Alginate-nanofibers fabricated by an electrohydrodynamic process.

INTRODUCTION

Electrospinning is a simple and widely used technique for producing micrometer to nanometer-sized fibers of various polymers. Electrospinning relies on electric charges to form ultrafine fibers from conical droplets of polymer solution ejected from a nozzle tip (1), (2). Nanometer-sized fibers have the potential for a range of highly useful applications, such as conductive polymeric biosensors, filter membranes, biomedical scaffolds, wound dressing materials, artificial organs, nanoelectronics, nanocomposites, and chemically protective clothing (3). In particular, nanofibers produced by electrospinning show promise for the production of polymeric scaffolds that mimic the structure and biological functions of the naturally occurring extracellular matrix (ECM).

Even though there are well-established manufacturing methods for fabricating nano-sized fibers, a commercially available mass production system with a high production rate is another issue (4). There has been some research into how to achieve high Throughput industrial production by stabilizing the spun jets in single-nozzle and multiple-nozzle systems without interrupting environmental conditions (4), (5). Zussman and coworkers (4) demonstrated experimentally and numerically that jets from multiple nozzles show higher repulsion to other neighboring jets due to Columbic forces than jets spun by a single-nozzle process. For a single-nozzle process, Deitzel et al. (5) studied the possibilities of decreasing the whipping instability caused by charged jets. That research focused on how to dampen the instability of spun fibers and control the deposited area of submicron poly(ethylene oxide) (PEO) nanofibers using a substrate with an electrostatic lens element (5). The results indicated an approach to controlling or even eliminating the bending instability inherent in conventional single- and multiple-nozzle electrospinning processes.

Dosunmu et al. (6) demonstrated an electrospinning process using multiple polymer jets projecting onto a porous tubular surface. Fiber production from multiple jets was compared with fiber production from a single-syringe nozzle jet. Fibers deposited on the porous tube from multiple jets had a significantly greater production rate than those from the single-nozzle jet. Yarin and Zussman (7) introduced needleless electrospinning of multiple nanofibers using a normal magnetic field to eliminate clogging of the nozzles during multiple-jet spinning. Multiple spinnerets increased the production rate and offered the potential for electrospinning bicomponent and multicomponent nanofibers. To obtain blended nanofibers of uniform thickness, a multiple-jet electrospinning device was manufactured with a rotating grounded target collector.

To predict the stability of electrospun jets fabricated with multiple nozzles, Kim et al. (8) introduced the electric field concentration factor (EFCF), defined as the jets' degree of convergence to a spinning axis. The EFCF parameter is used to compare the experimental results for single- and multiple-nozzle electrospinning processes. Stability analysis of electrospinning has demonstrated that by using a cylindrical electrode connected to multiple nozzles, the initial stream line and jets of nanofibers from the nozzles could have stable jet motion without interrupting charged jets nearby, changing the environmental conditions such as airflow or interfering with nearby dielectric or conductive materials.

Natural polymers used as biomedical scaffolds have the advantage of being very similar to macromolecular substances that the biological environment" can recognize and deal with metabolically, whereas synthetic polymers may cause problems due to toxicity and lack of recognition by cells. A potential problem with natural polymers used as biomaterials is their processability characteristics. If the material is to be used as the ECM of connective tissues such as tendons, ligaments, skin, blood vessels, or bone, the materials must be processed into a fiber shape. However, according to Lu et al. (9), biopolymers such as alginate, collagen, chitosan, silk, and eggshell membrane are extremely difficult to fabricate in micro/nanofiber form using an electrospinning process. To overcome these problems, there have been some efforts to blend biopolymers with biocompatible synthetic polymers, which may enhance processability (9-11). Although this approach does increase the processability for electrospinning, a new approach is required to obtain a high production rate for a short period of time. We have applied a multiple-nozzle electrospinning system supplemented with auxiliary cylindrical electrodes to achieve high productivity and stable electrospinnability of the material system.

In this article, we used a multiple-nozzle electrospinning system assisted with supporting electrodes to improve the processability of alginate, a natural biomaterial that is difficult to form into nano-sized fibers. Alginate has distinctive properties such as nontoxicity, biocompatibility, biodegradability, and hydrophilicity. It is widely used in biomedical applications such as wound dressings, tissue engineering scaffolds, and drug delivery carriers (12), (13). However, fabricating alginate-nanotibers in a general electrospinning process is difficult because alginate solution tends to congeal at very low polymer concentrations (10).

To enhance its electrospinnability, we mixed alginate with PEO and lecithin as a processing agent. We used a multiple-nozzle electrospinning system connected with auxiliary electrodes to improve the production rate of micro/nanofibers. To observe the effect of the electrodes in the multiple nozzle system, we measured the production rate and fiber size uniformity with and without the electrodes. The electrospun alginate-fiber webs we produced showed potential for use as a biomedical scaffold.

EXPERIMENTAL

Materials

Sodium alginate (SA; made up of [alpha]-(1 [right arrow]4)-L-guluronic acid (G) and [beta]-(1 [right arrow] 4)-D-mannuronic acid (M) plus a natural polysaccharide obtained from marine brown algae), PEO ([M.sub.w] = 9,000,000) and calcium chloride were purchased from Sigma-Aldrich (St. Louis, MO). The viscosity of the SA was medium. Lecithin was supplied by Doosan Biotech (Korea). Several SA/PEO solutions with different concentration ratios were prepared by dissolving SA and PEO in distilled water. The concentration ratios of SA/PEO were 1/1, 2/1, 3/1. 2/1, 2/2, and 3/2 wt%. We used the notation SaPa where "a" and "b" are the weight percent of SA and PEO, respectively. For example, S1P1 means a solution with SA 1 wt% and PEO 1 wt%. The SA/PEO mixtures were stirred at room temperature overnight to obtain homogeneous solutions.

Characterization

The viscosity of the SA/PEO solutions was measured at room temperature using a viscometer (LVDVE 230, Brookfield Engineering Laboratories, MA) equipped with an SC4-31 spindle and 13R chamber. We used a pH/conductivity meter (Orion 4 Star, Thermo Scientific, MA) to measure the conductivity of solutions. The fiber morphology of the alginate nanofiber was observed with a scanning electron microscope (SEM) (Nova nanoSEM 200, FEI, Netherlands). The average fiber diameter and diameter distribution were obtained by analyzing randomly selected fibers from SEM images with a custom code image analysis program (Scope Eye II, TDI, Korea).

Alginate nanofibers were cross-linked by soaking them in 5 wt% (w/v) [CaCl.sub.2] solution for the cell test and rinsing them with deionized water to remove the excess [CaCl.sub.2] (9), (14), (15). The nanofiber scaffolds were sterilized with 70% ethanol and prewarmed with Hank's balanced salt solution. Human dermal fibroblasts (HDFs) were cultured in Dulbecco's Modified Eagle's Medium supplemented with 10% fetal bovine serum. The HDFs were seeded on the 1 X 1 cm nanofiber at a density of 6.4 X [10.sup.4] cells/ nanofiber. HDFs on the nanofiber were cultured for up to 5 d at 37[degrees]C in an atmosphere of 5% [CO.sub.2]. The cells were fixed with 2.5% glutaraldehyde for 1 h and dehydrated through a series of ethanol dilutions. The HDFs on the nanofiber were sputter coated with Pt, and the morphology of the cell attachment was observed using a SEM 3 and 5 d after seeding.

Multiple-Nozzle Electrospinning System With an Auxiliary Electrode

In the electrospinning process, a high electric potential is applied to a droplet of the blended solution at the tip of a syringe needle. Then nanofiber is formed by the electrical repulsive force between positive charges in the droplet on the needle tip. The multiple-nozzle electrospinning system consisted of three nozzles (syringe needles), and each nozzle was attached to a conical auxiliary electrode to stabilize the Taylor cone and initial spun solution, as shown in Fig. 1. To minimize the interference of the electric field distribution of the three nozzles, the nozzles were placed 120 mm apart in a triangular shape. We used several different combinations of distance from needle to collector (15, 20, and 25 cm), voltage (0-40 kV DC), and flow rate (0.2, 0.5, 1.0, and 1.5 ml/h) to evaluate the effect of these parameters. All experiments for the electrospinning system were carried out at 30[degrees]C using S2P2 solution with 0.7 wt% lecithin added.

[FIGURE 1 OMITTED]

RESULTS AND DISCUSSION

Effect of Lecithin on the Electrospinnability of SA/PEO

In general, electrospinning is strongly dependent on the properties of the solution. Figure 2 shows the solution viscosity and conductivity as functions of the blend composition. As depicted in Fig. 2, the viscosity and conductivity increase as the content of SA in 2 wt% PEO increases. As Bhattarai et al. (10) observed, it is extremely difficult to create nanofibrous structures by electrospinning alginate solution due to its high viscosity. To overcome this processability problem, they added Triton X-100 as a surfactant in an alginate/PEO solution to control sol-gel transition, and this interacted with the alginate solution to reduce the solution's viscosity. The electrospun alginate nanofibers fabricated in this manner were suggested for use as biomedical scaffolds, however, the presence of the Triton X-100 surfactant may cause cell damage during the cell culturing process. Esquisabel et al. (16) used lecithin as a surfactant to prepare alginate-(poly-L-lysine) (PLL) microcapsules, the size of which was heavily dependent on the amount and type of lecithin. This indicates that lecithin could play an effective role as a surfactant to control the size of alginate-PLL microcapsules. On the basis of these results, we used lecithin as a surfactant in our SA/PEO electrospinning system.

[FIGURE 2 OMITTED]

To determine a suitable composition of PEO, various ratios of SA/PEO solution were electrospun with lecithin. Figure 3 shows the morphology of the electrospun webs produced for various SA/PEO ratios. A PEO content of 1 wt% with various SA concentrations produced beads and beaded fiber. However, electrospinning a solution with 2 wt% PEO successfully produced nanofibers. This shows that the composition of the PEO solution plays an important role in forming beaded fibers and nanofibers. The best composition of PEO in our SA/PEO blend system was a composition ratio of S2P2. Moreover, the addition of 0.7 wt% lecithin as a surfactant decreased the spattering of alginate droplets on the target, as shown in Fig. 4. We used S2P2 with 0.7 wt% lecithin as an appropriate composition in the multiple nozzle system to attain a high nanofiber production rate.

[FIGURE 3 OMITTED]

[FIGURE 4 OMITTED]

Effect of an Auxiliary Electrode

As other research has indicated, by controlling the shape and strength of the macroscopic electric field between a spinneret and a grounded target, we should be able to control the electrospinning process through basic electrostatic principles (5), (8). We connected various auxiliary electrodes (specially designed conical-type electrodes) to the syringe nozzles (17) to stabilize the initial spun jet solution and the Taylor cone at a nozzle tip, which could be important for attaining a high production rate. Using a conical-type auxiliary electrode resulted in a stable initial spun jet without sacrificing the high voltage drop between the nozzle and the target electrode. In this work, we used a conical-type electrode for each nozzle, as shown in Fig. 1, to obtain stable processability at each nozzle. Figure 5a shows the contours of the EFCF near the nozzle with and without an auxiliary electrode, both on the same scale. The EFCF, which was defined as the degree of convergence of an initially spun solution to a spinning axis, has been calculated as [E.sub.r]/||E||, where [E.sub.r] is the r-directional component of the electric field in the cylindrical coordinates and E is electrical field (8). The factor can vary across the range [+ or -]1 where a positive sign means a divergence of the electric field at the calculated position from the spinning axis and a negative sign means a convergence of the electric field. From Kim et al. (8), the meniscus of the Taylor cone at a nozzle tip can be influenced by the shape of ellipsoidal contour of the electric field near the nozzle tip. Calculated results show that the contour of EFCF for an auxiliary electrode exhibits a broad region at the nozzle tip relative to a normal nozzle, and this could lead to stabilizing the initial spun jets at a the nozzle tip, although interference from the other nearby charged jets and unsteady processing were present.

[FIGURE 5 OMITTED]

As shown in Fig. 5b, the electrospun jet with the auxiliary electrode was stable and went straight to the collector, whereas it was unstable and easily diverted for the normal nozzle. This indicates that the jet of the nozzle with the auxiliary electrode was concentrated on the collector during electrospinning because the Taylor cone with the auxiliary electrode was more stable than the normal nozzle. Therefore, we expect that it should be possible to increase total production through the use of an auxiliary electrode.

Processability of Multiple Nozzles

To find suitable processing conditions for the multiple nozzle system with or without auxiliary electrodes, we investigated the flow rate and DC voltage with respect to the morphology of the electrospun material. As shown in Fig. 6, there were three regions related to beads, beaded fibers, and nanofibers. It is well known that flow rate and applied DC voltage play important roles in the formation of nanofibers in electrospinning. From the figure, we estimate that, regardless of the auxiliary electrode, the critical DC voltages (from 20 to 23 kV) to stably fabricate alginate/PEO nanofiber were very similar to each other. Therefore, the auxiliary electrode used in this electrospinning system could assist in the enhancement of the production rate of electrospun alginate/PEO nanofibers without an increase in the supplementary applied DC voltage.

[FIGURE 6 OMITTED]

Figure 7a and b shows SEM images of the elelectrospun mats of alginate/PEO nanofibers deposited on the rectangular target. The image in Fig. 7a was obtained using the standard electrospinning method with three nozzles. The distance from the nozzles to the target was 150 mm, the spinning voltage was 27.5 kV at the nozzles, and the flow rate for each nozzle was 0.5 ml/h. The image in Fig. 7b was obtained using a three-nozzle system with three supplementary conical electrodes. Comparison of these two figures shows a reduction in the diameter of the spun fibers from 246 [+ or -] 83 nm for the standard electrospinning in Fig. 7a to 174 [+ or -] 62 nm for the modified electrospinning in Fig. 7b. As described by Fridrikh et al., the simplified diameter of the terminal jet ([d.sub.t]) can be determined from the equation,

[FIGURE 7 OMITTED]

[d.sub.t][infinity][([gamma][epsilon]).sup.1/3][(Q/I).sup.2/3][(1/ln [kappa]).sup.1/3] (1)

where [gamma] is the surface tension, [epsilon] is the dielectric constant, Q is the flow rate, l is the current, and [kappa] is the ratio of the initial jet length to the diameter of the nozzle (18). In this equation, if the current of whipped fibers in both electro-spinning systems is similar, the size reduction for the electrospinning process using an auxiliary electrode is reasonable because the length of the initial spun jet for the process using the auxiliary electrode is longer than that of the normal spinning process. However, this simple estimate is not completely correct because the current has a complex dependence on the voltage applied to the nozzles and auxiliary electrodes. A more detailed analysis of the reduction in size of the electrospun fibers with the auxiliary electrode will be the subject of future research. The uniform size of spun fibers from the electrospinning process supplemented by auxiliary electrodes may be the result of a stable induced electric held condition under the conical electrodes stabilizing the chaotic motion by guiding the initial jets.

To evaluate the effects of using the multiple nozzle system supplemented with auxiliary cylindrical electrodes on the mass production rate, the spun jets were collected on an aluminum target foil for various times and different nozzle-target distances. The target was 50-mm high and 300-mm wide. Figure 8a shows the weight of nanofibers collected as a function of time; it is clear that the collection rate is larger with the auxiliary electrode than without it. Figure 8b shows weight change of nanofiber on the collector for various distances between the collector and the needle tip. The deposited weight decreased as the distance increased with or without the auxiliary electrode. However, the increased fraction of weight became larger as the distance increased. This result indicates that the auxiliary electrode makes the electrospinning stable, regardless of distance, and dramatically improves the production rate.

[FIGURE 8 OMITTED]

For the cell experiment, the fiber strand formation of alginate nanofiber was maintained well after the [CaCl.sub.2] treatment (13). The cell morphology of the HDFs was observed using a SEM. The HDFs cells were embedded in the alginate nanofiber mat and maintained in the culture for 3 d. As shown in Fig. 9, the HDFs were initially attached as round shapes; after 5 d, they had elongated and were well spread out on the alginate nanofiber.

[FIGURE 9 OMITTED]

CONCLUSIONS

We fabricated nanofibers of SA blended with PEO and lecithin as a surfactant by using a modified multiple-nozzle electrospinning system supplemented with auxiliary conical electrodes. Alginate nanofibers with a composition of S2P2 provided the most stable electrospinning process. Using 0.7 wt% of lecithin as a surfactant produced good uniform bead-free nanofibers. For scaling up the production rate of the natural biocompatible nanofibers, the modified multiple-nozzle electrospinning system presented a stable initial stream line and jets of nanofibers from the nozzles. This stable jet motion was not influenced by nearby charged jets, environmental conditions such as airflow, or interference from nearby dielectric or conductive materials. This system achieved excellent processability and a high production rate of alginate nanofibers without sacrificing applied voltage loss. The multiple-nozzle electrospinning system with an auxiliary electrode is extremely practical for obtaining a high production rate for electrospun nanofibers. The cell culturing results indicate that electrospun alginate-based nanofiber mats have good potential for use as biomedical materials.

ACKNOWLEDGMENTS

The authors are grateful to Dr. S. A. Park for her assistant of cell culturing test. This work was supported by research funds from Chosun University, 2008.

REFERENCES

(1.) H. Yoshimoto, Y.M. Shin, H. Terai, and J.P. Vacanti, Biomaterials, 24, 2077 (2003).

(2.) Z. Ma, W. He, T. Yong, and S. Ramakrishna, Tissue Eng., 11, 1149 (2005).

(3.) W.E. Teo and S. Ramakrishna, Nanotechnology, 17, R89-R106 (2006).

(4.) S.A. Theron, A.L. Yarin, E. Zussman, and E. Kroll, Polymer, 46, 2889 (2005).

(5.) J.M. Deitzel, J. Kleinmeyer, D. Harris, and N.C. Beck Tan, Polymer, 42, 261 (2001).

(6.) O.O. Dosunmu, G.G. Chase, W. Kataphinan, and D.H. Reneker, Nanotechnology, 17, 1123 (2006).

(7.) A.L. Yarin and E. Zussman, Polymer, 45, 2977 (2004).

(8.) G.H. Kim, Y.S. Cho, and W.D, Kim, Eur. Polym. J., 42, 2031 (2006).

(9.) J.W. Lu, Y.L. Zhu, Z.X. Guo, P. Hu, and J. Yu, Polymer, 47, 8026 (2006).

(10.) N. Bhattarai, Z. Li, D. Edmondson, and M. Zhang, Adv. Mater., 18, 1463 (2006).

(11.) G.E, Wnek, M.E. Carr, D.G. Simpson, and G.L. Bowlin, Nana. Lett., 3, 213 (2003).

(12.) Y. Qin, J. Appl. Polym. Sci., 91, 1641 (2004).

(13.) X.L. Yan, E. Khor, and L.Y. Lim, J. Blamed. Mater. Res., 58, 358 (2001).

(14.) Y. Qin, H, Hu, and A. Luo, J. Appl, Polym. Sci., 101, 4216 (2006).

(15.) T. Takei, S. Sakai, H. Ijima, and K. Kawakami, Biotechnol. J., 1, 1014(2006).

(16.) A. Esquisabel, R.M. Hernandez, M. Igartua, A.R. Gascon, B. Calwo, and J.L. Pedraz, J. Microencap., 17, 363 (2000).

(17.) G.H. Kim, H.S. Han, J.H. Park, and W.D. Kim, Polym. Eng. Sci., 47, 707 (2007).

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

GeunHyung Kim, (1) Ko-eun Park (2)

(1) Department of Mechanical Engineering, Chosun University, GwangJu, Korea

(2) Division of Nano-Machinery, KIMM, Daejeon, Korea

Correspondence to: GeunHyung Kim; e-mail: [e-mail]

Contract grant sponsor: Chosun University.

DOI: 10.1002/pen.21472

Published online in Wiley InterScience (www.interscience.wiley.com).