• Polymer 45 (2004) 7597–7603


  •   
  • FileName: 31.pdf [read-online]
    • Abstract: Polymer 45 (2004) 7597–7603www.elsevier.com/locate/polymerMelt-electrospinning part I:processing parameters and geometric propertiesJason Lyons*, Christopher Li, Frank KoDepartment of Materials Science and Engineering, Drexel University, Philadelphia, PA 19104, USA

Download the ebook

Polymer 45 (2004) 7597–7603
www.elsevier.com/locate/polymer
Melt-electrospinning part I:
processing parameters and geometric properties
Jason Lyons*, Christopher Li, Frank Ko
Department of Materials Science and Engineering, Drexel University, Philadelphia, PA 19104, USA
Received 18 August 2004; received in revised form 30 August 2004; accepted 31 August 2004
Available online 11 September 2004
Abstract
The effects of various melt-electrospinning parameters on the morphology and fiber diameter of polypropylene of different tacticities were
studied. The effect of the electric field strength at various melt flow indexes of polypropylene on fiber uniformity, morphology, and diameter
was measured. It was shown that the molecular weight was the predominant factor in determining the fiber diameter of the collected fibers.
Observations prove that the tacticity also influences the fiber diameter. Atactic polymers having molecular chains incapable of crystallization
tend to produce larger diameter fibers than isotactic polymers capable of crystallization even at lower molecular weights. The polymer
volume, at a given time, supplied to the electric field affected fiber diameter. Those systems supplying the smallest volume, at a given time,
produced the smallest fiber diameter.
q 2004 Elsevier Ltd. All rights reserved.
Keywords: Polypropylene; Melt electrospinning; Tacticity
1. Introduction of the polymers being electrospun may leave remnants that
are not compatible within the industry. In the intent of
The utilization of electrostatic forces to deform materials cleaner processing, environmental safety, and productivity,
in the liquid state goes back many centuries [1]. Throughout there is a persistent desire to produce fibers by alternative
the 20th century, there have been a number of papers methods. Thus, in spite of the many potential applications,
dedicated to the study of electrohydrodynamic atomization environmental and health limitations, as well as pro-
[2–8]. Electrospinning is simply an extension of this ductivity complications do exist as a result of solvent
technology applied to higher viscosity fluids. Several based electrospinning systems. The use of molten polymers
researchers [9–16] performed experiments using polymeric to produce electrospun mats becomes a subject of great
solutions and were capable of producing fibers ranging in interest. In spite of the potential benefits of melt-electro-
diameter from a few nanometers to several micrometers. spinning, little progress has been made in the past twenty
Most of these fibers were being collected as nonwoven, years. Larrondo and Manley [19–21] were the first to
random fiber mats. These fibrous structures can potentially electrospin a molten polymer more than two decades ago.
be used in a variety of applications including filtration They were capable of spinning polypropylene (melt flow
devices, solar sails, reinforcement, nonwetting textile indexes 0.5–2.0) and succeeded in making fibers that were
surfaces, wound dressings, vascular grafts, and tissue greater than 50 mm in diameter. Their inability to spin sub-
scaffolds [16–18]. However, a vast majority of the fibrous micrometer diameter fiber was attributed to the large
structures were produced by solvent based electrospinning. increase in viscosity that could be many orders of magnitude
Certain chemicals that are used as solvents to dissolve many greater than that of a polymer solution. The electric field
strength used in their experiments was 3–8 kV cmK1 at a
* Corresponding author. Tel.: C1-570-650-4282; fax: C1-215-895-
spinning distance ranging from 1–3 cm. They observed that
6760. the polymer melt experienced a large initial decrease in
E-mail address: [email protected] (J. Lyons). diameter when placed in an electric field and as the electric
0032-3861/$ - see front matter q 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.polymer.2004.08.071
7598 J. Lyons et al. / Polymer 45 (2004) 7597–7603
field strength increases, the fiber diameter decreases. Other [8] or the small distance that the jet traverses before
groups [22,23], and the University of Massachusetts at contacting the collection device. The solidified fibers are
Dartmouth, have conducted research on melt-electrospin- deposited randomly on the surface of the grounded
ning polymers including poly(ethylene terephthalate) and collection plate. It has been shown that fiber diameter can
polyethylene. These groups reported a wide range of be controlled by adjusting the processing parameters such as
obtainable fiber diameters yet limited progress has been electric field strength, polymeric viscosity, and flow rate
made. A full understanding of the melt-electrospinning [11–14].
process, and its potential to replace solution electrospinning,
has not yet been realized. It is the object of this study to 1.2. Melt-electrospinning requirements
determine the feasibility of producing electrospun fibers, of
varying fiber diameters and morphologies, from the melt In conventional textile fiber formation from the melt,
and recognize trends revolving around the molecular weight small diameter fibers are made through simultaneous
of the polymer, the electric field strength, and the polymer’s control of the spinnerette diameter and the take up speed
tacticity. of the godet rollers. When the fiber passes through the godet
rollers, each rotating at a different speed, a shearing action
1.1. The electrospinning process (drawing) occurs on the molecular chains, thus inducing
molecular orientation while decreasing fiber diameter as
Electrospinning is a simple technique for the production much as 500%. In order to successfully produce nano or
of nano to micro scale fibers depending on the medium used. sub-micrometer diameter fibers through melt-electrospin-
The use of electrospinning to produce fibers from solution, ning, drawing of the polymer must occur as a result of the
without using pressure, was first reported in a patent issued electrostatic forces acting on the jet. The forces needed to
in 1934 by A. Formhals [25]. This technique incorporates create a reduction in diameter to the nanometer level are of
the generation of a strong electric field between the great interest and are currently being investigated.
polymeric melt within the extruder and a metallic collecting
device. Fig. 1 shows a schematic design of a melt-
electrospinning system. As a voltage is applied, a cone 2. Experimental
forms at the apex of the capillary or spinnerette. At a critical
voltage, the electrostatic forces acting on the jet overcome 2.1. Fiber spinning experiments
the surface tension and viscoelastic properties of the melt
resulting in a fine fiber extruded from the cone residing at The materials utilized during these experiments can be
the spinnerette. Similar to solution based electrospinning, seen in Table 1. All samples were used as received and they
the main driving force for fiber formation is the attenuation were purchased from Sigma-Aldrich. Polypropylene was
of the spin line under electrostatic forces. While in transit, chosen because of its relative ease to process from the melt
the jet diameter is continually reduced due to the and it large range of available molecular weights ranging
electrostatic forces acting on it until the point when the from the 10s of thousands to the 100s of thousands. In
viscosity once again overcomes the electrostatic forces as a addition, polypropylene is available in different tacticities.
result of jet solidification from cooling. Unlike similar By electrospinning different tacticities of polypropylene, the
experiments conducted by other researchers in melt- effect of molecular conformation on fiber diameter can be
electrospinning [23,24], these particular experiments did examined.
not exhibit a bending instability. The lack of a bending The polymers were processed through a 3/4 00 single
instability may be attributed to the extremely high viscosity screw Brabender table-top extruder with four heating zones
associated with the melt as has been demonstrated by Taylor at 200 8C and a 1.5 mm spinnerette. In these experiments the
spinnerette was grounded and the positive charge was
applied to a copper collection plate that was placed at
varying distances ranging from 2–5 cm. At greater dis-
tances, a much higher potential will be required. The electric
Table 1
List of polymers
Polymer Mw Mn
Isotactic polypropylene 580,000 165,700
Isotactic polypropylene 190,000 50,000
Isotactic polypropylene 106,000 21,000
Isotactic polypropylene 12,000 5000
Atactic polypropylene 19,600 5400
Atactic polypropylene 14,000 3700
Fig. 1. A schematic diagram describing the melt-electrospinning technique.
J. Lyons et al. / Polymer 45 (2004) 7597–7603 7599
field strength, expressed in terms of voltage/centimeter, 3. Results
required to extrude a jet from the cone ranged from 6–
15 kV cmK1. Higher applied voltages, at short distances,
3.1. Effect of molecular weight on fiber diameter
will result in electrical discharge between the spinnerette
and the collection plate.
It was observed that the molecular weight had a
significant impact on the feasibility of producing fibers
2.2. Characterization electrostatically at various electric field strengths. With a
sufficiently high molecular weight, weaker electric field
The morphology of the electrospun fibers was examined strengths (O10 kV cmK1) were incapable of producing
through field emission environmental scanning electron fibers. Fig. 2 shows the surface morphology of the
microscopy (Phillips XL-30 ESEM). A beam strength of polypropylene fibers obtained using an applied electric
15 kV with a spot size of 3 was used to take the field 15 kV cmK1 at a constant collection plate distance of
micrographs. The average fiber diameter and the respective 2 cm. Polypropylene with a high molecular weight did not
distributions were determined from 100 measurements of form a Taylor cone when placed in the electric field. A fiber
random fibers at each spinning condition. nearly the width of the spinnerette hole was slowly pulled to
Fig. 2. The morphology and fiber diameter distribution of polypropylene fibers at an electric field strength of 15 kV cmK1 at a collection plate distance of 2 cm.
The figure also shows the average, standard deviation, maximum, and minimum values of the fiber diameter. All values are reported in micrometers.
7600 J. Lyons et al. / Polymer 45 (2004) 7597–7603
the collection plate. It was seen that the highest molecular fiber diameter ditribution of the 12,000 Mw isotactic
weight polymers formed the largest diameter fibers. It was polypropylene. Fibers that were smaller than 1 mm have
also observed in all experiments, that a wide variation of been obtained, however, a majority of the fibers are above
diameter was present. In some instances there were standard 1 mm. The kurtosis of these graphs show a possible bi-modal
deviations upwards of 50%. An example of an average fiber distribution that may be attributed to the variation of
distribution is shown in Fig. 3. This distribution shows the polymer volumes being supplied to the spinnerette as a
result of inconsistent flow rates as such low RPM’s of the
extruding screw.
3.2. Effect of electric field strength on morphology and
diameter
As a result of the high viscosity of the polymeric melts, it
was required to work at considerably large electric field
strengths. Each molecular weight polypropylene was
electrospun at an electric field strength of 10, 12.5, and
15 kV cmK1 at a spinning distance of 2 cm. This field
strength is upwards of 10 times stronger than those reported
in solution electrospinning. Weaker field strengths were not
strong enough to overcome the surface tension and
viscoelastic forces of the molten polymer; at higher voltages
at this distance, electrical discharge would occur. As
expected, it was seen in all samples that the fiber diameter
decreased as the electric field strength increases. Fig. 4
shows the relationship between electric field strength and
fiber diameter for several of the electrospun polymers. In
this figure, the effects of molecular weigh and tacticity can
also be seen.
3.3. Effect of polymer volume at the spinnerette tip
In order to supply the appropriate amount of polymer to
the spinnerette, it was necessary to have the extruder at the
lowest RPM output. At times, the Brabender extruder
supplied polymer to the spinnerette faster than the
electrostatic forces could carry it away. As a result, it
became convenient to place the polymer directly on the
spinnerette to melt. In this experiment, the extruder was no
longer supplying a continuous volume at a given time to the
spinnerette. Thus, it was observed that the Taylor cone
continually decreased in size due to the reduction in
available polymer and smaller and smaller fibers were
produced from the diminishing cone as seen in Fig. 5.
4. Discussions
4.1. Molecular weight
From the results, it is evident that the molecular weight
plays a significant role in the feasibility of electrospinning
polymeric fibers. This finding is comparable and consistent
with past research claiming that solution concentration is
Fig. 3. The fiber diameter distribution of a 12,000 Mw polypropylene at
2 cm. (A) 10 kV cmK1, kurtosisZ1.22, (B) 12.5 kV cmK1, kurtosisZ3.62, the most dominant parameter in electrostatic spinning [26].
(C) 15 kV cmK1, kurtosisZ1.12. The graphs show a possible bi-modal It was observed that the 580,000 Mw polypropylene resulted
distribution. in fiber diameters in access of 400 mm. In addition, it was
J. Lyons et al. / Polymer 45 (2004) 7597–7603 7601
Fig. 4. (A) Chart showing the effect of the electric field strength on collected fiber diameter for selected electrospun polymers. The 190,000 and 580,000
molecular weight polymers were omitted due to graph distortion when inserted. (B) Effect of tacticity for electrospun polypropylenes. (C) Effect of molecular
weight for atactic polypropylene. (D) Effect of molecular weight for isotactic polypropylene (error bars represent the 190,000 Mw sample. Error bars for the
other samples can be seen in chart A).
observed that the 165,700 Mn isotactic polypropylene and result in larger fiber diameters than those obtained from
the 5400 Mn atactic polypropylene produced fiber diameters similar molecular weight polymers capable of crystal-
larger than similar tacticity polymers with smaller Mn lization. Therefore, the tacticity of the polymer has a
values. The polymer with the largest Mn will be subjected to significant effect on the fiber diameter. It is also possible that
the highest degree of polymer chain entanglement. It is variations in fiber diameter between polymers of different
therefore more difficult for the electrostatic forces to pull on tacticities are related to the memory effect of the polymer.
individual polymer chains. As a result, larger fiber diameter
will be formed, as was observed experimentally. As the 4.2. Electric field strength
molecular weight continually decreases, fiber diameter
decreases as shown in Fig. 6 for isotactic polypropylene. Consistent with past electrospinning research [10–12], an
It is seen that the molecular weight has an exponential effect increase in the electric field strength decreases the average
on the fiber diameter for isotactic polypropylene. There fiber diameter for all of the polymers examined. When a
were not enough data points to verify the same trend in the steady amount of polymer is being supplied to the
atactic samples. Fibers were not produced with electric field spinnerette, an increase in the electric field strength exposes
strengths less than 15 kV cmK1 for the 580,000 Mw isotactic the polymer droplet to larger forces, therefore further
sample. reducing the fiber diameter. It was observed that the angle
The molecular weight interaction was not exclusively (from jet axis) of the cone, at the spinnerette orifice,
responsible for determining the collected fiber diameter. As increases as the electric field strength increases as a result of
seen in Fig. 2, both atactic polypropylene samples formed more material being pulled away; consistent with past
larger diameter fibers than all but the 580,000 Mw isotactic research [10,11].
sample. Since atactic polypropylene possesses a random It is worthy to note that in some instances, while using
positioning of the methyl group off of the main molecular the 12,000 Mw polymer, that sub-micrometer fibers were
backbone, it is impossible to crystallize. The inability of the obtained as seen in Fig. 7. In all instances, these fibers are
polymer chains to closely pack due to steric hindrances may the result of branches from a larger fiber within the sample.
7602 J. Lyons et al. / Polymer 45 (2004) 7597–7603
Fig. 7. Sub-micrometer diameter fibers branching from 12,000 Mw isotactic
polypropylene.
This is conceivable because at the speeds that the jet is
traveling, it is not completely solidified once leaving the
spinnerette. As the jet travels further into the electric field, it
is exposed to stronger field strengths. If the molecular
weight of the polymer is low enough, it is possible that a
side jet can be created from the molten jet leading to smaller
diameter fibers. These fibers did not make up a majority of
the sample but they represent the only fibers to break the
1 mm barrier.
The surface morphology, of a majority of the polymers
Fig. 5. ESEM micrographs of 19,600 Mw melt-electrospun polypropylene at
an electric field strength of 10 kV cmK1 at 2 cm. (A) 10 s, (B) 20 s, (C)
electrospun, consists of smooth cylindrical fibers. It is
30 s. believed that the smooth surface is in part due to the partial
solidification that occurs as soon as the jet leaves the
spinnerette. Also, there is no solvent evaporation that may
lead to inconsistencies on the fiber surface.
4.3. Spinning volume at a given time
In order to study the effect of spinning volume, a polymer
chip was placed directly on the spinnerette orifice and
melted. Since no flow rate was being applied, the volume
would continually reduce as the polymer was transferred to
the collection device in the form of fibers. The diameters of
the collected fiber as a function of time can be seen in Fig. 5.
Similar to increasing the electric field strength a smaller
Taylor cone was formed as a result of the decreasing
volume. This cone was then exposed to larger field strengths
ultimately leading to a decrease in fiber diameter. This
observation suggests for the possibility of forming sub-
Fig. 6. Graph showing the exponential increase in fiber diameter with micrometer diameters consistently under specific experi-
increasing molecular weight for isotactic polypropylene. mental parameters for certain polymers from the melt if a
J. Lyons et al. / Polymer 45 (2004) 7597–7603 7603
small enough volume could be supplied to the spinnerette on also are extended to the Koerner Fellowship offered at
a consistent basis. This may also explain the large standard Drexel University for partial assistance in this study. The
deviations that result from the melt-electrospinning process. invaluable assistance of David Von Rohr for ESEM
assistance is very much appreciated.
5. Conclusions
Electrospinning polypropylene of various tacticities and References
molecular weight, from the molten state, was successfully
completed resulting in fiber diameters from several hundred [1] Gilbert W, de Magnete 1600.
nanometers to several hundred micrometers depending on [2] Zelany J. Phys Rev 1914;3:69.
[3] Zeleny J. Proc Cambridge Philos Soc 1915;18:17.
the electrospinning parameters and some important obser-
[4] Zeleny J. Phys Rev 1917;10:1.
vations were made. The surface morphology and diameter [5] Macky WA. Proc R Soc A 1931;133:565.
distribution of the fibers were studied as a function of [6] Nolan JJ. Proc R Irish Acad 1926;37A:28.
various electrospinning parameters including electric field [7] Vonnegut B, Neubauer RL. J Colloid Sci 1952;7:616.
strength, supplied volume, and molecular weight. The [8] Taylor GI. Proc R Soc Lond A 1969;313:453.
molecular weight of polymer was the dominant parameter [9] Baumgarten PK. J Colloid Interface Sci 1971;36(1):71.
[10] Deitzel J, Beck Tan NC, Kleinmeyer J, Rehrmann J, Tevault D,
determining the feasibility of electrostatically producing Reneker DH, Sendijarevic I, McHugh A. Generation of Polymer
polypropylene fibers from the melt. Although several trends Nanofibers through Electrospinning 1999, Army Research Labs.
were observed, many other parameters of the polymer and [11] Deitzel J, Kleinmeyer J, Harris D, Beck Tan NC. Polymer 2001;42:
electrospinning process must be examined including 261.
temperature, flow rate, spinnerette diameter, di-electric [12] Srinivasan G, Reneker DH. Polym Int 1995;36:195.
[13] Reneker DH, Chun I. Nanotechnology 1995;7:216.
constant, thermal conductivity, and surface energy. On the
[14] Chun I, Reneker DH, Fong H, Fang X, Deitzel J, Fan NB, Kearns K.
basis of observations made in the preliminary study, it J Adv Mater 1999;31(1):36.
would be helpful to develop a model relating fiber diameter [15] Deitzel J, Kleinmeyer J, Hirvonen JK, Beck Tan NC. Polymer 2001;
to the empirical processing parameters. Also, characteriz- 42:8163.
ation of the mechanical properties and structural analysis of [16] Doshi J, Reneker DH. J Electron 1995;35:151.
the collected fibers needs to be performed to gain insight on [17] Bognitzki M, Czado W. Adv Mater 2001;13(1):70.
[18] Jim HJ, Fridrikh SV, Rutledge GC, Kaplan DL. Biomacromolecules
the structure-properties relationship of melt-electrospun 2002;3:1233.
fiber. Accordingly, modeling of the melt-electrospinning [19] Larrondo L, Manley SJ. J Polym Sci: Polym Phys Ed 1981;19:909.
process will be the focus of subsequent studies. [20] Larrondo L, Manley SJ. J Polym Sci: Polym Phys Ed 1981;19:921.
[21] Larrondo L, Manley SJ. J Polym Sci: Polym Phys Ed 1981;19:933.
[22] Kim JS, Lee DS. Polymer 2000;32(7):616.
[23] Chun I, PhD Dissertation, University of Akron; 1995.
Acknowledgements
[24] Rangkupan R, Reneker DH. J Met Mater Sci Res Inst 2003;12(2):81.
[25] Formhals A, US Patent, 1,975,504; 1934.
This work was made possible, in part, by the State of [26] Sukigara S, Ghandi M, Ayutsede J, Micklus K, Ko F. Polymer 2003;
Pennsylvania under The Nanotechnology Institute. Thanks 44(19):5721.


Use: 0.1569