Melt electrospinning process optimization of polylactic acid nanofibers

Abstract

Melt electrospinning is a safe and simple technique for the production of micro and nanofibers which can be an alternative to conventional solvent electrospinning. The effects of various melt-electrospinning parameters, including molecular weight, electric field strength, flow rate and temperature on the morphology, and fiber diameter of polylactic acid nanofibers were studied. An orthogonal design was used to examine process parameters. It was shown that molecular weight was the predominant factor in determining the obtainable fiber diameter of the collected fibers and is the most effective parameter on the average fiber diameter of melt electrospun PLA nanofibers. The flow rate has the less important impact. Mean fiber diameter (MFD) increased by increasing MW and flow rate, but decreased by increasing electric field strength and temperature. MFD of optimized fibers was below 100 nm and the result of statistical analysis using Design Expert software was in good agreement with the experimental condition.

Introduction

The most remarkable advantage of electrospinning is that it can be performed with various polymers both in solution and in the melt [1]. In melt electrospinning, instead of a solution, the polymer melt is introduced into the capillary tube. Melt electrospinning can offer several advantages over solution electrospinning such as, no requirement of dissolution of polymers in organic solvents and their removal/recycling, environmentally safe and a higher throughput with no loss in mass by solvent evaporation [2–5]. Some chemicals that are used as solvents for dissolving the polymers being electrospun may leave remnants that are not compatible. Therefore, in spite of the many potential applications, environmental and health limitations, as well as productivity complications do exist with solvent based electrospinning systems. Many solvents that are used in electrospinning are often costly and this is a major part of the cost factor associated with the process. Melt electrospinning can offer several advantages over solution electrospinning such as, no requirement of dissolution of polymers in organic solvents and their removal/recycling, environmentally safe and a higher throughput with no loss in mass by solvent evaporation, and generation of sub-micron scale fibers of polymers which lack appropriate solvents at room temperature (for example polyethylene and polypropylene) [6,7]. Therefore, there is great interest towards the use of molten polymers to produce electrospun mats as these are also less expensive. Melt electrospinning becomes a viable technology for the production of fibers for biomedical application as it avoids toxic solvents, thereby enabling the fibers to be directly spun on the desired surface or material. If melt blowing technology could be extended to submicron fiber sizes, it would provide a much easier, faster, and cheaper alternative to electrospinning. Hence, a major focus of melt blowing research should arguably be to extend the technology to nanofibers due to the potential to penetrate new markets and enhance current product offerings [8]. Regardless of the potential benefits of melt-electrospinning, very little progress has been made in the past 20 years. One of the most promising biopolymers able to replace the synthetic polymers for industrial applications is poly(lactic acid) (PLA). PLA is a linear aliphatic thermoplastic polyester derived from completely renewable resources such as sugar, cane, beet, etc. The most common route for industrial production of high molecular weight PLA is the ring opening polymerization (ROP) of lactide monomer formed from lactic acid, which is produced by fermentation of renewable agricultural resources [9]. PLA possesses interesting physical properties together with biocompatibility and biodegradability properties, which are all strongly influenced by its stereochemistry and molecular weight. In addition, it is easily processed into a desired configuration on standard plastics equipment to yield molded parts, films, or fibers [8]. It has been proposed that change of some parameters may influence the diameters of the fibers [10]. It was previously shown that the strength of the film/mat of fibers produced by electrospinning is sensitive to fiber diameter [11]. Moreover, size of the fibers along with morphology influences the hydrophobic behavior of polymers [12]. Therefore, controlling the mean fiber diameter which is a function of process parameters is crucial. Different diameters result from different approaches and electrospinning equipment used. Addition of Irgatec® (BASF, peroxide free polymer modifier) to viscosity reduction caused significant reduction of fibers to nanoscale [10]. Gas assisted melt electrospinning resulted in a 20-fold decrease in the final fiber diameter [13]. In this study, melt electrospun fibers of PLA were prepared and the effects of electrospinning parameters on the mean fiber diameter (MFD) of electrospuned fibers were investigated using Taguchi method. To the best of our knowledge, this is the first work that studies the melt electrospinning process parameters of PLA nanofibers.

Materials and methods

Materials

PLA with molecular weights of 30, 60, and 75 kDa was purchased from Sigma Aldrich. All other chemicals and reagents were of analytical grade.

Preparation of PLA electrospun fibers

Melt electrospinning was used for the production of PLA fibers. A proper amount of PLA pellets were loaded into the heating tube made of plastic and was heated to 175 ℃ for about 20 min to be completely molten. The molten polymer was pushed using the syringe pump to eject through a grounded metallic spinneret. A gas-assisted melt electrospinning nozzle (air with temperature of 110 ℃ and velocity at nozzle of 300 m/s) was used to produce thinner fibers (Figure 1). The ejected polymer melt was collected with the collector.

Figure 1.

Schematic of the gas-assisted melt electrospinning nozzle.

Characterization of fiber morphology

The morphology of electrospun scaffolds was characterized by scanning electron microscopy (SEM; Vega II XMU instrument Tescan, Czech Republic). Specimens were sputter-coated with gold for 20 s using a desk sputter coater (DSR1, Nanostructured Coating Co., Iran) and imaged with a back-scattering detector. At least 250 fiber measurements from more than 10 SEM images were used in order to ensure reproducible statistics when measuring fiber size.

Design of experiments

Effects of four factors A (temperature), B (electric field strength), C (flow rate) and D (molecular weight) on the mean fiber diameter were examined. Taguchi method is known as one the widely used tools in engineering applications, which helps in optimizing the engineering experiments through determination of best choices of factors influenced by a physical phenomenon, with the minimum number of experiments [14,15]. Therefore, requiring the minimum amount of experiments makes this engineering optimization tools an efficient one. Details of three levels for each factor are shown in Table 1. In order to cover a full representation of full-factorial experiment, Taguchi method employs some standard tables of experiments, called orthogonal arrays [16]. Nine experiments in accordance with the L9 orthogonal array were performed using Taguchi method. The results were analyzed using Design Expert (DX7) software.

Table 1.

Factors and their levels.

Level	A (℃)	B (kV/cm)	C (µL/min)	D (kDa)
1	175	2	20	30
2	185	3	40	60
3	195	4	60	75

Results and discussion

The experimental process conditions and their responses are presented in Table 2. Figure 2 shows the SEM images of prepared fibers in each experiment. The effects of the four factors on the mean fiber diameter can consequently be listed in the following order: Mw > Electric field strength > Temperature > Flow rate.

Table 2.

Design of experiments and their results.

EXP	A (℃)	B (kV/cm)	C (µL/min)	D (kD)	MFD (nm)
1	195	3	20	75	350
2	175	4	60	75	340
3	175	2	20	30	100
4	185	3	60	30	90
5	195	4	40	30	160
6	175	3	40	60	290
7	185	4	20	60	270
8	185	2	40	75	400
9	195	2	60	60	250

Figure 2.

SEM images of melt electrospun fibers (scale bar = 2 µm).

Effect of molecular weight on fiber diameter

Figure 3(a) shows the effect of MW on the mean fiber diameter of PLA fibers. As can be seen, MFD increases by increasing the MW of PLA. Molecular weight of the polymer is a direct reflection of melt viscosity. Its value is decided by the relative molecular mass and its distribution, the length of polymer chain, the conformation of chain, the interaction between different chains, and some other factors. Differences in the three MW values signify a considerable difference in melt viscosity. Consequently, MW is found to be the most significant control factor affecting average diameters. Lyons et al. [17] also reported a significant relationship between fiber diameter and Molecular weight of polypropylene fibers. In another aspect, MW is the only factor about the material while the electric field strength, temperature and flow rate are external factors, which effect on melts cannot act directly.

Figure 3.

Effects of process parameters on MFD.

Effect of electric field strength on fiber diameter

Figure 3(b) shows the relationship between electric field strength and mean fiber diameter. As expected, it was seen that the fiber diameter decreased as the electric field strength increases. Consistent with past electrospinning research [18–20], an increase in the electric field strength decreased the mean fiber diameter. Weaker field strengths were not strong enough to overcome the surface tension and viscoelastic forces of the molten polymer. When a steady amount of polymer is being supplied to the spinnerette, an increase in the electric field strength exposes the polymer droplet to larger forces, therefore further reducing the fiber diameter.

Effect of flow rate on fiber diameter

It was suggested that a minimum value for melt flow rate is required to form the drop of polymer at the tip of the needle for the sake of maintaining a stable Taylor cone [21]. It was also noted that the diameter of the fiber obtained is directly related to the flow rate as the case for solution electrospinning. As shown in Figure 3(c), mean fiber diameter increases by increasing flow rate. An increase in flow rate increased the amount of material flowing through the tip, which in turn resulted in increased fiber diameter.

Effect of temperature on fiber diameter

The effect of temperature on MFD of PLA fibers is shown in Figure 3(d). When spinning temperature is above the glass transition temperature of PLA (55 ℃), the average fiber size continues to decrease as temperature increases. Temperature usually has a significant effect on viscosity of melts. In the case of low viscosity, polymer chains move freely, with some of them being attracted to the collector along the force of the high electrical field, thus forming a fiber. If the viscosity is high, a large amount of chains move together, causing the fiber to be thicker.

Optimization

The results of analysis of variance (ANOVA) for mean fiber diameter is shown in Table 3. The results demonstrated that the model is highly significant, as is obvious from Fischer’s F-test with very low probability values (p < 0.05). To test the fit of the model, the determination coefficient (R²) was evaluated. The value of R² for MFD was 0.99. The high value of R² indicates that the model is adequate to describe MFD. The optimum conditions (fibers with minimum MFD) can be determined from the response table of the ANOVA analysis. The optimal parameters for the PLA fibers, with minimum diameter were found to be temperature 175 ℃, electric field strength 4 kV/cm, flow rate 40 µL/min, and molecular weight of 30 kDa. Experiments were performed several times using the optimized conditions. Under these conditions the mean fiber diameter was reached to 90 ± 12 nm. The result of software prediction was 86.8 nm. Comparing the experimental result with the response provided by the software shows that they are in good agreement. Figure 4 shows the image of PLA fibers at optimum conditions.

Table 3.

The ANOVA analysis for mean fiber diameter.

Source	Sum of square	DOF	Mean square	p-Value
Model	6073.22	8	759.15	0.0048
Temperature	12.52	2	6.26	0.0330
Electric field strength	324.08	2	162.04	0.0122
Flow rate	8.69	2	4.34	0.0408
Molecular weight	5727.93	2	12863.97	0.0051
Pure error	0	0
Total	6073.22	8

Significant at the 5% level (p < 0.05).

Unites of variables: temperature (℃); electric field strength (kV/cm); flow rate (µL/min), molecular weight (kDa).

Figure 4.

SEM image of optimum scaffold.

Conclusion

In summary, using an orthogonal design method, the effects of four factors (temperature, electric field strength, flow rate, and MW of the polymer) on the mean fiber diameter of PLA fibers were compared at three levels. Melt electrospinning results showed that MW is the most effective parameter. This may be attributed to melt viscosity, which varied distinctly as a function of MW at the three levels. Melt electrospun PLA fibers, were prepared at optimum conditions with average fiber diameter of less than 100 nm. These fibers could be good candidates for biomedical applications, because of their very high surface-to-volume ratio, and a relatively defect-free structure.