Electrospinning is a fascinating technique used to fabricate micro- to nano-scale fibers from a wide variety of materials. Molecular entanglement of the constituent polymers in the spinning dope is essential for successful electrospinning. We present a protocol for utilizing rheology to evaluate the electrospinnability of two biopolymers, starch and pullulan.
Electrospinning is a fascinating technique to fabricate micro- to nano-scale fibers from a wide variety of materials. For biopolymers, molecular entanglement of the constituent polymers in the spinning dope was found to be an essential prerequisite for successful electrospinning. Rheology is a powerful tool to probe the molecular conformation and interaction of biopolymers. In this report, we demonstrate the protocol for utilizing rheology to evaluate the electrospinnability of two biopolymers, starch and pullulan, from their dimethyl sulfoxide (DMSO)/water dispersions. Well-formed starch and pullulan fibers with average diameters in the submicron to micron range were obtained. Electrospinnability was evaluated by visual and microscopic observation of the fibers formed. By correlating the rheological properties of the dispersions to their electrospinnability, we demonstrate that molecular conformation, molecular entanglement, and shear viscosity all affect electrospinning. Rheology is not only useful in solvent system selection and process optimization, but also in understanding the mechanism of fiber formation on a molecular level.
Electrospinning is a technique that is capable of producing continuous micro- to nano-scale fibers from a wide variety of materials. It has gained increasing academic and industrial interest1. Though the setup and practice of electrospinning seem straightforward, the ability to predict electrospinnability and control fiber properties remains a challenge. The reason may lie in the fact that there are many factors influencing the electrospinning process2 and the process, especially the path travelled by the fiber, is chaotic1. Often an empirical “cook-and-look” approach is used for screening potential electrospinnable materials. However, to gain better control over the electrospinning process and resultant fiber properties, a more complete understanding of the mechanisms that govern electrospinnability is required. Several researchers have found that molecular entanglement of polymers in the spinning dope is an essential prerequisite for successful electrospinning3-5.
Rheology is a powerful tool to probe molecular conformation and interaction in polymer dispersions. For instance, McKee et al. investigated the molecular conformation of linear and branched poly(ethylene terephthalate-co-ethylene isophthalate) copolymers in a solvent containing chloroform/dimethyl terephthalate (7/3, v/v), and determined that the polymer concentration had to be 2-2.5x the entanglement concentration for successful electrospinning4.
There is currently renewed interest in fibers from biopolymers because of their advantages in biodegradability, biocompatibility, and renewability vis-à-vis their synthetic counterparts. Yet practitioners confront many challenges arising generally from their structural complexity, difficulty in thermal processing and inferior mechanical properties. Starch, found in plant tissues, is among the most abundant and inexpensive biopolymers on earth. Pure starch fibers fabricated using an electro-wet-spinning apparatus were recently described6. Pullulan is a linear polysaccharide produced extracellularly by certain bacteria. The regular alternation of (1→4) and (1→6) glucosidic bonds are believed to be responsible for several distinctive properties of pullulan, including excellent fiber/film forming capability7,8. Electrospinning of pullulan fibers from aqueous dispersion has been reported by a number of researchers9,10. In our previous publications, the electrospinnability of two biopolymers, starch11 and pullulan12, has been discussed. This report focuses on demonstrating the protocol for utilizing rheological principles in the investigation of the electrospinnability of these two biopolymers.
1. Spinning Dope Preparation
2. Steady Shear Rheology
3. Electrospinning Parameter Variation
Figure 1. Schematic drawing of the electro-wet-spinning setup. The biopolymer dispersion is extruded from a syringe pump. A high voltage DC power supply provides high voltage to the blunt needle and grounds the coagulation bath. The polymer jet from the needle tip travels through a straight path and then develops a rapid whipping path (aka whipping instability).
4. Morphological Characterization
Flow curves of the biopolymer dispersions as a function of biopolymer concentration and DMSO concentration in solvent were obtained. Two representative figures show the flow curves of starch (Figure 2A) and pullulan (Figure 2B) as a function of their concentration in pure DMSO solvent. The specific viscosities were plotted against biopolymer concentration (Figure 3A for starch and Figure 3B for pullulan). From these plots, entanglement concentrations were obtained as the intercept of the fitted lines in the semidilute unentangled and the semidilute entangled regimes.
Figure 2. Flow curves of (A) Gelose 80 starch and (B) pullulan in pure DMSO as a function of concentration (%, w/v) at 20 °C. In both figures, starch and pullulan of low concentrations were less viscous to produce sufficient torque at low shear rates. These unreliable data were thus not plotted. In general, the two biopolymers showed Newtonian behavior at low concentrations, i.e. the apparent shear viscosity was independent of shear rate. Shear thinning became apparent as their concentration increases, especially beyond 10% (w/v). Yet the shear thinning behavior was weak. The 15% and 20% (w/v) pullulan dispersions only showed the early phase of the power law region upon high shear rates, while the starch dispersions did not show significant reduction in viscosity over the shear rate range from 0.1 to 100 sec-1. Reprinted with permission from ref 11, Copyright (2012) American Chemical Society, and with permission from ref 12, Copyright (2014) Elsevier.
Figure 3. Plot of specific viscosity versus (A) Gelose 80 starch and (B) pullulan concentration in pure DMSO. The slopes of the fitted lines in semidilute unentangled (left side) and the semidilute entangled (right side) regimes indicate the concentration dependence of specific viscosity, aka scaling dependence4. Pullulan showed stronger concentration dependence than starch in the entangled regime. The intercept of the two fitted lines was termed as the entanglement concentration (ce) at which biopolymers start to overlap in the dispersion. Starch required a higher concentration than pullulan to start entangling. Reprinted with permission from ref 11, Copyright (2012) American Chemical Society, and with permission from ref 12, Copyright (2014) Elsevier.
Electrospinning was attempted for all of the biopolymer dispersions, and results judged in terms of electrospinnability, i.e. jet forming ability during electrospinning, and morphology of the fibers formed. A dispersion of good electrospinnability formed a stable and continuous jet that resulted in continuous and smooth fibers without droplets. A dispersion that was not able to electrospin could not form a stable jet or develop whipping instability. Either tiny droplets or thick fibers were deposited into the coagulation bath. Figure 4 shows representative good and poor fibers evaluated from their appearance. Figure 5 summarizes evaluation of electrospinnability at varying concentrations of DMSO in solvent and the biopolymer in dispersion for starch and pullulan, respectively. In addition to entanglement concentrations, shear viscosities at 100 sec-1 were plotted against biopolymer concentration, where regions of electrospinnability were denoted (Figure 6).
Figure 4: Scanning electron micrographs of good (left) and poor (right) starch and pullulan fibers. Good fibers are smooth, continuous, and randomly oriented, while poor fibers may have beads, breaks, and droplets as shown in the figure (red circles). (a) 10% (w/v) Gelose 80 starch in 95% (v/v) DMSO, (b) 8% (w/v) Gelose 80 starch in 80% (v/v) DMSO, (c) 17% (w/v) pullulan in 40% (v/v) DMSO, and (d) 9% (w/v) pullulan in 80% (v/v) DMSO. Reprinted with permission from ref 11, Copyright (2012) American Chemical Society, and with permission from ref 12, Copyright (2014) Elsevier.
Figure 5. Evaluation of electrospinnability of (A) Gelose 80 starch and (B) pullulan dispersions as a function of DMSO concentration in solvent and biopolymer concentration in the dispersion: good electrospinnability (circles), poor electrospinnability (diamonds), and unable to electrospin (X's). Shaded areas roughly represent electrospinnable regions. Entanglement concentrations are also approximately labeled. Reprinted with permission from ref 11, Copyright (2012) American Chemical Society, and with permission from ref 12, Copyright (2014) Elsevier.
Figure 6. Shear viscosity (at 100 sec-1) of (A) Gelose 80 starch and (B) pullulan dispersions as a function of biopolymer concentration in different DMSO/water solvents. Shaded areas roughly represent the electrospinnable region. Reprinted with permission from ref 11, Copyright (2012) American Chemical Society, and with permission from ref 12, Copyright (2014) Elsevier.
Rheology is an essential tool to study the processing of polymers, including conventional fiber spinning and electrospinning13. From the steady shear rheological studies, polymer conformation and their interactions in different solvents can be resolved (Figures 2 and 3). At concentrations not high enough for biopolymer molecules to overlap with one another, their concentration dependence was around 1.4 (Figure 3), which was in good agreement with reported values of other polymers in good solvent3,4. After the biopolymer molecules start to entangle, the specific viscosity showed a much higher dependence on concentration. A greater n value indicates a stronger intermolecular interaction. Many random coil polysaccharides showed similar concentration dependence, with an n value of approximately 3.3 14. Pullulan showed stronger interaction than starch in solvents of high DMSO content, possibly due to the molecular nature of the two biopolymers. The starch used had some highly branched components (~20% amylopectin), while the pullulan should be linear. Of course, the molecular weights, which were unknown, would also have an influence.
The entanglement concentration would depend on the conformation of the biopolymer in the dispersion. For example, the entanglement concentration of starch in 92.5% (w/v) DMSO aqueous solution is much lower than that in pure DMSO11. It implies that starch molecules exist in a more extended conformation in 92.5% (w/v) DMSO aqueous solution so that they occupy a larger hydrodynamic volume and tend to overlap more easily. The entanglement concentrations of pullulan did not vary as drastically as those of starch with varying solvent quality, probably because both water and DMSO are good solvents for pullulan and have little effect on the molecular conformation. Water, which is not a good solvent for starch, made the scenario much complicated, since undissolved starch molecules would affect the rheological response.
To spin good fibers, the concentration had to be 1.2-2.7 and 1.9-2.3x the entanglement concentration for starch and pullulan, respectively (Figures 4 & 5). This range is narrower for pullulan, probably also due to less conformation difference in the solvents. It was interesting to note that a dispersion at entanglement concentration, when polymers start to entangle with one another, was not electrospinnable. Probably, high shear force involved in electrospinning impede chain overlap and long range polymer interaction that might have already been established at static and low shear conditions, and thus an enhanced and sufficient entanglement is required. In addition, shear viscosity also played an important role (Figure 6). The electrospinnable starch and pullulan dispersions fall into a similar range of shear viscosity at 100 sec-1, with an upper boundary of 2.2 Pa·sec.
The procedure described herein can be modified in correspondence with the equipment and materials used in other studies. The dissolution of polymers is the first critical step in this protocol, because we found that partially dissolved starch dispersions (e.g., in 85% (w/v) DMSO) produced unstable steady shear viscosity data that prevented an accurate determination of ce. While conducting steady shear measurements, we prefer to start from the highest shear rate. By doing so, the dispersion is evenly distributed within the gap by the help of a high shear rate. The electrospinning step required much practice. Attention should be paid to the shape change of the droplet at the needle tip. Safety precautions during electrospinning should not be neglected. The main danger of electrospinning comes from the high voltage used in the process, although the current is relatively low. Electrospinning experiments should be performed in a fume hood in order to expel solvent vapor that may pose health hazards if one is exposed to it for a long time. Avoid close distance and even contact between the charged needle tip and the coagulation bath, because these will result in a short and fire hazard.
The rheological methods employed in the current study do have limitations. For instance, it should be noted that the actual shear rate involved in electrospinning is much higher than 100 sec-1 1. In addition to shear rheology studied, elongational rheology, which characterize the stretching of dispersion along the trajectory, may also play an important role15. The rheometer used in this study is not capable of characterizing elongational viscosity.
Rheological studies can provide valuable information on biopolymer conformation in dispersions and their processing properties. This protocol is potentially useful in electrospinning of many other biopolymers and their blends, in terms of solvent system selection, optimization of parameters, and fiber forming mechanism on a molecular level.
The authors have nothing to disclose.
This work is funded in part by the USDA National Institute for Food and Agriculture, National Competitive Grants Program, National Research Initiative Program 71.1 FY 2007 as Grant No. 2007-35503-18392, and National Institutes of Health, Institute for Allergy and Infectious Disease, R33AI94514-03.
Name of Reagent/ Equipment | Company | Catalog Number | Comments/Description |
Gelose 80 starch | Ingredion | Used as it is | |
Pullulan | Hayashibara Co. Ltd | Used as it is | |
Dimethyl Sulfoxide | BDH Chemicals | BDH1115-4LP | |
Ethanol | VWR International | 89125-172 | 200 proof |
Rheometer | TA Instruments | ARES | 50 mm cone and plate geometry |
Syringe (10 mL) | Becton, Dickinson and Company | 309604 | Syringe with Luer-Lok® Tip |
High voltage generator | Gamma High Voltage Research, Inc. | ES40P | |
Syringe pump | Hamilton Company | 81620 | |
Environmental scanning electron microscope | FEI Company | Quanta 200 | for starch fibers |
Environmental scanning electron microscope | Phenom-World | Phenom G2 Pro | for pullulan fibers |