Influence of solution properties on the roller electrospinning of poly(vinyl alcohol).
Fiber science and technology are developing forward increasingly into the 21st century. In recent years, nanotechnology has become a topic of great interest to researchers, especially nanofiber production is very important for scientists and companies because of their special properties such as much smaller fiber diameter, high surface area, small and controllable pore size etc. For that reason nanofibers have many important application areas such as filtration (1-7), wound dressing (8), composites (9-11), tissue engineering (12), biomedical devices (13), membrane (14), (15), sound absorbtion (16) etc.
Electrospinning process with a hollow needle is used to obtain ultra thin fibers (1-1000 nm). This process has been known for many years (17) but the scientific research and number of publications have increased recently. The electrospinning process was first patented by Formhals in 1934 (18). After Formhals' studies, many new methods have been developed about electrospinning to solve the technical problems such as the low production rates, needle clogging etc. In 2004, an electrospinning device with a multiple needle system was developed allowing high production rates (19-21). Nevertheless, it did not meet a commercial application because of needle clogging and other production problems. In the same year. He et al. (22) applied the vibration technology to electrospinning. This method was able to produce finer nanofibers under lower applied voltage and increase in fiber strength by decreasing the applied voltage. Lukas et al. (23) developed a rod method to obtain nanofibers from solutions directly to improve production rate. Another needleless approach increasing production rate 12 times compared to needle electrospinning was proposed by Yarin and Zussman (24). In 2005, Tomaszewski and Szadkowski (25) designed three types of multijet electrospinning heads (series, elliptic and concentric) to improve electrospinning method. They used poly(vinyl alcohol) in water solution as the spinning liquid. The concentric electrospinning head was selected as the best type with respect to both the efficiency and quality of the process. In the same year Warner et al. (26) developed a new vertical electrospinning method which has 10-500 times higher efficiency when compared with conventional electrospinning method. Recently, Dosunmu et al. (27) invented a novel needleless method for electrospinning multiple jets based on a cylindrical porous tube. The solution was electrified and pushed by air pressure through the tube. They claimed that mass production rate from this method was 250 times greater than conventional electrospinning method. The roller electrospinning process was developed by Jirsak et al. (28) in 2003 at the Technical University of Liberec, Czech Republic. This method is the first one commercialized under the brand name Nanospider by the Elmarco Company in Liberec. In this study, electrospinning of PVA by the Nanospider method was investigated.
Since poly(vinyl alcohol) nanofibers have many important applications such as membrane, filtration, wound dressing, drug delivery, the effect of polymer and solution properties on spinnability and nanofiber properties were studied using the device based on a hollow needle (29), (30). Lee et al. (31) investigated the role of molecular weight of atactic PVA in the structure formation and properties of PVA nanofabric. They determined that the molecular weight of PVA had a marked influence on the structure and properties of nanofabrics. Jun et al. (32) also investigated the effect of molecular weight on fiber properties. They obtained PVA nanofibers by electrospinning with needle and discussed bead formation and fiber diameter when using blends of different molecular weight of PVA. In another study, Park et al. (33) studied the effect of preparation parameters on rheological behavior and microstructure of aqueous mixtures of hyaluronic acid/poly(vinyl alcohol). They found that rheological properties depend on the degree of hydrolysis of PVA. Lee et al. (34) found the rheological responses show time-dependence under low shear rate values and they also observed the chain mobility of molecules decrease as the relaxation time increases, which is important to spin nanofibers. Lin et al. (35) obtained much more uniform PVA nanofibers with using chitosan as a thickener. There are another several works about poly(vinyl alcohol) nanofiber production with needle electrospinning (36), (37). Generally PVA polymers with molecular weights ranging from 70.000 to 195.000 were electrospun (38-42).
However there is a significant difference between mechanisms of electrospinning in the needle based and roller based process. Therefore, results of previous studies are not fully transferable to the roller process. In the needle spinning, the polymer solution is delivered to the tip of a metal capillary, which is linked with a source of high voltage. Due to electric field, the solution is formed into droplets or fibers which arc transported to a (usually grounded) collector electrode. Depending on molecular weight of polymer and it's concentration in solution, only beads, fibers and beads or only fibers are created. Shenoy et al. (43) found that the chain entanglement characterized by entanglement number [([n.sub.e]).sub.SO ln] significantly influences beads or fiber formation. If the value of [([n.sub.e]).sub.SO ln] is below 2, only beads occur. If the value is greater than 2.5 only fibers are created.
The roller spinning device (Fig. 1) consists of a slowly rotating roller partially immersed in the polymer solution and a collector electrode. A high voltage source is connected to the solution. Collector electrode is usually grounded. The roller is covered by the polymer solution which is always fresh due to rotation of the roller. Many Taylor cones (44) are simultaneously created on the roller surface producing nanofibers. The nanofibers are then transported towards the collector electrode. A supporting textile or nontextile sheet moves usually along the collector electrode and is covered by nanofiber layer so that the production process is continuous.
[FIGURE 1 OMITTED]
Creation of Taylor cones on the surface of roller was described by Lukas et al. (45). Spinnability of a specific polymer solution depends on it's ability to create relatively stable jets. If the jets break due to Rayleigh instability, the life of Taylor cones is extremely short and no spinning process occurs. Therefore, electrospraying (a method of generating a small droplets through electrostatic charging) is not typical for the roller process and single beads hardly occur. Beads can be found in the product only together with fibers.
Three different samples of atactic poly(vinyl alcohol) (PVA) were studied in this work and three solutions of different concentrations in water were prepared from each sample. And distilled water was used as a solvent. The concentrations of solutions were chosen so that three groups of solutions showed similar values of viscosity measured at low shear rate (Table 1).
TABLE 1. The properties of PVA materials. Solution Molecular Hydrolysis Tacticity Concentration Viscosity weight grade (%) (wt%) (Pas) ([M.sub.w]) (Shear (g/mol) rate 100 1/s) PVA11 67.000 88 Atactic 12.5 0.2 PVA12 67.000 88 Atactic 14.5 0.48 PVA13 67.000 88 Atactic 17 0.95 PVA21 80.000 88 Atactic 10 0.21 PVA22 80.000 88 Atactic 12 0.48 PVA23 80.000 88 Atactic 14 0.98 PVA31 150.000 88 Atactic 7.8 0.25 PVA32 150.000 88 Atactic 9 0.48 PVA33 150.000 88 Atactic 10.5 0.98
The conductivity and surface tension of the solutions were determined by a conductivity meter (Radelkis, OK-102/1) and Du Nouy Ring method (Kruss) using platinium ring and highly precise electronic balance respectively. Dependence of viscosity of the solutions on shear rate was measured using Haake RotoViscol Rheometer. Polymer solutions were then electrospun using roller (needleless) spinner is shown in Fig. 1. The spinning process parameters are shown in Table 2.
TABLE 2. Process parameters of the roller eleclrospinning. Roller Roller Roller Velocity of Distance Voltage length diameter velocity backing between (kV) (mm) (mm) (rpm) material the electrodes (m/min) (mm) 145 20 3.2 0.2 110 60
Relative humidity inside the spinner was 35% and temperature 26[degrees]C. The nanofibers produced were collected on the polypropylene (PP) antistatic spunbond nonwoven backing material moving along the collector electrode by a constant velocity 0.2 m/min.
The fiber morphology and diameters of the poly(vinyl alcohol) nanofibers were determined using a scanning electron microscopy (SEM). The TESCAN Digital Microscopy Imaging SEM using an accelerating voltage of 30 kV was employed to take the SEM photographs.
The quality of nanofibers (diameter, uniformity and morphology) was evaluated using SEM photographs and image analysis of Lucia 32G computer software. The average fiber diameters were calculated from 200 measured values and only fibers were used to measure fiber diameter not include nonfibrous elements.
Two kinds of average values were calculated from measured diameters.
[d.sub.i] = fiber diameter
[n.sub.i] = number of fibers with diameter [d.sub.i]
[A.sub.n] = [[SIGMA][n.sub.i][d.sub.i]/[SIGMA][n.sub.i]] (number average) (1)
[A.sub.w] - [[SIGMA][n.sub.i][d.sub.i.sup.2]/[SIGMA][n.sub.i][d.sub.i]] (weight average) (2)
The formulas (1) and (2) are used in macromolecular chemistry to evaluate numeric and weight average molecular weights. The ratio [A.sub.w]/[A.sub.n] characterizes the width of the molecular weight distributions and optimum value should be very close to one for uniform fibers. Similarly, this ratio is used here to characterize the width of fiber diameter distribution.
Beside fibers, some nonfibrous particles may be created during the electrospinning process. These are droplets mainly occuring when the molecular weight of polymer is low or thick fibers and foil-like formations if the molecular weight is over optimum value. The latter may occur due to low degree of fibers elongation. The amount of nonfibrous particles in the nanofiber layer was evaluated using Lucia 32G image analysis software and expressed as the percentage of the sample area occupied by nonfibrous particles.
Polymer throughput was measured from the velocity of supporting PP spunbond and area weight of nanofiber layer in grams of polymer per minute per length of spinning roller (grams/min/m).
RESULTS AND DISCUSSION
Measurement and Analysis of Solutions Properties
The values of electric conductivity and surface tension of poly(vinyl alcohol) solutions are shown in Tables 3 and 4. Whereas differences in molecular weight, concentration and viscosity of PVA solutions were considerable, differences in electric conductivity and surface tension of studied solutions were rather small.
TABLE 3. Conductivity values of PVA solutions. Solution Molecular Concentration (wt%) Electric conductivity weight (mS/cm) ([M.sub.w]) (g/mol) PVA11 67.000 12.5 2.20 PVA12 67.000 14.5 1.88 PVA13 67.000 17 1.94 PVA21 80.000 10 2.23 PVA22 80.000 12 1.92 PVA23 80.000 14 2.16 PVA31 150.000 7.8 2.38 PVA32 150.000 9 2.36 PVA33 150.000 10.5 2.45 TABLE 4. Surface tension values of PVA solutions. Solutions Surface tension value (mN/m) Temperature ([degrees]C) PVA11 49.1 20.3 PVA12 50.2 20.5 PVA13 56.2 20.3 PVA21 51.0 20.1 PVA22 52.8 20.5 PVA23 54.7 20.0 PVA31 54.7 20.1 PVA32 57.8 20.1 PVA33 62.0 20.0
The studied solutions show significant differences in rheologic behavior. Viscosity of PVA 80.000 and PVA 150.000 which are spinnable show a strong decrease in effective viscosity with increasing shear rate while nonspinnable PVA 67.000 does not (Figs. 2-4). This difference shows lack of polymeric character of PVA 67.000 which leads to a weak polymer network and polymer jet.
[FIGURE 2 OMITTED]
[FIGURE 3 OMITTED]
[FIGURE 4 OMITTED]
Spinnability of Solutions
The solutions of PVA with molecular weight 67.000 (PVA11, PVA12 and PVA13) were not spinnable while all other solutions were spinnable. For PVA 67.000, no Taylor cones appeared on the surface of spinning roller at any process parameters such as voltage, roller speed, distance of roller to collector, relative humidity, temperature etc. And all the solutions of PVA 80.000 and 150.000 were spinnable. PVA throughputs of the spinning process expressed in grams of solid spun polymer per minute and one meter length spinning roller (g/min/m) are shown in Fig. 5.
[FIGURE 5 OMITTED]
The articles dealing with the needle electrospinning do not study the process performance (throughput). It is well known that the spinning or spraying occurs in rather a narrow range of solution dosage. After having found optimum dosage, the nanofiber materials are prepared and studied.
The performance of the roller electrospinning process strongly depends on the properties of specific polymer solution. It was observed that the spinning performance relates to the number of Taylor cones present on the spinning roller. Number of Taylor cones per spinning area of roller ([m.sup.-2]) was counted using camera record (46). Taylor cones are created on the surface of polymer solution by the mechanism described by Lukas et al. (45). If the Taylor cone is stable, it moves together with the surface of rotating roller and produces a jet. Although the method to quantify the number of Taylor cones on the roller is only in development, we observe that more cones give greater performance. The number of cones depends on their average life time which is influenced by the stability of jets. As soon as the jet breaks, the cone disappears. Stability of jets depends on the strength of the polymer network (degree of polymer-polymer interactions) which among the others depends on both molecular weight and concentration of polymer solution.
PVA of the low molecular weight (in this case of 67.000 g/mol) does not create a polymer network strong enough to create stable jets. Increase in polymer concentration does not eliminate this insufficiency, although the low-shear-rate viscosities are as high as those of spinnable polymers. Rheological behavior of low molecular weight polymer solutions (PVA11, PVA12 and PVA13 in Figs. 2-4) is apparently more different than that of spinnable polymers. It is closer to Newtonian behavior of nonpolymer solutions. This suggests low polymer-polymer interactions and low strength of polymer network. Our results are not fully consistent with Shenoy et al. work about entanglement number of polymer (43). There are two possible reasons for this inconsistency: The first one is the differences in mechanisms of needle and needleless electrospinning. Second, the Shenoy's et al. conclusions were made for the polymers showing no polymer-polymer interactions. This is apparently not true in the case of PVA.
As shown in Fig. 5, increasing the concentration of the lower molecular weight PVA polymer solution proportionately increases spinning performance in terms of throughput, but with the higher molecular weight PVA solution after a concentration of 7.8%, there is little difference in spinning performance with 9 and 10.5% PVA solutions. The distributions of fiber diameters in Figs. 6 and 7, values of the ratio [A.sub.w]/[A.sub.n] in Table 5 and the SEM photographs in Fig. 8, show rather a narrow distribution of the fiber diameters.
[FIGURE 6 OMITTED]
[FIGURE 7 OMITTED]
[FIGURE 8 OMITTED]
To determine the quality of nanofiber webs, we calculated two kinds of average diameters such as number average and weight average. When Table 5 was analyzed 150.000 PVA at 9% concentration has the best fiber diameter uniformity. It was also determined about there has been a significant difference between the uniformity coefficient results statistically.
Amount of nonfibrous particles expressed as a percentage of photograph area occupied by nonfibrous particles from total photograph area is shown in Fig. 9. There are SEM photos of nonfibrous particles present in the fiber layer in Fig. 10.
[FIGURE 9 OMITTED]
[FIGURE 10 OMITTED]
Except for the 80.000 g/mol PVA at 10% concentration, which had noteably smaller fiber diameters, the different molecular weights and different concentrations did not have much influence on fiber diameters and their distributions, nonfibrous particles are formed namely if the molecular weight exceeds an optimum value which seems to lay close 80.000 in the case of used PVA samples.
In this study, the effect of polymer molecular weight and some solution properties such as conductivity, surface tension and rheological behavior were investigated on the roller electrospinning of poly(vinyl alcohol) concerning spinnability, process performance, fiber diameters, diameter distribution and nonfibrous particles. Solutions of PVA 80.000 and PVA 150.000 which are spinnable show decrease in effective viscosity with increasing shear rate while in the case of nonspinnable PVA 67.000 the decrease is smaller or none. It suggests a low level of intermolecular entanglements and weak polymer network in the case of PVA 67.000. Molecular weight has an important effect on the spinnability while concentration has not. On the contrary, spinning performance increases with polymer concentration. Increase in molecular weight and concentration did not have much effect on fiber diameter and distribution except 80.000 g/mol PVA at 10% concentration. Nanoflbers show rather narrow distribution of the fiber diameters and 150.000 PVA at 9% concentration is the best concerning fiber diameter distribution, nonfibrous particles are formed, if the molecular weight exceeds an optimum value. This appears to lay close to 80.000 in the case of used PVA samples.
The authors thank Nonwoven Department of Textile Engineering Faculty, Technical University of Liberec in Czech Republic for providing work conditions such as laboratory, device etc.
(1.) W. Simm, U.S. Patent 3994,258 (1976).
(2.) D.R. Salem, Structure Formation in Polymeric Fibers, Vol. 225, Hanser Gardner Publications, Ohio, USA (2001).
(3.) H.Y. Chung, J.R.B. Hall, M.A. Gogins, D.G. Crofoot, and T.M. Weik, U.S. Patent 6924,028 (2005).
(4.) E.E. Koslow, U.S. Patent 6872,311 (2005).
(5.) I. Krucinska, E. Klata, and M. Chrzanowski, New Textile Materials for Environmental Protection NATO Advanced Research Workshop-Intelligent Textiles for Personal Protection and Safety, Zadar-Croatia (2005).
(6.) C. Shin, G.G. Chase, and D.H. Reneker, Colloids. Surf. A., 262, 211 (2005).
(7.) J. Hruza, Presented at the 5th World Textile Conference AUTEX, Portoroz, Slovenia, June 27-29 (2005).
(8.) I. Krucinska, A. Blasinska, A. Komisarczyk, P. Kiekens, M. Chrzanowski, L. Szosland, and G. Shoukens, II. in Proceedings of the International Technical Textiles Congress, Istanbul, Turkey, Meta Basim Press, 1, 13-15 July (2005).
(9.) K.J.J.R. Balkus, J.P. Ferraris, and S. Madhugiri, U.S. Patent 20,030,168,756(2003).
(10.) E.P. Giannelis, Appl. Organomet. Chem., 12, 675 (1998).
(11.) S. Adanur and B. Ascioglu, II. in Proceedings of the International Technical Textiles Congress, Istanbul, Turkey, Meta Basim Press, 29, 13-15 July (2005).
(12.) E.G. Russel, Presented at the 7th Annual Textile Conference AUTEX, Tampere University of Technology Publications, Tampere, Finland (2007).
(13.) A.G. Scopelianos, U.S. Patent 5,522,879 (1996).
(14.) H. Liu and Y. Hsieh, J. Polym. Sci. Part B: Polym. Phys., 40, 2119 (2002).
(15.) B. Chu, B. Hsiao, and D. Fang, U.S. Patent 6,713,011 (2005).
(16.) K. Kalinova, Presented at the 7th World Textile Conference AUTEX, Tampere-Finland (2007).
(17.) F.R.S. Rayleigh, Philos. Mag., 44, 184 (1882).
(18.) A. Formhals, U.S. Patent 1975,504 (1934).
(19.) W.S. Lee, S.M. Jo, S.G. Go, and S.W. Chun, U.S. Patent 6616,435 (2003).
(20.) B. Ding, E. Kimura, T. Sato, S. Fujita, and S. Shiratori, Polymer, 45, 1895 (2004).
(21.) K. Fujihara, M. Kotaki, and S. Ramakrishna, Biomaterials, 29, 4139 (2005).
(22.) J. He, Y. Wan, and J. Yu, Int. J. Nonlinear Sci. Numer. Simul., 5, 253 (2004).
(23.) D. Lukas, S. Torres, and X. Qin, Presented at the 11th International Conference, TU Liberec (2004).
(24.) A.L. Yarin and E. Zussman, Polymer, 45, 2977 (2004).
(25.) W. Tomaszewski and M. Szadkowski, Fibers Textiles Eastern Europe, 13, 22 (2005).
(26.) S. Warner, S. Ugbolue, M. Jaffe, and P. Patra, National Textile Center Report, Project No. F05-MD01, USA (2005).
(27.) O.O. Dosunmu, G.G. Chase, W. Kataphinan, and D.H. Reneker, Nanotechnology, 17, 1123 (2006).
(28.) O. Jirsak, F. Sanetrnik, D. Lukas, V. Kotek, L. Martinova, and J. Chaloupek, Patent WO 2005024101 (2005).
(29.) X. Wang, D. Fang, K. Yoon, B.S. Hsiao, and B. Chu, J. Membr. Sci., 278, 261 (2006).
(30.) K.H. Hong, Polymer, 47, 43 (2007).
(31.) J.S. Lee, K.H. Choi, H.D. Ghim, S.S. Kim, D.H. Chun, H.Y. Kim, and W.S. Lyoo, J. Appl. Polym. Sci., 93, 1638 (2004).
(32.) Z. Jun, H.Q. Hou, J.H. Wendorff, and A. Greiner, J. E-Polymers Art, 38 (2005).
(33.) H.O. Park, J.S. Hong, K.H. Ahn, S.J. Lee, and S.J. Lee, Korea-Australia Rheol. J., 17, 79 (2005).
(34.) E.J. Lee, K.S. Dan, and B.C. Kim, J Appl. Polym. Sci., 101, 465 (2006).
(35.) T. Lin, J. Fang, H.X. Wang, T. Cheng, and X.G. Wang, Nanotechnology, 17, 3718 (2006).
(36.) B. Ding, H.Y. Kim, S.C. Lee, D.R. Lee, and K.J. Choi, Fibers Polym., 3, 73 (2002).
(37.) R. Khajavi and R. Damerchely, Pak. J. Biol. Sci., 10, 314 (2007).
(38.) W.K. Son, J.H. Youk, T.S. Lee, and W.H. Park, Mater. Lett., 59, 1571 (2005).
(39.) S. Adanur and B. Ascioglu, J. Ind. Text., 36, 311 (2007).
(40.) P. Supaphol and S. Chuangchote, J. Appl. Polym. Sci., 108, 969 (2008).
(41.) S. Sakai, K. Antoku, T. Yamaguchi, and K. Kawakami, J. Membr. Sci., 325, 454 (2008).
(42.) J. Zeng, A. Aigner, F. Czubayko, T. Kissel, J.H. Wendorff, and A. Greiner, Biomacromelocules, 6(3), 1484 (2005).
(43.) S.L. Shenoy, W.D. Bates, H.L. Frisch, and G.E. Wnek, Polymer, 46, 3372 (2005).
(44.) G.I. Taylor, Proc R Soc London Ser: A, 280 (1964).
(45.) D. Lukas, A. Sarkar, and P. Pokornz, J. Appl. Phys., 103(8), 1 (2008).
(46.) T.A. Dao and O. Jirsak, Nanofibers for the 3rd Millennium-Nanofor Life, Prague, 39 (2009).
Funda Cengiz, (1) Tuan Anh Dao, (2) Oldrich Jirsak (2)
(1) Engineering and Architecture Faculty, Department of Textile Engineering, Suleyman Demirel University, Cunur, Isparta 32260, Turkey
(2) Nonwoven Department of Textile Engineering Faculty, Technical University of Liberec, Halkova 6, 46117, Czech Republic
Correspondence to: F. Cengiz; e-mail: firstname.lastname@example.org
Contract grant sponsor: Czech Ministry of Industry and Commerce; contract grant number: 1H-PK2/46.
Published online in Wiley InterScience (www.interscience.wiley.com).
[C] 2009 Society of Plastics Engineers
|Printer friendly Cite/link Email Feedback|
|Author:||Cengiz, Funda; Dao, Tuan Anh; Jirsak, Oldrich|
|Publication:||Polymer Engineering and Science|
|Date:||May 1, 2010|
|Previous Article:||Electrical characterization of the boron trifluoride doped poly(3-aminoacetophenone)/p-Si junction.|
|Next Article:||Synthesis and antibiofouling properties of novel crosslinkable terpolymers containing semifluoroalkyl substituted aromatic side chains.|