**4.1. Solvents for electrospun PU nanofibers**

Electrospinning solution properties are also directly related to the solvent type. The type of solvent and their concentration influenced the morphology of electrospun nanofiber mat significantly. Process involves the stretching of the solution caused by repulsion of the charges at its surface. Thus, if the conductivity of the solution is increased, more charges can be carried by the electrospinning jet [13], which increases the stretching of the polymer solution. As much as solution conductivity, the dielectric constant of a solvent has also a significant influence in electrospinning process. Higher dielectric property reduced the bead formation and the diameter of the resultant electrospun fibers [47].

Most commonly used solvents that dissolve PUs are highly polar organic solvents such as N,N-dimethylformamide (DMF), dimethylsulfoxide (DMSO), N-Methylpyrrolidone and tetrahydrofuran (THF) [50, 51]. Ketones such as acetone, methyl ethyl ketone, cyclohexanone are partial solvents for TPUs. Aliphatic alcohols such as ethanol and isopropanol cause a slight swelling; aliphatic esters such as ethyl acetate and butyl acetate strongly cause severe swelling of TPUs [50]. Investigation of the effects of solvents in electrospinning of TPU appeared in the study of Mondal [29] and Cay et al. [52]. Mondal used four different solvents (THF, DMF, N,N-dimethylacetamide, and DMSO) and reported that the morphology of the resultant nanofibers changed significantly with the solvent selection. Solvent conductivity and vapor pressure of the solvents were found to be the important factors (**Table 1**). In the study of Cay et al., the effects of the incorporation of ethyl acetate or tetrahydrofuran into TPU/DMF solvent system were investigated. The solutions of TPU in pure DMF and DMF/EA were found to be easily electrospinnable. DMF seemed to be the best solvent to dissolve TPU pellets but diluting TPU solutions with 10 or 20% of EA ensured positive effect on fiber diameter by means of achieving thinner fibers. Incorporation of THF to DMF led to thicker fibers compared to TPU/DMF solutions. With the increasing THF volume fraction, electrospinning is restricted due to high viscosity and low conductivity.

A mixed-solvent system of THF and N,N-dimethylacrylamide (DMAA) in the study of Kidoki et al. to investigate the relationships between the structural features and mechanical properties of electrospun segmented PU (SPU) meshes. They studied the polymer concentration and solvent mixing ratio to achieve different formulations and investigated the operation parameters such as applied voltage, tip to collector distance, and feeding rate. SPU was electrospun from the mixed solvent of THF and DMAA with different mixing ratios [DMF content: 5, 10, and 30% (v/v)]. An increase in DMAA ratio significantly affects the degree of bonding between SPU fibers at contact sites and leads to thinner fibers formation. The porosity of the electrospun SPU meshes decreased with increasing DMF ratio according to the porosimetric characterization. The pore size distribution exhibited three representative peaks of approximately 5, 20, and 70 μm void sizes. Increasing DMAA ratio markedly decreased the proportion of the 20 μm void. In addition to these, an increase in DMAA ratio induced an increase in elasticity of the mesh. The authors pointed that electrospun SPU meshes using a mixed-solvent system with low- and high-boiling point solvents may be useful in the engineering of SPU-fiber-based matrices or scaffolds [57].

### **4.2. Solution and processing conditions for electrospun PU nanofibers**

The diameter of the nanofibers produced by electrospinning is a key parameter for most of the applications. The diameter of the nanofibers defines the structural features such as pore sizes and specific surface areas. These features affect the selectivity of filters, the permeability of filters, catalytic activities in systems using nanofibers to immobilize catalysts, or the cell proliferation in tissue engineering relying on nanofiber-based scaffolds [1]. Solution parameters, especially polymer concentration and the spinning parameters including feeding rate, applied voltage, tip to collector distance, have the strongest impact on the fiber diameter. The properties of PUs can vary in a wide range according to the structure of PU, thus below investigated parameters were particular for the selected PUs.


Most commonly used solvents that dissolve PUs are highly polar organic solvents such as N,N-dimethylformamide (DMF), dimethylsulfoxide (DMSO), N-Methylpyrrolidone and tetrahydrofuran (THF) [50, 51]. Ketones such as acetone, methyl ethyl ketone, cyclo

hexanone are partial solvents for TPUs. Aliphatic alcohols such as ethanol and isopropanol cause a slight swelling; aliphatic esters such as ethyl acetate and butyl acetate strongly cause severe swelling of TPUs [50]. Investigation of the effects of solvents in electrospinning of TPU appeared in the study of Mondal [29] and Cay et al. [52]. Mondal used four different solvents (THF, DMF, N,N-dimethylacetamide, and DMSO) and reported that the morphol

ogy of the resultant nanofibers changed significantly with the solvent selection. Solvent conductivity and vapor pressure of the solvents were found to be the important factors (**Table 1**). In the study of Cay et al., the effects of the incorporation of ethyl acetate or tetrahydrofuran into TPU/DMF solvent system were investigated. The solutions of TPU in pure DMF and DMF/EA were found to be easily electrospinnable. DMF seemed to be the best solvent to dissolve TPU pellets but diluting TPU solutions with 10 or 20% of EA ensured positive effect on fiber diameter by means of achieving thinner fibers. Incorporation of THF to DMF led to thicker fibers compared to TPU/DMF solutions. With the increasing THF volume fraction, electrospinning is restricted due to high viscosity and low conductivity. A mixed-solvent system of THF and N,N-dimethylacrylamide (DMAA) in the study of Kidoki et al. to investigate the relationships between the structural features and mechanical properties of electrospun segmented PU (SPU) meshes. They studied the polymer concentration and solvent mixing ratio to achieve different formulations and investigated the operation parameters such as applied voltage, tip to collector distance, and feeding rate. SPU was electrospun from the mixed solvent of THF and DMAA with different mixing ratios [DMF content: 5, 10, and 30% (v/v)]. An increase in DMAA ratio significantly affects the degree of bonding between SPU fibers at contact sites and leads to thinner fibers formation. The porosity of the electrospun SPU meshes decreased with increasing DMF ratio according to the porosimetric characterization. The pore size distribution exhibited three representative peaks of approximately 5, 20, and 70 μm void sizes. Increasing DMAA ratio markedly decreased the proportion of the 20 μm void. In addition to these, an increase in DMAA ratio induced an increase in elasticity of the mesh. The authors pointed that electrospun SPU meshes using a mixed-solvent system with low- and high-boiling point solvents may be useful in the engineering of SPU-fiber-based matrices or

**4.2. Solution and processing conditions for electrospun PU nanofibers**

proliferation in tissue engineering relying on nanofiber-based scaffolds [

investigated parameters were particular for the selected PUs.

The diameter of the nanofibers produced by electrospinning is a key parameter for most of the applications. The diameter of the nanofibers defines the structural features such as pore sizes and specific surface areas. These features affect the selectivity of filters, the permeability of filters, catalytic activities in systems using nanofibers to immobilize catalysts, or the cell

eters, especially polymer concentration and the spinning parameters including feeding rate, applied voltage, tip to collector distance, have the strongest impact on the fiber diameter. The properties of PUs can vary in a wide range according to the structure of PU, thus below

scaffolds [57].

22 Aspects of Polyurethanes



1]. Solution param


**Table 1.** Properties of selected solvents [52–56].

Electrospun Polyurethane Nanofibers http://dx.doi.org/10.5772/intechopen.69937 23 A 32 factorial experimental design was used to investigate the tip to collector distance and applied voltage effect for PU nanofibers by Yanilmaz et al. [27]. They used TPU (PU, 270,000 g/mol) supplied from Coim Co. and chose THF for 10% (w/w) TPU solution, 5, 7.5, and 10 cm for tip-to collector distances and 10, 15, and 20 for applied voltages. From their design, it was seen that the distance and applied voltage had significant effects on the fiber diameter and it depends on the applied voltage and distance; furthermore, the interaction of these factors affects fiber diameter significantly.

Akçakoca Kumbasar et al., used DMF for TPU (Pellethane 2103-80AE) solutions and investigated the effect of TPU concentration (6, 8, 10, 12, and 14% (w/w) represented in **Figure 3**), tip to collector distance (8, 10, 12, and 15 represented in **Figure 4**), feeding rate (0.3, 1.5, 2.5, and 3.5 ml/h), and applied voltage (7, 10, 13 kV) on nanofiber diameters. Their results showed that 6% of TPU concentration is too low and 14% is too high for smooth nanofibers and as expected nanofiber diameter increased with increasing TPU concentration. With the increasing tip to collector distance, they were able to achieve better nanofiber morphologies and thinner nanofibers. They showed that decreasing the tip to collector distance caused insufficient solvent evaporation, which makes membrane-like surface instead of nanofibrous surface. Their results also revealed that increase in the feeding rate caused too much polymer deposition on the collector in a short time and that caused conjunction of nanofibers. They concluded that in case of thinner and smoother nanofiber production, 10% of TPU concentration, 20 cm tip to collector distance, 0.3 ml/h feeding rate, and 13 kV applied voltage are suitable [58].

Zhuo et al., also investigated the process parameters, including the applied voltage, feeding rate, and solution concentration. They synthesized PU from PU resin (number-average molecular weight = 180,000 g/mol), based on Poly(e-caprolactone) diol containing a 75% softsegment content, and 4000 soft-segment length by the bulk polymerization method. They found that 12.0 kV was a critical value for their synthesized PU. For preparing uniform PU nanofibers, diameters ranging from about 700 to 50 nm used 5.0 wt% PU/DMF solutions. However, when the applied voltages were increased to a high value, such as 20.0 or 25.0 kV, the diameters of nanofibers were not uniform and many loops were formed. Higher feeding rate (e.g., 0.1 mm/min) caused larger fibers compared to lower feeding rate (e.g., 0.06 mm/min), and smaller and uniform nanofibers were observed with lower feedings. The fiber diameters increased with the increasing solution concentration. They used five PU/ DMF solutions (3.0, 5.0, 7.0, 10.0, and 12.0 wt%), did not observe jet formation in the PU/ DMF solutions above 12.0 wt% because of the higher viscosity, whereas in a too diluted solution (e.g., <3.0 wt%), the jet broke into droplets and they observed electrospraying instead of electrospinning. Finally, they concluded that uniform PU nanofibers could be produced by using 5.0–7.0 wt% of PU/DMF solutions, applied voltages of 10–15 kV, and feeding rates of 0.06–0.08 mm/min [38]. They also observed that increasing the applied voltage caused stickier fibers [38].

Andrews et al. investigated some other spinning parameters for Poly(ether urethane), Tecoflex® SG-80A (Thermedics Polymer Products, Wilmington, MA), including flow rate,

A 32

24 Aspects of Polyurethanes

suitable [58].

stickier fibers [38].

affects fiber diameter significantly.

 factorial experimental design was used to investigate the tip to collector distance and applied voltage effect for PU nanofibers by Yanilmaz et al. [27]. They used TPU (PU, 270,000 g/mol) supplied from Coim Co. and chose THF for 10% (w/w) TPU solution, 5, 7.5, and 10 cm for tip-to collector distances and 10, 15, and 20 for applied voltages. From their design, it was seen that the distance and applied voltage had significant effects on the fiber diameter and it depends on the applied voltage and distance; furthermore, the interaction of these factors

Akçakoca Kumbasar et al., used DMF for TPU (Pellethane 2103-80AE) solutions and investigated the effect of TPU concentration (6, 8, 10, 12, and 14% (w/w) represented in **Figure 3**), tip to collector distance (8, 10, 12, and 15 represented in **Figure 4**), feeding rate (0.3, 1.5, 2.5, and 3.5 ml/h), and applied voltage (7, 10, 13 kV) on nanofiber diameters. Their results showed that 6% of TPU concentration is too low and 14% is too high for smooth nanofibers and as expected nanofiber diameter increased with increasing TPU concentration. With the increasing tip to collector distance, they were able to achieve better nanofiber morphologies and thinner nanofibers. They showed that decreasing the tip to collector distance caused insufficient solvent evaporation, which makes membrane-like surface instead of nanofibrous surface. Their results also revealed that increase in the feeding rate caused too much polymer deposition on the collector in a short time and that caused conjunction of nanofibers. They concluded that in case of thinner and smoother nanofiber production, 10% of TPU concentration, 20 cm tip to collector distance, 0.3 ml/h feeding rate, and 13 kV applied voltage are

Zhuo et al., also investigated the process parameters, including the applied voltage, feeding rate, and solution concentration. They synthesized PU from PU resin (number-average molecular weight = 180,000 g/mol), based on Poly(e-caprolactone) diol containing a 75% softsegment content, and 4000 soft-segment length by the bulk polymerization method. They found that 12.0 kV was a critical value for their synthesized PU. For preparing uniform PU nanofibers, diameters ranging from about 700 to 50 nm used 5.0 wt% PU/DMF solutions. However, when the applied voltages were increased to a high value, such as 20.0 or 25.0 kV, the diameters of nanofibers were not uniform and many loops were formed. Higher feeding rate (e.g., 0.1 mm/min) caused larger fibers compared to lower feeding rate (e.g., 0.06 mm/min), and smaller and uniform nanofibers were observed with lower feedings. The fiber diameters increased with the increasing solution concentration. They used five PU/ DMF solutions (3.0, 5.0, 7.0, 10.0, and 12.0 wt%), did not observe jet formation in the PU/ DMF solutions above 12.0 wt% because of the higher viscosity, whereas in a too diluted solution (e.g., <3.0 wt%), the jet broke into droplets and they observed electrospraying instead of electrospinning. Finally, they concluded that uniform PU nanofibers could be produced by using 5.0–7.0 wt% of PU/DMF solutions, applied voltages of 10–15 kV, and feeding rates of 0.06–0.08 mm/min [38]. They also observed that increasing the applied voltage caused

Andrews et al. investigated some other spinning parameters for Poly(ether urethane), Tecoflex® SG-80A (Thermedics Polymer Products, Wilmington, MA), including flow rate,

**Figure 3.** (a) 6%, (b) 8%, (c) 10%, (d) 12%, and (e) 14% w/w electropsun TPU nanofibers [58].

**Figure 4.** SEM images of the TPU nanofibers at tip-to collector distance: (a) 8, (b) 10, (c) 12, and (d) 15 cm [58].

relative spray height, spray distance, traverse speed, mandrel speed, grid voltage, and mandrel voltage. Dimethylacetamide (DMAc):2-butanone [methyl ethyl ketone (MEK)] were used in the ratio of 1:1.68 to obtain 12.5% w/v PU solution. Their results indicated that inter-fiber separation was significantly affected by flow rate, spray distance, grid voltage, and mandrel voltage, but not by relative spray height, traverse speed, and mandrel speed whereas fiber diameter was significantly affected by flow rate and mandrel voltage, there was no significant difference brought about by changes in relative spray height, spray distance, traverse speed, mandrel speed, and grid voltage. Void fraction was significantly affected by flow rate but not by relative spray height, spray distance, traverse speed, mandrel speed, grid voltage, and mandrel voltage. Fiber orientation on the external surfaces was significantly affected by traverse speed and mandrel speed, but not by flow rate, relative spray height, spray distance, grid voltage, and mandrel voltage. They indicated that the volumetric flow rate was the sole spinning parameter that affects the scaffold thickness [59].

Demir et al. prepared a segmented polyurethaneurea based on poly(tetramethylene oxide)glycol cylcoaliphatic diisocyanate, and unsymmetrical diamine. They used 2-methyl-1,5 diaminopentane (DAP), dibutylamine (DBA), DMF as solvents and studied electrospinning behavior of produced elastomeric polyurethaneurea copolymer in solution. They observed fiber diameters increased with the increasing solution concentration and lower concentration favored beads, and increased concentrations favored curly fibers. Salt addition increased the solution conductivity, which led to increase in mass flow. They found viscosity and solution temperatures were dominant factors and improving the fiber morphology was possible with increasing solution temperature, and it was quicker to electrospun these solutions compared to the solutions that were at room temperature [20].

The effect of tetraethylammonium bromide (TEAB) salt on the spinnability of polyurethane (PUR, Larithane LS 1086, aliphatic elastomer based on 2000 g/mol, linear polycarbonated diol, isophorone diisocyanate, and extended isophorone diamine) was investigated in the study of Cengiz et al. They used a roller electrospinning method. They found that the conductivity, viscosity, spinning performance increased with salt concentration. Also, solution viscosity decreased with shear rate. PU including 1.82 wt% TEAB gives the best spinning performance although 0.87 wt% TEAB is the optimum value related to fiber properties such as diameter, uniformity, and morphology given the ideal PU nanoweb structure [60]. In another study of Cengiz et al., they discussed the effects of 1,1,2,2 tetrachlorethylen (TCE), a non-solvent addition on the independent (electrical conductivity, dielectric constant, surface tension, and the rheological properties of the solution etc.) and dependent parameters (number of Taylor cones per square meter (NTC/m2 ), spinning performance for one Taylor cone (SP/TC), total spinning performance (SP), fiber properties such as diameter, diameter uniformity, non-fibrous area). The effect of non-solvent concentration on the dielectric constant, surface tension, rheological properties of the solution, and also spinning performance were statistically important. Beside, non-solvent concentration affects the quality of fiber and nanoweb structure [61].

Yalcinkaya et al., measured the jet current and jet life in roller electrospinning of PU (molecular weight 2000 g/mol, Larithane LS 1086; Novotex, Italy) in their study. They analyzed the relationships between jet current and jet life and number of Taylor cones/m2 (NTC/m2 ), spinning performance (SP), and fiber properties (diameter, non-fibrous area) and determined the effects of PU and TEAB concentrations on jet current and jet life. They observed that jet current increases with PU and TEAB concentration, while jet life decreases. NTC/m2 and spinning performance increased with jet current and decreased with jet life, and they observed that jet current movement gives an idea about jet life [62].

relative spray height, spray distance, traverse speed, mandrel speed, grid voltage, and mandrel voltage. Dimethylacetamide (DMAc):2-butanone [methyl ethyl ketone (MEK)] were used in the ratio of 1:1.68 to obtain 12.5% w/v PU solution. Their results indicated that inter-fiber separation was significantly affected by flow rate, spray distance, grid voltage, and mandrel voltage, but not by relative spray height, traverse speed, and mandrel speed whereas fiber diameter was significantly affected by flow rate and mandrel voltage, there was no significant difference brought about by changes in relative spray height, spray distance, traverse speed, mandrel speed, and grid voltage. Void fraction was significantly affected by flow rate but not by relative spray height, spray distance, traverse speed, mandrel speed, grid voltage, and mandrel voltage. Fiber orientation on the external surfaces was significantly affected by traverse speed and mandrel speed, but not by flow rate, relative spray height, spray distance, grid voltage, and mandrel voltage. They indicated that the volumetric flow rate was the sole spinning parameter that affects the scaffold

**Figure 4.** SEM images of the TPU nanofibers at tip-to collector distance: (a) 8, (b) 10, (c) 12, and (d) 15 cm [58].

thickness [59].

26 Aspects of Polyurethanes

Among above the traditional electrospinning method, an uncommon laser-heated electrospinning which is represented in (**Figure 5**), was used by Takasaki et al. and they investigated the effect of the spinning conditions including the applied voltage, the laser power, the laser irradiation point, and the laser beam width on the diameter of TPU microfibers. The average diameter of electrospun TPU fibers decreased with decreasing applied voltage and increasing laser power. A narrower laser beam reduced the variation in the fiber diameters. A PU microfiber with an average diameter of 2.4 μm and a coefficient of variation of 8% was obtained using a 0.9 mm wide laser beam [63].

**Figure 5.** Schematic diagram of laser-electrospinning system. Modified from Takasaki et al. [63].

### *4.2.1. Melt electrospinning*

Although most electrospinning researches are based on polymer solutions, there are also several researchers that use polymer melts. Generally, most electrospinning conditions also affect the molten polymer in electrospinning [9]. Melt electrospinning is especially attractive for tissue engineering scaffold manufacturing process, because it is possible to eliminate potentially cytotoxic solvents in contrast to electrospinning from a polymer solution. However, there are some technical challenges related to the need for a well-controlled high-temperature setup and there is a difficulty in developing an appropriate polymer. In the study of Karchin et al., a biodegradable and thermally stable PU, produced from 1,4-butanediamine and 1,4-butanediol, (CH2 ) 4 -content diisocyanates and polycaprolactone. These aliphatic PU formulations were used in the melt electrospinning [42]. A catalyst-purified PU based on 1,4-butane diisocyanate, polycaprolactone, and 1,4-butanediol in a 4/1/3 M ratio respectively, yielded a non-toxic polymer that could be electrospun from the melt. The authors concluded that this electrospun polymer contained point bonds between fibers and its mechanical properties were analogous in vivo soft tissues.

### **4.3. Collectors for electropsun PU nanofibers**

In most electrospinning setup, the collector plate is made out of a conductive material and covered by a piece of aluminum foil, which is electrically grounded so that there is a stable potential difference between the source and the collector [9]. Rotating drum with different diameters [47, 59, 64–69], parallel electrodes [70], rotating wire drum collector [71], rotating tube collector with knife-edge electrodes below [66], disc collector [72–74] array of counter-electrodes [75], rotating drum with sharp pin inside [76], ring collector placed in parallel [77] could be also used for different purposes [13]. Despite the random and centered collection of nanofibers on the collector plate, the most basic form of getting aligned nanofiber deposition is through the use of a rotating mandrel. Schematic representation is given in **Figure 6**. Beside aligned nanofibers, the texture of the fiber mesh may also be varied by using patterned collectors.

Andrews et al. deposited the PU nanofibers onto 3 mm diameter stainless steel mandrels that were pre-coated with a saturated NaCl solution to facilitate fiber formation and subsequent removal of the scaffold [59]. Pedicini and Farris used a grounded flat aluminum foil target for isotropic TPU fiber mats for tensile tests and infrared spectroscopy experiments, oriented electrospun TPU nanofiber samples for IR dichroism studies were collected onto a rotating stainless steel drum [64]. Two types of electrodes with tines were used in the study of Banuskeviciute et al. [79]. The electrode consists of eight separate plates. In every plate, tines were set at equal distances. Tines were different by the width and shape of every electrode. They showed that the type of electrode had an influence on the structure of the electrospun TPU mats, but not on the diameter of formed fibers. Rotating mandrels could be used for electrospun vascular graft for the regeneration of blood vessel (**Figure 6b**). Theron et al. [65] spun small diameter vascular graft prototypes (1.6 mm nominal ID) using the apparatus including rotating/translating mandrel. Tubes were removed from the mandrels by swelling in EtOH and dried. Thandavamoorthy et al. [80] self-assemble electrospun PU nanofibers into honeycomb patterns on the collector surface. Residual charges on the collected fibers and the electrical property of the collector screen influenced the self-alignment of fibers. They electrospun PU nanofibers over these substrates while keeping all other process parameters constant. When cotton, a natural fiber with very poor electrical conducting properties, was used as the collecting surface, a 3-D honeycomb pattern deposition was produced (**Figure 7**). Similar results were observed with the glass substrate.
