delineating the areas of different jet behavior as a function of the applied field strength and flow rate. The figures display two regions: a mixture of nanofibers and droplets, and nanofibers. At low voltages, dripping was observed, but as the electrical field was increased, nanofibers were obtained after passing through the region consisting of a mixture of droplets and nanofibers. In Fig. 5a and b, the result showed that the critical voltage ([V.sub.c]) of normal electro-spinning was relatively low compared with the processes using auxiliary electrodes. This is because of the difference in the concentrated electric field strength of the tip between the normal electrospinning and modified processes using auxiliary electrodes, as calculated electric field strength. From the calculated and experimental results, we can choose T-5 electrode as the most appropriate electrode to fabricate nanofibres because of the stability of initial spun jets and low critical voltage.

[FIGURE 6 OMITTED]

The morphology of the nanofibers electrospun on a PET film was examined under optical microscopy (Fig. 6a and b). The PCL solution was electrospun with and without a conical electrode (T-5) and the same voltage (15.4 kV) was applied during both processes. The average nanofiber was 0.91 [+ or -] 0.14 [micro]m for normal electrospinning (Fig. 6a) and 0.95 [+ or -] 0.2 [micro]m with the T-5 auxiliary electrode (Fig. 6b). The diameters of the electrospun nanofibers were similar with and without the auxiliary electrode. This indicates that although a strong electrical field was applied to the nozzle tip, the applied voltage did not significantly affect the diameter of electrospun fibers. According to Ramakrishuna and coworkers [14], as the concentration of polymer solution increased, the diameter of electrospun nanofibres also increased because of the coupled parameters such as electrical conductivity, viscosity, and surface tension of the solution. While increasing the applied voltage on high polymer concentration, the diameter of fibers slightly changed, but while lowering the polymer concentration, the influence of applied voltage was negligible to change of diameter of electrospun fibers. In addition, polymer concentration, molecular weight, and electrical conductivity of solvents played an important role in controlling the morphology of the electrospun nanofibres, while the voltage and feeding rate were of less effect compared to those parameters [14].

[FIGURE 7 OMITTED]

[FIGURE 8 OMITTED]

Figure 7 shows the selectively deposited regions and micrograph images of the electrospun fibers processed using the auxiliary electrode (T-5) and moving x-y collector. The fibers were deposited on the PET film over 2 h under an electrical field of 15.3 kV using a conical electrode (T-5). Figure 7a shows the nanofibers deposited randomly on the PET film using the fixed x-y stage, and the magnified picture shows electrospun nanofibers. Figure 7b is a region of linear deposition obtained using the stage that was swung in the y-direction at a speed of 25 cm/sec. Figure 7c shows a cross section of the electrospun fibers deposited using the stage that was swung first in the x-direction and then in the y-direction at the same speed. The deposited mat was 126 [micro]m thick and average diameter of the deposited fibers was 763 [+ or -] 96 nm. Figure 7d showed a round shape deposited area using the moving target with a speed of 15 cm/sec. The concentrated deposited shapes showed that the electrospun nanofibers could selectively coat on a substrate and it might be possible to fabricate user-defined three-dimensional scaffolds, which may possess controlled mechanical properties due to their fiber orientation.

To evaluate the mechanical properties of nanofibers deposited in the y-direction shown in Fig. 7b, we measured the tensile properties of specimens parallel and perpendicular to the y-direction of the stage, moving at two different speeds (25 and 35 cm/sec). The deposition time was at least 2 h to ensure sufficient thickness and thickness uniformity of the stacked nanofiber mat. The tensile behavior of the electrospun mat was measured using a Tytron (MTS) at a crosshead speed of 10 mm/min.

According to Rutledge and coworkers, the mechanical properties of an electrospun nanofiber web are closely related to the fiber orientation within the web, bonding between fibers, and slip of one fiber over another, rather than the mechanical properties of individual fibers [7]. However, if we assume that the slippage between fibers and fiber bonding is uniform over the nanofiber web, it is reasonable to assume that the tensile properties are related only to the fiber orientation within the mat.

The stress-strain curve for a nanofiber mat fabricated using the x-y moving system at 35 cm/sec is shown in Fig. 8. In the figure, "||" and "[perpendicular to]" indicate that the specimen is parallel to the y-direction of the stage and perpendicular to the moving direction, respectively. The specimens are shown in Fig. 8. Using the tensile test, we determined that the manufactured mat had orthotropic properties, which we believe resulted from the orientation of the nanofibers.

Additional tensile properties of the nanofiber mat, such as the maximum strength, percentage of breaking strain, and Young's modulus, are shown in Table 2. In the table, [S.sub.1] and [S.sub.2] denote movement velocities of 25 and 35 cm/sec, respectively. Although we did not see significant orientation of the PCL fibers under the microscope, the tensile properties indicate that the electrospun fiber mats made using the moving system were oriented in the direction of stage movement. This is more convincing for the greater movement velocity.

To observe the effect of the designed nanofiber mat to the cell proliferation, the HDFs cells were embedded in the both randomly dispersed nanofiber mat and the aligned nanofiber mat. Shown in the Fig. 9, the cells were well spread on the nanofibers matrixes and showed interactions varying in the alignment of nanofibers in the webs. When the Fig. 9a and b were compared, the linearly oriented nanofibers in the mat to the moving direction of the x-y stage provided the good alignment of the cells, while randomly dispersed proliferation of the cells for a normally spun fiber mat.

[FIGURE 9 OMITTED]

CONCLUSIONS

To obtain design-based nanofiber mats, we used an electrospinning method with an auxiliary electrode to stabilize an initial spun jet solution at a nozzle tip and x-y moving collector system, which could be controlled by various predetermined computer aided designs. To determine the most appropriate auxiliary electrode, we calculated EFCFs for several kinds of electrodes having various shapes and dimensions. From the predicted and experimental results, we can choose a converging conical electrode (T-5 electrode) in terms of EFCF and critical voltage to be required to generate nanofibres. Using the electrospinning process with the T-5 electrode and x-y collector system, various shapes of nanofiber mats were fabricated, and the deposited nanofiber mats showed mechanically orthotropic tensile properties attributable to the fiber orientation and also caused the cell alignment to the direction of the fibers alignment. This simple system may have potential for fabricating design-based polymer scaffolds directed by a predetermined architecture.

ACKNOWLEDGMENT

The help from Dr. S.A. Park for cell culturing is greatly appreciated.

REFERENCES

1. W.J. Li, C.T. Laurencin, E.J. Caterson, R.S. Tuan, and F.K. Ko, J. Biomed. Mater. Res., 60, 613 (2002).

2. D.W. Hutmacher, J. Biomater. Sci. Polym. Edn., 12, 107 (2001).

3. D.W. Hutmacher, M. Sittinger, and M.V. Risbud, Trends Biotechnol., 22, 354 (2004).

4. M.E. Gomes, J.S. Godinho, D. Tchalamov, A.M. Cunha, and R.L. Reis, Mater. Sci. Eng. C. 20, 19 (2002).

5. J.F. Mano, C.M. Vaz, S.C. Mendes, R.L. Reis, and A.M. Cunha, J. Mater. Sci. Mater. Med., 10, 857 (1999).

6. Z. Ma, M. Kotaki, R. Inai, and S. Ramakrishna. Tissue Engineering. 11, 101 (2005).

7. M. Wang, H.J. Jin, D.L. Kaplan, and G.C. Rutledge. Macro-molecules, 37, 6856 (2004).

8. J. Doshi and D.H. Reneker, J. Electrostat., 35. 151 (1995).

9. D.H. Reneker and I. Chun, Nanotechnology, 7, 216 (1996).

10. A.L. Yarin, S. Koombhongse, and D.H. Reneker. J. Appl. Phys., 89, 3018 (2001).

11. M.M. Bergshoef and G.J. Vancso. Adv. Mater., 11, 1362 (1999).

12. S.V. Fridrikh, J.H. Yu, M.P. Brenner, and G.C. Rutledge, Phys. Rev. Lett., 90, 144502 (2003).

13. G.H. Kim, J. Polym. Sci. B, 44, 1426 (2006).

14. S.-H. Tan, R. Inai, M. Kotaki, and S. Ramakrishuna. Polymer, 46, 6128 (2005).

Geun Hyung Kim, Houkseop Han, Jong Ha Park, Wan Doo Kim

Department of Future Technology, Korea Institute of Machinery and Materials, Yuseong-gu, Daejeon 305-343, Korea

Correspondence to: Geun Hyung Kim; e-mail: [e-mail]

Contract grant sponsor: Korea Institute Machinery and Materials.

TABLE 1. Dimensions of various auxiliary electrodes.

      Dimension of auxiliary
      electrode (mm)          Length of
Type  a   b   c               a nozzle (mm)

T-1   22  11  20              20
T-2   22  22  20              20
T-3   22  33  20              20
T-4   22  11  20              16
T-5   22  11  20              24

                         No aux. electrode  T-1   T-2   T-3   T-4   T-5

[E.sub.at a tip](kV/mm)  3.6                0.93  0.90  0.87  0.29  1.12

FIG. 2. (a) the EFCF at a normalized spinning line from the nozzle tip
to ground for the various electrodes (w/o auxiliary electrode, T-1, T-2,
and T-3) and (b) the EFCF at a normalized spinning line from the nozzle
tip to ground for the various electrodes (T-4, T-1, and T-5). In table,
the maximum strengths of the electrical field at the nozzle tip for
various electrodes were shown.

TABLE 2. Tensile properties of nanofiber PCL mats obtained using the
electrospinning process with a T-5 electrode.

Sample                         Young's            Ultimate tensile
(n = 6)                        modulus (MPa)      strength (MPa)

[S.sub.1(||)]                  17.3 [+ or -] 8.1  2.54 [+ or -] 0.5
[S.sub.1([perpendicular to])]  13.7 [+ or -] 4.0  1.91 [+ or -] 0.3
[S.sub.2(||)]                  19.8 [+ or -] 3.7  2.63 [+ or -] 0.3
[S.sub.2([perpendicular to])]  13.5 [+ or -] 3.3  1.67 [+ or -] 0.5

Sample                         Break
(n = 6)                        strain (%)

[S.sub.1(||)]                  59.6 [+ or -] 12
[S.sub.1([perpendicular to])]  52.7 [+ or -] 5
[S.sub.2(||)]                  83.2 [+ or -] 15
[S.sub.2([perpendicular to])]  38.8 [+ or -] 7