Polymeric Nanofibers - American Chemical Society

Chapter 21

Downloaded by STANFORD UNIV GREEN LIBR on October 12, 2012 | http://pubs.acs.org Publication Date: February 23, 2006 | doi: 10.1021/bk-2006-0918.ch021

Mechanical Behavior of Nonwoven Electrospun Fabrics and Yarns Sian F. Fennessey, Angelo Pedicini, and Richard J. Farris* Polymer Science and Engineering Department, Silvio O. Conte National Center for Polymer Research, University of Massachusetts at Amherst, Amherst, M A 01003

Continuous, electrospun nanofibers are expected to out— perform their conventional counterparts in the reinforcement of composites due to their increased surface area available for adhesion and high aspect ratio (l/d). The mechanical behavior of nonwoven, electrospun nanofibers was examined as a function of fiber morphology, alignment, and degree of molecular orientation. Fiber alignment and molecular orientation was improved with use of a high-speed, rotating collection device. The ultimate strength and modulus of nanofiber yarns have been shown to improve with addition of twist and through post drawing treatments.

Introduction Electrospinning, a fiber spinning technique that relies on electrostatic forces to produce fibers in the nanometer to micron diameter range, has been extensively explored as a method to prepare fibers from polymer solutions or melts [1]. The main feature of the electrospinning process is that it is a simple means to prepare continuous fibers with unusually large' surface to volume ratios and porous surfaces [2,3]. Due to the chaoticoscillation of the electrospinning jet, a characteristic feature of the electospinning process, randomly oriented and 300

© 2006 American Chemical Society

In Polymeric Nanofibers; Reneker, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2006.


301 isotropic structures in the form of nonwoven nanofiber mats or webs are often generated due to a lack of control over the forces driving fiber orientation and crystallization. Recent efforts have been made to control the spatial orientation of electrospun fibers for use with ID device fabrication, which requires well aligned and highly ordered architectures through redesign of the collection apparatus [4]. Molecular orientation has been observed in electrospun fibers collected onto a parallel plate and a rotating drum collection apparatus, although the orientation has not been quantified [5]. Progress in understanding the electrospinning technique has allowed for recent engineering efforts in processes used to collect electrospun fibers for various applications, however, very limited work has addressed the mechanical properties of electrospun fiber mats [6]. The mechanical properties and reinforcing behavior of continuous nanofibers are expected to differ significantly from their finite length, conventional counterparts. Composites containing continuous fibers perform better than those prepared containing short fibers, particles, or whiskers since the reinforcement effect of a fiber is dependent upon its length to diameter ratio, according to the Halpin-Tsai equations [7] and Christensen's equation [8] of composite modulus prediction. Christensen's equation suggests that composites containing short fibers, particles, or whiskers are only expected to produce moderate improvements over non-reinforced polymer materials. Short fibers initially reinforce the composite, however, the reinforcing effect reaches a plateau above which short fibers have a limited ability to reinforce the matrix. As the fiber length increases, the reinforcing effect is initially increased and the level of the plateau is raised. Therefore, there is a greater possibility of increasing the modulus and strength of a composite by using a continuous fiber with a high aspect ratio (l/d) as reinforcement, rather than a high modulus short fiber. In addition to a high aspect ratio (l/d), the homogeneous dispersion of fibers throughout the matrix, and good interfacial adhesion and load-transfer between the matrix and fiber lead to the improved strength and modulus of a composite [7a, 9]. Continuous, electrospun fibers are expected to have improved interfacial adhesion in comparison to conventional fibers due to electrospun fibers' large surface to volume ratio resulting in an increase in surface area available to adhere to a matrix material. Due to the electrospun fibers' irregular void structure among fibers of the fabric and hairiness of the yarns, electrospun fabrics and yarns are expected to take advantage of the mechanical interlocking mechanism of load transfer. The strength of a composite is based on adhesion between the filler and the matrix; i f the adhesion is poor, the composite strength and modulus will be essentially that of the matrix material. In the present work, the mechanical behavior of electrospun polyurethane and polyacrylonitrile fibers is investigated and the use of electrospun fibers for the reinforcement of thin films and nanocomposites is proposed. The effect of


302


morphology, crystallinity, molecular orientation, and fiber alignment on the mechanical behavior of polyurethane fibers is determined. With consideration of the results found for the polyurethane system, the electrospinning process is optimized for the fabrication of fibers for reinforcement applications. Carbon precursor fibers from polyacrylonitrile are aligned with various degrees of orientation with the use of a rotating target. The mechanical properties of twisted polyacrylonitrile nanofiber yarns are determined as a function of twist angle. The effect of drawing on the modulus and ultimate strength of aligned polyacrylonitrile nanofiber yarns is determined.

Experimental

Electrospinning Set-up The electrospinning apparatus consists of a polymer solution reservoir, a high voltage power supply, and a grounded target. Solution is loaded into a pipette and a wire electrode is immersed into the solution. The solution is electrospun between 9.5-15 k V horizontally across a gap of 15-20cm and collected onto a stationary target. When the polymer reservoir is empty, the process stops, the pipette is refilled and electrospinning is resumed. The solution can also be loaded into a syringe that is oriented onto a dual syringe pump, and an electrode is clipped onto the needle. The needle, electrode and grounded target are all enclosed in order to reduce the effect of air currents on the trajectory of the electrospining jet. The flow rate of solution to the needle tip is maintained so that a pendant drop remains during electrospinning. The solution is electrospun between 8-16kV horizontally across a gap of 13-16cm onto a rotating target [5a, 6a, 10, 11] as described elsewhere [12].

Microscopy Electrospun fibers were observed by field emission scanning electron microscopy (FESEM) and polarized optical microscopy. Samples were mounted onto S E M plates, sputter coated with gold, and examined using a JOEL J S M 6320FXV electron microscope operating at an accelerating voltage of 5kV. Measured fiber diameters include a 5% random error. Electrospun fibers were also examined using an Olympus BX51, polarizing optical microscope, to detect birefringence and with a 1 order red plate to determine the elongation sign of st


303 the fiber by determining the directions of vibration of the fast and slow components [13].


IR Dichroism Dried, electrospun fiber bundles were examined using a Perkin Elmer Spectrum 2000 infrared spectrometer (FTIR) with a polarized wire-grid to measure dichroism as a function of collection take-up speed onto a rotating target. Spectra were acquired with the draw direction of the electrospun fibers positioned both parallel and perpendicular to the electric vector direction of the polarizer. Care was taken to examine the same region of fibers in both instances. Spectra were recorded over the range of 700-4000cm with typically 64 scans. The dichroic ratio, D, and Herman's orientation function, f, were determined (Equation 1 and 3). _I

D = A||/AL

(1)

2

D = 2 cot oc

(2)

0

2

/ = [3 - 1]/ 2 = (D - 1)(D + 2)1 ( D - 1)(D + 2) 0

(3)

0

where A j | is the absorbance when the electric vector direction of the polarizer is oriented parallel to the fiber draw direction and Aj. is the absorbance when the electric vector is oriented perpendicular to the fiber draw direction, D is the dichroic ratio of an ideally oriented polymer, a is the transition moment angle, and O is the angle of the molecular segment relative to the fiber axis. The transition moment angle is the angle between the chain axis and the direction of the dipole moment change for the group of interest. 0

X-ray Diffraction Dried, electrospun fibers were examined by wide angle X-Ray diffraction ( W A X D ) as a function of collection take-up speed onto a rotating target. Pin hole collimated, monochromated C u K a radiation was used and diffraction patterns of electrospun fiber bundles were collected with a G A D D S detection system (Brucker) in air. Calcite was used as a reference to aid analysis. Diffraction patterns were collected after 10 hours of exposure per sample and a


304 background was subtracted. The Herman's orientation function, / determined using the primary equatorial arcs (Equation 4). 2

was

2

/ = [ 3 < c o s 0 > - l ] / 2 = j / | s i n O | [ 3 c o s 0 - l ] / 2 d < D / J / I s i n O l d O (4)


where O is the azimuthal angle between the axis of the molecular segment and of the fiber and / is the scattering intensity of the reflection at that angle.

Twisted Yarns Approximately 32cm x 2cm unidirectional tows of electrospun nanofibers were collected using a rotating target, linked together, and twisted using a Roberta electric spinner by Ertoel. The twisted yarns were rinsed in deionized water for 24 hours and then dried under vacuum at 100°C. The twist per centimeter (tpcm), denier, and the angle of twist of the yarn were determined.

Mechanical Testing The mechanical behavior of dried, electrospun fibrous mats was determined according to A S T M 1708D with dumbbell shaped tensile specimens using an Instron 4411 and crosshead speed of 10 mm/min (50%/min. strain rate) in tension at room temperature. The mechanical behavior of dried, twisted yarns of electrospun nanofibers was examined using an Instron 5564 with a crosshead speed of 2mm/min (10%/min. strain rate) in tension at room temperature. Samples were mounted on paper tabs and had a 20mm gage length. The crosssectional area was calculated from the denier and the density of P A N from the literature [14]. The initial modulus, ultimate strength, and elongation at ultimate strength was measured.

Drawing Uniaxially aligned yarns were drawn using an Instron 5564 with a crosshead speed of speed of 2mm/min (10%/min. strain rate) in tension at 85°C in deionized water. Samples were mounted with a 20mm gage length and elongated to various draw ratios. The fibers were dried, remounted with a 20mm gage length, and the mechanical behavior was examined as previously described.


305

Polyurethane Nanofibers


Introduction Although elastomeric nanofibers have no inherent utility in the field of fiberreinforced composites, elastomeric polymers can be electrospun to produce highly porous nanofiber membranes that may prove to be useful as filter media and for medical applications [15]. A i r and vapor transport properties of electrospun polyurethanes have been compared to commercial membranes, such as expanded polytetraflouroethylene (ePTFE) and Gore-Tex® membrane laminates, and the tensile properties of electrospun thermoplastic polyurethane were determined [16]. A fundamental understanding of the mechanical properties and behavior of electrospun fibers and nonwoven nanofiber fabrics is critical to membrane applications. In addition, predictive tools to estimate mechanical properties of electrospun material relative to their bulk analogs will allow for the design of further applications for electrospun materials. In the present work, the mechanical behavior of electrospun polyurethane nonwoven fabric mats was characterized and compared to that of the bulk material. Differences in the mechanical response of the material as a function of morphology and molecular orientation was observed. The tensile strength of the examined electrospun polyurethane systems was compared to the respective bulk material strength by factoring the density of the electrospun material relative to the bulk.

Materials A thermoplastic polyurethane elastomer Pellethane® 2103-80AE was received from an industrial source and dimethylforamide (DMF) was obtained from Sigma Aldrich Co. The polymer was used as received and D M F was dried prior to use. Solutions were prepared at room temperature with mixing. Solution cast polyurethane films were prepared from 6wt% solution of polymer in THF and dried at room temperature for at least 72 hours prior to use.

Mechanical Behavior of Electrospun Thermoplastic Polyurethane Electrospinning of Pellethane® 2103-80AE onto a stationary target results in the production of randomly oriented, entangled nanofibers with circular cross



306

Figure 1. (a)Entangled, electrospun Pellethane® 2103-80AE collected onto a stationary target with (b) circular cross sections. Reproduced with permission from reference 6c. Copyright 2003Elsevier.

sections (Figure 1). The fibers do not have smooth surfaces, but rather the surface is rough and contains contours. The electrospun fibers are weakly birefringent as observed under cross polars, although no crystallinity is detected by wide angle X-ray diffraction ( W A X D ) and no melting endotherm is observed by differential scanning calorimetery (DSC) of the electrospun fibers. During the electrospinning process the spinning jet undergoes extremely large draw ratios, therefore, it would be expected that the resulting collected fibers would contain some molecular orientation. Since D M F is a high boiling solvent, it is not expected that D M F is completely evaporated during the electrospinning process. A surface, or skin, forms on the jet due to the diffusion of moisture from humidity in the environment into the fiber and solvent remains in the core of the fiber. The fibers are unconstrained after collection onto the target and tend to relax. The polymer chains are able to rearrange and any molecular orientation that may have been induced by the electrospinning process is lost. The electrospun fabric of the polyurethane system has a film-like character due to the adhesion of the fibers to one another, forming junction points throughout the mat. The inter-fiber adhesion plays a significant role in the mechanical integrity of the electrospun polyurethane material. The cross sectional area of the electrospun mat was corrected for its density relative to the bulk material in the stress calculations (Table I). The stress-strain curve for the electrospun polyurethane is not typical of elastomeric materials, as is that for the bulk sample (Figure 2). The initial slope of the stress-strain curve for the electrospun material is lower than for the bulk due to the relatively small number of fibers bearing load at low strains. Fibers are initially unoriented, at angles to the direction of the applied load, and rotate to align with the direction of the


307 Table I. Density corrected, mechanical properties of Pellethane® 210380AE Pellethane® 2103-80AE Solution Cast Film


Electrospun fiber mat Isotropic

Density correction factor (%)

Ultimate Strength (MPa)

1.0

48.6

26.4

30.7

Elongation at break (%) 532 253

Strain (%) Figure 2. Mechanical behavior of a solution cast film (—) versus an electrospun fabric mat(

/

/

\

*

t

-'

*'

,

,

,

,

,

,

4

6

8

10

12

14

Surface Velocity of Rotating Target (m/s)

Figure 6. Orienation parameter determined by dichroism (7) and WAXD Q as a function of the surface velocity (m/s) of the rotating target. Reproduced with permission from reference 12. Copyright 2004 Elsevier.


315 180 . . 160 S

,40

f

— !

I" 55

80

"S

60

&


3

10

U

>

T A

40

f

A

(A

fT ® &

O

[* a

4

^ —IE

4

s

>

20 0

12

Angle of Twist (degree) Figure 7. Mechanical properties of twisted Polyacrylonitrile nanofiber yarn as a function of twist angle (°). Reproduced with permission from reference 12. Copyright 2004 Elsevier

The stress-strain behavior of the yarns appeared to be similar to that of commercially produced P A N fibers. Commercial P A N fiber has an ultimate strength of approximately 512MPa (after post-treatment), according to the literature [25]. Commercial P A N precursor fibers are drawn prior to stabilization which decreases the fiber diameter and reduces the probability of encountering a critical flaw in a given test length. The ultimate strength of commercial P A N precursor fibers is approximately three times larger than the yarns of electrospun nanofibers; the twisted yarns were not drawn. P A N precursor fiber with a diameter of 155|im, as measured by laser diffraction, and an average denier of 80 was prepared by dry-jet solution spinning in our laboratory and the mechanical properties were measured prior to post treatment. The initial modulus and ultimate strength of the dry-jet solution spun fiber was 2.6 ±0.1GPa and 56 ±13MPa, respectively. The initial modulus and ultimate strength of the twisted electrospun P A N yarn with a twist angle of 1.1° and 11° are both approximately 1.5 times, and 2.2 and 2.9 times greater than that of the dry-jet solution spun P A N fiber prior to post-drawing, respectively [12].

Post-treatment of Polyacrylonitrile Fiber Yarns The effect of post stretching of unidirectional electrospun P A N fiber yarns at elevated temperature on the mechanical properties was investigated.


316


Unidirectional tows of P A N nanofibers prepared by electrospinning onto a rotating target with a surface velocity of 9.8m/s were drawn above the glass transition temperature to 100% elongation. The initial modulus and ultimate strength increase with draw ratio from 3.0 ± 0.3 GPa and 114+11 M P a at zero draw to 4.8 + 0.5 GPa and 253 ± 47 MPa at 100% draw, respectively (Figure 8). The elongation at break decreases from 25 + 6% at zero draw to 9.4 ± 2% at 50% draw, and then remains approximately constant with increasing draw to

-5

15

35

55

75

95

Draw Ratio (%)

Figure 8. Mechanical properties of uniaxially aligned Polyacrylonitrile as a function of elongation (%)

100% draw. Unidirectional tows could not be drawn more than 100% at 8590°C and provide reproducible results. The ultimate strength of unidirectional yarns drawn to 100% is approximately half that of commercial P A N precursor fibers, which are drawn to a maximum of 600% elongation. The degree of orientation determines the mechanical properties of P A N fibers. It is also known that the mechanical properties of carbon fibers made from P A N fibers depend on the orientation imparted by the latter. The electrospinning process with rotating collection system provides a technique to prepare oriented carbon precursor fiber yarns with fiber diameters in the sub micron range.

Conclusions Electrospun fibers collected onto a stationary target are isotropic, with little or no molecular orientation, and entangled. The molecular orientation and


317


alignment of electrospun fibers is controlled and altered through the use of a rotating collection device. The ultimate strength and modulus of electrospun yarns is increased with the addition of twist to a critical twist angle. Post drawing of electrospun fibers increases their axial orientation and mechanical strength. Aligned, electrospun fibers with a high degree of orientation are expected to surpass conventional fibers in the reinforcement of composites due to the increased surface area available for adhesion, their high aspect ratio (l/d), and expected property improvement with smaller diameter fibers.

References 1. 2.

Larrondo, L. and R.S. Manley. J. Polym. Sci. 1981, 19, 921. Deitzel, J.M.; Harris, K . D . and N.C. Beck Tan. Polymer 2000, 42, 261272. 4. (a) Theron, A . ; Zussman, E.; Yarin, A . L . Nanotechnology 2001, 12, 384. (b) Li, D.; Xia, Y . Nano Lett. 2003, 3, 555. (c) Zussman, E.; Theron, A . ; Yarin, A . L . Appl. Phys. Lett. 2003, 82, 973. (d) Li, D.; Wang, Y.; Xia, Y. Adv. Mater. 2004, 16, 361. 5. (a) Lee, S.; Yoon, J.; Suh, M . Macromol. Res. 2002, 10, 282. (b) Dersch, R.; Liu, T.; Schaper, A . K . ; Greiner, A.; Wendorff, J.H. J. Polym. Sci.: Part A 2003, 41, 545. (c) Kameoka, J.; Orth, R.; Yang, Y . ; Czaplewski, D.; Mathers, R.; Coates, G.W.; Craighead, H . G . Nanotechnology. 2003, 14, 1124. (d) Sundaray, B . ; Subramanian, V . ; Natarajan, T.S.; Xiang, R.; Chang, C.; Fann, W. Appl. Phys. Lett. 2004, 84, 1222. 6. (a) Lee, K . ; Kim, H . ; Ryu, Y . ; Kim, K . ; Choi, S. J. Polym. Sci.: Part B 2003, 41, 1256. (b) Yang, K.S.; Edie, D.D.; Lim, D.Y.; Kim, Y.M.; Choi, Y . O . Carbon 2003, 41, 2039. (c) Pedicini, A.; Farris, R.J. Polymer 2003, 44, 6857. 7. (a) Kardos, J. L . High Performance Polymers. Baer, E. and A . Moet, Eds. New York, Hanser Publishers 1991, Ch 6. (b) Halpin, J.C. and J.L. Kardos. Polym. Eng. & Sci. 1976, 16, 344. (c) Halpin, J.C. Revised Primer on Composite Materials: Analysis. Pennsylvania, Technomic Publishing Co. 1984, Ch. 6. 8. Christensen, R . M . Mechanics of Composite Materials. New York, John Wiley & Sons Inc. 1979, Ch. 3. 9. Schwartz, M. Composite Materials: Properties, Nondestructive Testing, and Repair. New Jersey, Prentice Hall Inc. 1996, Ch 1. 10. MacDiarmid, A . G . ; Jones, W.E.; Norris, I.D.; Gao, J.; Johnson, A.T.; Pinto, N.J.; Hone, J.; Han, B.; Ko, F.K.; Okuzaki, H . ; Llaguno, M . Synth. Met. 2001, 119, 27.



318 11. (a) Ding, B . ; K i m , H . ; Lee, S.; Lee, D.; Choi, K . Fibers and Polymers 2002, 3, 73. (b) Kang, Y . ; K i m , H . ; Ryu, Y.; Lee, D. ; Park, S. Polymer (Korea) 2002, 26, 360. 12. Fennessey, S.F.; Farris R.J. Polymer 2004, 45, 4217. 13. Chamot, E.; Mayson, C. Handbook of Chemical Microscopy Volume 1. New York, John Wiley & Sons 1938, Ch.10. 14. (a) Brandrup, J. and E . H . Immergut. Polymer Handbook. 3rd Ed. New York, John Wiley & Sons Inc. 1989, V/57. (b) Brandrup, J.; Immergut, E.H. and E.A. Grulke. Polymer Handbook. 4th Ed. New York, John Wiley & Sons Inc. 1999, VII/11. 15. (a) W.L. Simm, "Apparatus for the Production of Filter by Electrostatic Fiber Spinning," U.S. Patent 3,994,258 (1976). (b) D. Groitzsch, E. Fahrback, "Microporous Multilayer Nonwoven Materials for Medical Applications," U.S. Patent 4,618,524 (1986). (c) C.H. Bamford, K . G . Al—

16. 17. 18. 19. 20. 21.

22. 23.

24. 25.

Lamee, "Functionalisation of Polymers," U.S. Patent 5,618,887 (1997). (d) G.E. Martin, I.D. Cockshott, F.J.T. Fildes, "Fibrillar Lining for Prosthetic Device," U.S. Patent 4,044,404 (1977). H . Schrueder-Gibson, P. Gibson, K . Sencal, M . Sennett, J. Walker, W. Yeomans, D. Ziegler, P.P. Tsai, J. of Adv. Mater. 2002, 34, 44. S.L. Cooper; A . V . Tobolsky. J. of Applied Polym. Sci. 1966, 10, 1837. Estes, G . M . ; Seymour, R.W.; Cooper, S.L. Macromolecules 1971, 4, 452. Chung, D.D.L. Carbon Fiber Composites. Boston, Butterworth-Heinemann 1994, Ch 2. Fitzer, E.; Manocha, L . M . Carbon Reinforcements and Carbon/Carbon Composites. New York, Springer-Verlag 1998, Ch1. (a) Chun, I.; Reneker, D.H.; Fong, H . ; Fang, X . ; Deitzel, J.; Beck Tan, N.; Kearns, K . Adv. Mater. 1999, 1, 36. (b) Kim, C.; Yang, K.S. Appl. Phys. Lett. 2003, 83, 1216. (a) Bahl, O.; Mathur, R.; Kundra, K . Fibre Sci. Tech. 1981, 15, 147. (b) Chari, S.; Bahl, O.; Mathur, R. Fibre Sci. Tech. 1981, 15, 153. (a) Bashir, Z.; Church, S.; Waldron, D. Polymer 1994, 35, 967. (b) Bashir, Z.; Tipping A.; Church, S. Polymer International1994, 33, 9. (c) Bashir, Z.; Atureliya, S.; Church, S. J. Mater. Sci. 1993, 28, 2731. (d) Dalton, S.; Heatley, F.; Budd, P. Polymer 1999, 40, 5531. (e) Zbinden, R. Infared Spectroscopy of High Polymers. New York, Academic Press 1964, Ch V . (f) Jasse B . ; Koenig, J. J. Macromol. Sci.- Rev. Macromol. Chem. 1979, C17,61. (a) Bashir, Z. Acta Polymer 1996, 47, 125. (b) Davidson, J.; Jung, H . ; Hudson, S.; Percec, S. Polymer 2000, 41, 3357. Moncrieff, R.W. Man-Made Fibres. 6th Ed. New York, John Wiley & Sons 1975.


Polymeric Nanofibers - American Chemical Society

Recommend Documents