Dynamic mechanical relaxations of electrospun poly(acrylonitrile-co-methyl acrylate) nanofibrous yarn

Abstract

The phase structure and dynamic mechanical properties of poly(acrylonitrile-co-methyl acrylate) (P(AN-co-MA)) nanofibers collected in the form of twisted yarn via the two-nozzle conjugated electrospinning method were investigated to study the effects of solution concentration and take-up velocity on the relaxation behavior of nanofibers yarn. The wide-angle X-ray diffraction analyses of P(AN-co-MA) nanofibers show a two-phase structure of nanofibers consisting of crystalline and amorphous phases and polymorphic transition from hexagonal to orthorhombic. Heating P(AN-co-MA) nanofibers at over the glass transition temperature led to an increased degree of both crystallinity and crystallite size with no polymorphic change. Three transitions (tan δ peaks) were observed in nanofibrous yarn prepared at different spinning dope concentrations and take-up speeds, except for the specimen prepared at a concentration of 14 wt% and collecting speed of 8 cm/min, wherein no α transition was observed due to improved molecular orientation. The temperature dependence of the dynamic Young’s modulus of nanofibrous yarn at different spinning dope concentrations was mainly affected by the diameter of the nanofiber as the morphological property and molecular orientation. Take-up speed was found to affect the γ and α transitions more than the β transition. Moreover, the maximum storage modulus was obtained at a take-up speed of 8 cm/min at all over recorded temperatures.

Keywords

poly(acrylonitrile-co-methyl acrylate)thermal transition storage modulus twisted nanofibrous yarn electrospinning X-ray diffraction dynamic mechanical analysis

Polyacrylonitrile (PAN) is an important and versatile polymer used to produce a large variety of products, including fibers for textiles,¹ as a precursor for carbon fiber,^2–4 fiber-reinforced composites,^5,6 ultrafiltration membranes,^7,8 proton exchange membranes,^9,10 and metal ion absorbent materials,^11,12 owing to the unique properties of PAN, such as low density, thermal stability, and high strength and modulus of elasticity, which have made it a preferred polymer in many applications.¹ Many studies have been published on the properties of PAN-based polymers influenced by structural morphology,^13–15 phase structure,^16,17 chain conformation,^18,19 and molecular dynamics in the forms of fibers, films, powders, and even single crystals.^20–22 The relaxations and/or thermal transitions in PAN have also been extensively investigated by using instrumental techniques, including dynamic mechanical analyses (DMAs), dielectric measurements, infrared absorption spectroscopy, and wide-angle X-ray diffraction (WAXD) at elevated temperatures.^21,22 The DMA curves of PAN often show more than one (tan δ) peak. In the nomenclature of DMA, the α relaxation represents the tan δ peak at glass transition temperature T_g of the polymer associated with backbone segmental motion, whereas the β peak arises from sub-T_g relaxations, usually of side chain or bulky group fluctuations.²³ At the low end of the temperature, a third relaxation, often close to room temperature, termed γ relaxation, was rarely but also reported. In Table 1, different interpretations are summarized about the nature of the transitions occurred in PAN in the literature.^{21,22,24–28} Furthermore, according to WAXD analyses, authors proposed several structural models of PAN polymer, including the one-phase paracrystalline structure; the two-phase system, such as the conventional semicrystalline polymer; the two-phase structure composed of both amorphous and the paracrystalline phases; the three-phase structure with two amorphous phases and a crystalline phase; and another three-phase model consisting of an amorphous, a paracrystalline, and a crystalline phase. Nowadays, electrospun nanofibers have a very high specific surface area due to their nanoscale diameter, but they are still manipulatable without resorting to microscopy due to their macroscale length, thus becoming a nanoscale material of industrial and scientific interest. To facilitate the diverse applications of nanofibers, such as carbon fibers, wound dressing, tissue engineering, filtration, separation membranes, and ion exchange membranes, it is necessary to have a better understanding of the relationship between the properties, processing, and microstructure of nanofibers and various affecting factors in practical conditions.^29,30 As one of the most extensively used polymers, PAN nanofibers have been extensively investigated to evaluate the effects of drawing, heat treatment, and surface modification on their performance.^2,31–36 However, knowledge about the phase-transition PAN-based nanoscale fibers, such as poly(acrylonitrile-co-methyl acrylate) (P(AN-co-MA)), which is widely used in industry, has not been clearly understood. Single nanofibers, because of their tiny diameter, are known to be significantly stronger in tension. However, the resulted nanofibrous yarn, as described in the next section, exhibit a disappointingly low strength.³⁷ Clearly, the superior performance of the nanofibers failed to be transferred into the corresponding yarn performance. It is, hence, critically important to understand the structural property transfer from individual nanofibers into the collected mass, the yarn. Therefore, this work represents such an attempt to connect the phase structure and dynamic mechanical behavior of electrospun P(AN-co-MA) nanofibers in the form of a twisted yarn structure and attempts to study the effects of solution concentration and take-up speed as important parameters of the electrospinning process on mechanical relaxations of nanofibers with DMA and hot-stage WAXD techniques.
Table 1.
Various explanations for transitions in the polyacrylonitrile (PAN) structure

Authors γ relaxation β relaxation α relaxation

Cotton and Schneider²⁴ – Unsure about the nature (88℃)^a Glass transition (118℃)

Kimmel and Andrews²⁵ – Glass transition from loosening of the van der Waals bonding (90℃) Glass transition from the breaking of the dipole–dipole association network (140℃)

Okajima et al.²⁶ – Amorphous glass transition (109℃) Transition in more ordered domains (160℃)

Minami et al.²⁷ – The motion in the paracrystalline domains (110℃) Glass transition due to molecular motion in the amorphous regions (160℃)

Kenyon and Rayford²⁸ Secondary transition of the amorphous phase (70℃) The release of dipole–dipole bonds in the paracrystalline phase (105℃) Transition in the amorphous phase (140℃)

Bashir²¹ – Glass transition from mesophase glass to the mesophase melt (100℃) Glass transition from amorphous glass to the amorphous rubber (150℃)

Sawai et al.²² Local mode motions of syndiotactic and short isotactic sequences, which take a planar zigzag conformation in paracrystalline phases (25℃) The molecular motion of helical sequences in paracrystalline phases (100℃) • The micro-Brownian motion in amorphous regions (for gel films in 150℃) • First-order crystal/crystal transition (for ultradrawn at-PAN in 150℃)

a
Transition temperatures (tan δ peaks) for PAN from dynamic mechanical analysis.

Experimental details

Materials

Industrial polyacrylonitrile copolymer, P(AN-co-MA) containing 5 mol% methyl acrylate, and highly pure dimethylformamide (DMF, 99%) were respectively supplied by Iran Polyacryle and Merck companies. The weight average molecular weight and the number average molecular weight of the received P(AN-co-MA) were $\bar{M_{w}} =$ 100,000 g/mol and $\bar{M_{n}} =$ 70,000 g/mol, respectively. All solutions of P(AN-co-MA) in DMF were prepared under constant mixing by a magnetic mixer (AREX Heating magnetic stirrer 230 V/50–60 Hz, Model #: F20500163) at room temperature.

Two-nozzle conjugated electrospinning

Figure 1(a) schematically illustrates the setup used for two-nozzle conjugated (TNC) electrospinning and yarn formation. It consists of a high-voltage power supply, two syringe pumps, a conductive hemisphere, and a take-up unit. Two different charged nozzles were placed at 14 cm distance opposite each other and a neutral surface was placed in the middle of the electric field formed between two syringe needles. The electrons on the surface of the conductive hemisphere are displaced in such a way that half of the surface became positively charged and the other half negatively charged due to columbic forces among charges. The jets of the polymer solution ejected from the charged nozzles with a feed rate of 0.96 ml/h moved slightly towards the part of the hemisphere surface with an opposite charge and were collected on it. Electrospun nanofibers were entangled with the end of a piece of yarn subjected in their path. With rotating and taking up the yarn, a spinning triangle is formed and, as shown in Figure 1(b), twisted continuous nanofibrous yarn is fabricated. The applied voltages of 7 and 9 kV, solution concentrations of 12, 14, and 17 wt%, and take-up speeds of 2.7, 8, and 11 cm/min were chosen to characterize the nanofibrous yarns based on suitable conditions for preparing uniform nanofibers with no beads. The average diameter analysis of nanofibers showed the values of 255 ± 34, 392 ± 41, and 625 ± 25 nm measured at concentrations of 12, 14, and 17 wt%, respectively.
Figure 1.
(a) The setup used to collect continuous twisted yarn composed of nanofibers. (b) A specimen of collected yarn and its constitutive nanofibers.

X-ray diffraction (WAXD)

A P(AN-co-MA) powder mount was prepared on a glass sample holder using front-loading techniques and was placed on a Bruker Advance Diffractometer equipped with a LynxEye Solid State Detector. The diffraction pattern was collected in continuous mode from 2θ = 10–50° using Cu radiation (λ = 1.5406 Å), a step size of 0.02° per step, and a count time of 2 seconds per step. Nanofiber samples electrospun at different applied voltage were cut into small pieces and placed in a pellet press with a diameter of 1 cm. A small amount of Quartz (SiO₂, approximately 5 mg) was used as an internal standard to correct for the sample displacement errors of the variation in the pellet thickness. The pellet was placed on an alumina sample holder and then loaded into the environmental sample chamber on the Inel powder diffractometer equipped with a CPS 120 position sensitive detector (PSD). The diffraction patterns were collected from 10 to 60° 2θ using Co radiation (λ = 1.7903 Å), a step size of 0.03°/step, and a count time of 4 seconds per step. An initial pattern was collected in reflective mode at 25℃ for 5 minutes. The temperature in the environmental chamber was increased to 100℃ using a rate of 2℃ per minute and a tolerance of 2℃. The sample was allowed to equilibrate at 100℃ for 10 minutes. After equilibration, a diffraction pattern of the sample was collected for 5 minutes. The diffraction data was evaluated using Jade 6.0 software and peaks associated with internal standard were subtracted from the patterns.

Dynamic mechanical analysis

DMA measurements were made on a DMA Q800 instrument operated at frequencies of 1 Hz in the temperature range of 20–160℃. The samples were deformed dynamically in a tensile mode. The effects of concentration and take-up speed on the variations of the dynamic storage modulus (E′) and loss tangent (tan δ) with temperature were measured at the lowest frequency of 1 Hz using the heating rate of 2℃/min. For 16 hours prior to performing dynamic mechanical tests, the samples were in a conditioning room at a temperature of 22 ± 0.5℃ and relative humidity of 65 ± 1%.

Results and discussion

WAXD analyses of P(AN-co-MA) nanofibers

Before discussion of the relaxation behavior of nanoscale fibers, the structures of P(AN-co-MA) powder and nanofibers electrospun at a concentration of 14 wt% and two different applied voltages of 7 and 9 kV were characterized. The room temperature WAXD pattern of P(AN-co-MA) powder (Figure 2) exhibits two peaks of d₁ = 5.27 Å and d₂ = 3.05 Å, commonly reported for diffraction patterns of PAN polymer as equatorial peaks,^{13–15,17,21,38,39} suggesting a similar supermolecular structure for both PAN and P(AN-co-MA) polymers. Thus, it seems to be conceivable to describe the phase structure of P(AN-co-MA) using the same model as that previously proposed for PAN.⁴⁰ Initial inspection of powder pattern shows that the ratio of d₁/d₂ is $\sim \sqrt{3} / 1$ , which is characteristic of the hexagonal packing of chains with the rod diameter of 2d₁/ $\sqrt 3$ = 6.085 Å.^13,17,38 Accordingly, P(AN-co-MA) macromolecules can be thought of as alternating longer fragments of a molecular rod of 0.6 nm diameter and shorter fragments fitting within a cylinder of a smaller diameter. The mean length of the ‘thicker’ fragments of such non-uniform ‘molecular rods’ would obviously depend on the number of ‘foreign’ units in the chain.⁴⁰ As summarized in Table 2, it is possible to investigate the PAN copolymer powder pattern (Figure 2) in more detail in two ways, based on models of hexagonal and pseudo-hexagonal (rectangular) unit cells.³⁸ As mentioned by Bashir,²¹ the presence of two peaks on the equator in the powder pattern indicates lateral ordering or periodicity, but the absence of proper meridional peaks (00l) indicates the absence of chain-axis order. This finding, as well as the broad amorphous peak observed in the diffraction pattern, demonstrates a two-phase structure of P(AN-co-MA) polymer of paracrystalline and amorphous phases. The crystallization index defined as the degree of order using the Gupta and Singhal method³⁹ was determined with the value of 36.6% for PAN copolymer.
Figure 2.
Wide-angle X-ray diffraction (WAXD) pattern of poly(acrylonitrile-co-methyl acrylate) (P(AN-co-MA)).

Table 2.
Characterization of diffraction pattern of poly(acrylonitrile-co-methyl acrylate) powder

Cell type d-spacing (Å) Miller indices (hkl) Average lattice parameters (Å)

a^a b^a c^a

Hexagonal 5.2701 (100) 6.0947 6.0947 –

3.0519 (110)

Pseudo-hexagonal (rectangular) 5.2701 (200) 10.5402 6.1040 –

3.0519 (020)

a
a, b, and c are the dimensions of the unit cell of crystalline lattice obtained with Jade 6 software.

Figure 3 shows WAXD patterns (recorded at 25℃ and 100℃) of P(AN-co-MA) nanofibers electrospun at two different applied voltages. Characterization of diffraction patterns of nanofibers (Table 3) reveals the appearance of new peaks in comparison to the X-ray pattern of P(AN-co-MA) powder, which may be an evidence of polymorphism variation in nanoscale fibers. The presence of equatorial peaks in the X-ray patterns of P(AN-co-MA) nanofibers accompanied by off-axis (hkl) and meridional (00l) peaks, which are normally found in crystalline polymers,²¹ exhibits two-phase morphology including crystalline and amorphous phases in nanofibers. Further analyses of WAXD patterns conducted to consider the orthorhombic polymorph for P(AN-co-MA) nanofibers with the lattice parameters mentioned in Table 3 show results close to the unit cell dimensions suggested by Bashir Z.³⁸ The main difference in WAXD profiles recorded at different temperatures is related to the broad and diffuse peak centered at 2θ≈20–21 in room temperature diffraction patterns of samples, which is sharpened and shifted to slightly larger d-spacing (lower 2θ) on heating. The crystallite size corresponding to the mentioned peak was increased from 40 to 109 Å and from 42 to 120 Å for applied voltages of 7 and 9 kV, respectively. This structural change has a prominent effect on the enhanced value of crystallinity with increasing temperature, especially at a higher applied voltage. Heating PAN fibers to over glass transition temperature resulted in releasing more and more nitrile groups from their bound state due to the dipole–dipole interaction of the nitrile groups and the increasing segmental mobility of molecular chains. By providing sufficient time for polymeric chains, because of the motion of molecular chains in amorphous regains, some smaller crystals might be disturbed while other larger crystals were formed so that the crystallinity of fibers was improved.^1,41 Moreover, the influence of short chain fragments of AN-MA type on the structural changes on annealing would consist of (1) facilitating parallelization of the molecular rods comprising ‘pure’ AN sequences (the ‘hinge’ effect); and (2) increasing the possibility of optimal pairing between nitrile dipoles of neighboring chains.⁴⁰ The peak of d = 2.09 Å (hkl: 032) observed at the applied voltage of 7 kV was vanished by the increasing voltage. Although this off-axis peak was not proper on the meridian, it may be a sign of chain-axis order. It seems that columbic forces and bending instability encouraged with voltage disturbed polymer molecules to orient in the chain-axis order. The results of crystallinity at room temperature explained the low crystallinity value of electrospun fibers and the inverse effect of applied voltage on the crystallization behavior of nanofibers. Low crystallinity of electrospun PAN nanofibers is a result of rapid solvent evaporation and fast solidification of polymer due to the high specific surface area of nanofibers.⁴² Crystallinity has been reported to vary with the applied voltage. The crystallinity of poly(L-lactic acid) (PLLA) nanofibers at a concentration of 5% decreased with increasing voltage from 15 to 25 kV; on the contrary, at a concentration 8% crystallinity increased with increasing voltage.⁴³ Therefore, there is an optimum electric field strength for a given polymer concentration at which a maximum value of crystallinity is obtained. As is known, applied voltage V (kV) provides the surface charge on the electrospinning jet. A higher voltage will lead to greater stretching and accelerating of the viscoelastic solution due to the greater columbic forces between charges in the jet, as well as the stronger electric field. Although electrostatic forces and bending instability occurred during electrospinning, they apply high elongated forces on the jet and thus may cause the polymer molecules to be more ordered and induce a greater crystallinity in the fiber, but the gap distance determines the time of travel for the jet and therefore influences its stretching level.^30,44 The lower distance of the two syringe needles at the setup used to collect nanofibrous yarn in comparison with the conventional electrospinning setup led to the decreased flight time of the electrospinning jet intensified at a higher voltage, due to enhanced acceleration of fibers. This fact was corroborated by the increasing of the cross-sectional area of nanofibers with voltage observed in previous work.⁴⁵ Since the orientation of the polymer molecules will take some time, the reduced flight time means that the fibers will be deposited before the polymer molecules have sufficient time to align themselves. On the other hand, the decrement of flight time would probably cause incomplete evaporation of solvent within nanofibers and, as a result, it allows the molecular chains to relax in the collected nanofibers and disturb the ordering of molecules.⁴⁶ Therefore, given sufficient flight time, the crystallinity of the fiber will improve with higher voltage, as suggested by Zhao and coworkers.⁴⁷ The low degree of crystallinity of nanofibers measured in this study concurs with the low tensile strength of as-spun nanofibers,⁴⁷ which improved with drawing and heat treatment, as previously studied. The tensile strength and modulus of hot-drawn PAN nanofiber yarn with a draw ratio of 4 in a boiling water bath increased from 50 to 350 MPa and 0.5 to 7 GPa, respectively; similarly, crystallinity increased with increasing the draw ratio.^48,49 Hence, the X-ray results reveal two important facts as follows. Electrospinning can induce a hexagonal-to-orthorhombic transition owing to electrostatic forces applied on the jet, and there is enhancement of crystallinity of nanofibers with temperature and no evidence of polymorphic change on heating of nanofibers. These results confirm others’ findings in the literature.^21,38 Accordingly, an applied voltage of 9 kV was chosen to prepare DMA test samples in different concentrations and at different take-up speeds. This is because of better stability of the electrospinning process and a more significant response of the crystallization manner with temperature and, thus, preferable study of phase structure transitions at this voltage.
Figure 3.
Wide-angle X-ray diffraction patterns recorded at different temperatures for poly(acrylonitrile-co-methyl acrylate) nanofibers collected at two applied voltages of (a) 7 kV and (b) 9 kV.

Table 3.
Wide-angle X-ray diffraction analyses of poly(acrylonitrile-co-methyl acrylate) nanofibers electrospun at different applied voltages

Voltage (kV) Temperature (^oC) 2θ d-spacing (Å) (hkl) Average lattice parameters (Å) Crystallinity (%)

7 25 21.07, 29.80, 41.05, 44.05, 50.72 4.89, 3.48, 2.55, 2.38, 2.09 (200), (211), (002), (420), (032) – 10.66

100 19.40, 29.58, 40.74, 43.91, 50.51 5.31, 3.50, 2.57, 2.39, 2.09 (200), (211), (002), (420), (032) a = 10.614 b = 10.907 c = 5.1405 18.55

9 25 20.07,40.47, 43.62 5.13, 2.58, 2.408 (200), (002), (420) – 6.54

100 19.31, 29.26, 33.39, 40.47, 43.58 5.33, 3.54, 3.11, 2.58, 2.41 (200), (211), (230), (002), (420) a = 10.67 b = 11.48 c = 5.17 32.47

Dynamic behavior of poly(acrylonitrile- co-methyl acrylate) nanofibrous yarns

Thermal transition

Figures 4 and 5 show the effects of concentrations and take-up speeds on the temperature dependences of the storage modulus (E′) and tan δ for P(AN-co-MA) nanofibrous yarns electrospun at 9 kV and measured at 1 Hz in the temperature range of 20–150℃. The overall trends of thermal transition peaks enhanced in tan δ plots are similar to PAN polymer reported in the literature.^{21,22,24–28} The transition temperatures shifted to a lower temperature for P(AN-co-MA) nanofibers in comparison to PAN homopolymer, owing to the presence of methyl acrylate comonomer in P(AN-co-MA). The introduction of such small amounts of a comonomer greatly enhances the internal mobility of polymer segments, reducing the sequences of acrylonitrile molecules capable of interacting with neighboring sequences.^39,40
Figure 4.
Dynamic mechanical analysis results of poly(acrylonitrile-co-methyl acrylate) nanofibrous yarns collected at a take-up speed of 8 cm/min with different concentrations of (a) 12 wt%, (b) 14 wt%, and (c) 17 wt% (storage modulus: dashed line - - -; tan δ: solid line —).

Figure 5.
Dynamic mechanical analysis results of poly(acrylonitrile-co-methyl acrylate) nanofibrous yarns prepared at a concentration of 14 wt% with take-up speeds of (a) 2.7 cm/min, (b) 8 cm/min, and (c) 11 cm/min (storage modulus: dashed line - - -; tan δ: solid line —).

The temperatures of thermal transitions that occurred in P(AN-co-MA) nanofibrous yarns are summarized in Table 4. As mentioned in the introduction, the sources of γ, β, and α transitions were attributed, respectively, to local mode motions of syndiotactic and short isotactic sequences in paracrystalline phases, molecular motion in paracrystalline phases, and transition in amorphous regions. However, the WAXD analyses of PAN copolymer nanofibers (Table 3) show that the ordered phases can be in the form of a crystalline region and thus γ and β transitions in P(AN-co-MA) nanofibers are related to motions in crystalline phases. The DMA analysis (Table 4) indicates the increased transition temperatures at ordered regions (γ and β) with concentration from 12 to 14 wt% at a constant take-up speed of 8 cm/min. It is well known that an increase in the concentration will result in more molecular chain entanglements and thus further thermal energy is required to transfer polymer chains from a glassy state to a rubbery state, that is, the temperature of transitions shifts to a higher value. The vanishing of the α transition related to molecular motions in the amorphous phase at the concentration of 14 wt% might be attributed to improved molecular orientation for this specimen, as subsequently described. More enhancement of concentration to 17 wt% had an insignificant effect on the temperature of γ and β transitions and led to the appearance of α transition at higher value in comparison to the concentration of 12 wt%. This indicates that an increased entanglement of polymer chains at high concentration promoted an amorphous region with less mobility of polymer chain segments.
Table 4.
Temperatures of thermal transitions acquired from peaks of tan δ plots for poly(acrylonitrile-co-methyl acrylate) nanofibrous yarns prepared at different concentrations and take-up speeds

Concentration (wt%) Take-up speed (cm/min) Transition temperature (^oC)

α β γ

12 8 121 76 41

14 – 85 58

17 138 86 57

14 2.7 98 83 51

8 – 85 58

11 123 88 48

The evaluation of take-up speed on thermal transitions in P(AN-co-MA) nanofibers (Table 4) reveals the slight influence of collecting speed on the β transition compared with the α and γ transitions, demonstrating the impressive variation of the amorphous phase and chain conformation with take-up speed. The increase of take-up speed has two main outcomes, namely the reduction of number of turns per unit length of yarn (twist level) by considering fixed rotational speed of the take-up unit and the enhancement of drawing force applied on nanofibrous yarn. The consequence of these effects was less deviation of nanofibers from the longitudinal axis of yarn and, as a result, a more effective drawing force was applied on nanofibers and improved orientation of molecular chains was achieved. The increase of molecular orientation amplified by the hinge action of short chain fragments facilitated parallel alignment of chain segments, giving rise to the presence of CN groups not involved with intramolecular interactions. The existence of such ‘unblocked’ CN groups especially available for intermolecular interactions would, along with the parallel chain alignment, promote the formation of dipole pairs.⁴⁰ Thus, the effect of such intermolecularly bonded nitrile groups would be an increase of order and a decrease of chain segment mobility, confirmed by the absence of a thermal transition in the amorphous phase at a take-up speed of 8 cm/min for a concentration of 14 wt% and the increase of the temperature of γ transitions, to some extent, respectively. Further increase of the take-up velocity from 8 to 11 cm/min resulted in a significant drop of γ transition temperature, a slight increase of transition temperature in the ordered phase (β), and a transition occurring in the amorphous phase at higher temperatures at the take-up speed of 2.7 cm/min, reflecting the ascending effect of molecular orientation on transition temperature, as mentioned above. The appearance of the α transition and decreasing of the γ transition temperature at a collecting speed of 11 cm/min indicate the disturbing effect of increased take-up speed on molecular orientation. This means that a molecular chain requires some time to align itself in the direction of applied force, which is lost with additional take-up speed, as previously reported.^31,50,51

Dynamic Young’s modulus

The dynamic Young’s modulus (E′, or the real storage modulus that depends on the elastic part of deformation)⁵² οf all samples prepared at different concentrations (Figure 6) decreased slowly up to the γ transition temperature and then more rapidly with increasing temperature due to the onset of significant molecular motions. The overall trend of storage modulus versus temperature is similar to that for PAN polymers reported in the literature.⁵³
Figure 6.
Storage modulus versus temperature of poly(acrylonitrile-co-methyl acrylate) nanofibrous yarns collected at (a) different concentrations of 12, 14, and 17 wt% at a take-up speed of 8 cm/min and (b) different take-up speeds of 2.7, 8, and 11 cm/min at a concentration of 14 wt%.

The response of the storage modulus at ambient conditions is in agreement with the reported results,⁴² which showed decreasing PAN nanofiber from 2.8 µm to about 100 nm causes an increase in elastic modulus from 0.36 to 48 GPa, with the largest increase for nanofibers smaller than 250 nm. The trend of the storage modulus with concentration and take-up speed might be interpreted as follows. Since the take-up and rotational speeds used to collect yarn are the same for all samples prepared at various concentrations (Figure 6(a)), the twist level can be considered equal among specimens, 1916 TPM (twist per meter) at a collecting speed of 8 cm/min. On the other hand, the diameter analysis of nanofibers shows the values of 255 ± 34, 392 ± 41, and 625 ± 25 nm measured at concentrations of 12, 14, and 17 wt%, respectively. A decrease in fiber diameters leads to an increase of twist angle due to a reduction of torsional and flexural rigidity of fibers, which plays a part in the arrangement of fibers in the yarn, that is, with decreasing fiber diameter from 625 to 255 nm, the flexural rigidity decreases 97%, and the surface-to-volume ration increases 145%. Moreover, when tension is applied to the twisted nanofibrous yarn during collection on the take-up roller, transverse compressive forces induced on the yarn would encourage cohesion of nanofibers in the yarn structure, especially for finer nanofibers. This phenomenon was demonstrated through the reduced area of cross-section at a lower concentration at fixed take-up speed and twist level in previous work.⁴⁵ Thus, the increase of twist angle and cohesion among fibers for finer nanofibers lead to a greater resistance of nanofibers to slippage,⁵⁴ which was intensified by enhancing the available surface of fibers to contact each other owing to the increase of surface-to-volume ratio at a lower diameter of nanofibers. Hence, the diameter of nanofibers as the morphological property influenced the dynamic Young’s modulus of nanofibrous yarn. As a result, finer nanofibers have a higher storage modulus. However, at temperatures greater than the γ transition, the storage modulus at concentration of 14 wt% was placed at a higher level than that of the concentration of 12 wt% because of the small amount of chain entanglements in nanofibers electrospun at low concentration and improved molecular orientation for this sample, which reduces macromolecular mobility in the crystalline phase, as mentioned earlier. Increasing the concentration to 17 wt% causes an insignificant effect on the γ, β, and α transitions as well as increasing the nanofiber diameter and leads to the lowest storage modulus for this sample. Accordingly, a concentration of 14 wt% was chosen for different take-up speeds. Figure 6(b) shows that the maximum value of the storage modulus was acquired at a take-up speed of 8 cm/min in the whole range of recorded temperatures. In regards to the decrease of twist level with increasing take-up speed, it might be stated that at a low twist level (collecting speed of 11 cm/min) the effect of fiber cohesion outweighs that of fiber obliquity. This gives rise to an increase in the yarn modulus, whereas at a high twist level (collecting speed of 2.7 cm/min) there is a further increase in cohesion that no longer produces an increase in the modulus because of the overwhelming effect of fiber obliquity.⁵⁴

Conclusion

In this study, in order to understand the structural properties the relationship between nanofiber and yarn, we investigated the effects of electrospinning parameters, including applied voltage, solution concentration, and take-up velocity, on the microstructure and dynamic behavior of P(AN-co-MA) nanofibrous yarn. The WAXD results reveal that electrospinning can induce a hexagonal-to-orthorhombic transition in nanofibers owing to electrostatic forces. Also, there is an enhancement of crystallinity with increasing temperature and no evidence of polymorphic change on heating of nanofibers, especially at a higher applied voltage of 9 kV. Thermal transitions expressed in terms of tan δ peaks occurred at a lower temperature in P(AN-co-MA) nanofibers than in PAN, owing to the flexible conformation of chain fragments of AN-MA type that gives rise to an increase of the effective mobility of chain segments. DMA results showed that there is the same trend in the nanofiber yarn dynamic Young’s modulus with increasing temperature. Also, it was shown that the fineness and molecular orientation of nanofibers enhanced the PAN nanofiber yarn storage modulus. The results of WAXD and DMA may have implications for the structural properties of PAN nanofiber as a precursor to PAN carbon nanofiber.

Authors	γ relaxation	β relaxation	α relaxation
Cotton and Schneider²⁴	–	Unsure about the nature (88℃)^a	Glass transition (118℃)
Kimmel and Andrews²⁵	–	Glass transition from loosening of the van der Waals bonding (90℃)	Glass transition from the breaking of the dipole–dipole association network (140℃)
Okajima et al.²⁶	–	Amorphous glass transition (109℃)	Transition in more ordered domains (160℃)
Minami et al.²⁷	–	The motion in the paracrystalline domains (110℃)	Glass transition due to molecular motion in the amorphous regions (160℃)
Kenyon and Rayford²⁸	Secondary transition of the amorphous phase (70℃)	The release of dipole–dipole bonds in the paracrystalline phase (105℃)	Transition in the amorphous phase (140℃)
Bashir²¹	–	Glass transition from mesophase glass to the mesophase melt (100℃)	Glass transition from amorphous glass to the amorphous rubber (150℃)
Sawai et al.²²	Local mode motions of syndiotactic and short isotactic sequences, which take a planar zigzag conformation in paracrystalline phases (25℃)	The molecular motion of helical sequences in paracrystalline phases (100℃)	• The micro-Brownian motion in amorphous regions (for gel films in 150℃) • First-order crystal/crystal transition (for ultradrawn at-PAN in 150℃)

Cell type	d-spacing (Å)	Miller indices (hkl)	Average lattice parameters (Å)
Hexagonal	5.2701	(100)	6.0947	6.0947	–
3.0519	(110)
Pseudo-hexagonal (rectangular)	5.2701	(200)	10.5402	6.1040	–
3.0519	(020)

Voltage (kV)	Temperature (^oC)	2θ	d-spacing (Å)	(hkl)	Average lattice parameters (Å)	Crystallinity (%)
7	25	21.07, 29.80, 41.05, 44.05, 50.72	4.89, 3.48, 2.55, 2.38, 2.09	(200), (211), (002), (420), (032)	–	10.66
100	19.40, 29.58, 40.74, 43.91, 50.51	5.31, 3.50, 2.57, 2.39, 2.09	(200), (211), (002), (420), (032)	a = 10.614 b = 10.907 c = 5.1405	18.55
9	25	20.07,40.47, 43.62	5.13, 2.58, 2.408	(200), (002), (420)	–	6.54
100	19.31, 29.26, 33.39, 40.47, 43.58	5.33, 3.54, 3.11, 2.58, 2.41	(200), (211), (230), (002), (420)	a = 10.67 b = 11.48 c = 5.17	32.47

Concentration (wt%)	Take-up speed (cm/min)	Transition temperature (^oC)
12	8	121	76	41
14	–	85	58
17	138	86	57
14	2.7	98	83	51
8	–	85	58
11	123	88	48

Footnotes

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The first author would like to acknowledge the financial support of Isfahan University of Technology.

References

Masson

. Acrylic fiber technology and applications, New York: CRC Press, 1995.

Sutasinpromprae

Jitjaicham

Nithitanakul

et al.

Preparation and characterization of ultrafine electrospun polyacrylonitrile fibers and their subsequent pyrolysis to carbon fibers. Polym Int 2006; 55: 825–833.

Sui

Sun

Yang

et al.

Highly aligned polyacrylonitrile-based nano-scale carbon fibres with homogeneous structure and desirable properties. Compos Sci Technol 2013; 87: 77–85.

Ren

Vancso

. Process optimization and empirical modeling for electrospun polyacrylonitrile (PAN) nanofiber precursor of carbon nanofibers. Eur Polym J 2005; 41: 2559–2568.

Tabil

Panigrahi

. Chemical treatments of natural fiber for use in natural fiber-reinforced composites: a review. J Polym Environ 2007; 15: 25–33.

Hou

et al.

Assembly of well-aligned multiwalled carbon nanotubes in confined polyacrylonitrile environments: electrospun composite nanofiber sheets. J Am Chem Soc 2004; 126: 15754–15761.

Zhao

Zhang

et al.

Nanofiltration membrane prepared from polyacrylonitrile ultrafiltration membrane by low-temperature plasma: I. Graft of acrylic acid in gas. J Membr Sci 2004; 232: 1–8.

Scharnagl

Buschatz

. Polyacrylonitrile (PAN) membranes for ultra-and microfiltration. Desalination 2001; 139: 191–198.

Wang

Yue

Economy

. Preparation of proton-conducting composite membranes from sulfonated poly (ether ether ketone) and polyacrylonitrile. J Membr Sci 2007; 291: 210–219.

10.

Carter

Wycisk

Yoo

et al.

Blended polyphosphazene/polyacrylonitrile membranes for direct methanol fuel cells. Electrochem Solid State Lett 2002; 5: A195–A197.

11.

Saeed

Haider

et al.

Preparation of amidoxime-modified polyacrylonitrile (PAN-oxime) nanofibers and their applications to metal ions adsorption. J Membr Sci 2008; 322: 400–405.

12.

Kampalanonwat

Supaphol

. Preparation and adsorption behavior of aminated electrospun polyacrylonitrile nanofiber mats for heavy metal ion removal. ACS Appl Mater Interface 2010; 2: 3619–3627.

13.

Bohn

Schaefgen

Statton

. Laterally ordered polymers: polyacrylonitrile and poly (vinyl trifluoroacetate). J Polym Sci 1961; 55: 531–549.

14.

Hobson

Windle

. Crystalline structure of atactic polyacrylonitrile. Macromolecules 1993; 26: 6903–6907.

15.

Liu

Ruland

. X-ray studies on the structure of polyacrylonitrile fibers. Macromolecules 1993; 26: 3030–3036.

16.

Hayakawa

Nishi

Arisawa

et al.

Dielectric relaxation in the paracrystalline phase in polyacrylonitrile. J Polym Sci Part A-2 Polym Phys 1967; 5: 165–177.

17.

Grobelny

Sokol

Turska

. A study of conformation, configuration and phase structure of polyacrylonitrile and their mutual dependence by means of WAXS and H BL-nmr. Polymer 1984; 25: 1415–1418.

18.

Rizzo

Auriemma

Guerra

et al.

Conformational disorder in the pseudohexagonal form of atactic polyacrylonitrile. Macromolecules 1996; 29: 8852–8861.

19.

Mathieu

Defranceschi

Lécayon

et al.

Insight into polyacrylonitrile molecular structure from calculated vibrational frequencies and infrared intensities of model oligomers. Chem Phys 1994; 188: 183–195.

20.

Cho

Park

Lee

et al.

Effect of heat treatment on structural changes and dynamic mechanical relaxation of polyacrylonitrile fibers. Polym Bull 1993; 30: 663–668.

21.

Bashir

. The hexagonal mesophase in atactic polyacrylonitrile: a new interpretation of the phase transitions in the polymer. J Macromol Sci Part B 2001; 40: 41–67.

22.

Sawai

Kanamoto

Yamazaki

et al.

Dynamic mechanical relaxations in poly (acrylonitrile) with different stereoregularities. Macromolecules 2004; 37: 2839–2846.

23.

Menard

. Dynamic mechanical analysis: A practical introduction, Boca Raton: CRC Press, 2008.

24.

Cotton

Schneider

. Transition temperatures in copolymers of acrylonitrile. Kolloid-Zeitschrift Zeitschrift Polymere 1963; 192: 16–21.

25.

Kimmel

Andrews

. Birefringence effects in acrylonitrile polymers. II. The nature of the 140 C transition. J Appl Phys 1965; 36: 3063–3071.

26.

Okajima

Ikeda

Takeuchi

. A new transition point of polyacrylonitrile. J Polym Sci Part A-1 Polym Chem 1968; 6: 1925–1933.

27.

Minami

Yoshihara

Sato

. Mechanical relaxation properties of polyacrylonitrile, I: relaxation phenomena of copolymers. Kobunshi Kagaku (English edition) 1972; 1: 125–125.

28.

Kenyon

Rayford

. Mechanical relaxation processes in polyacrylonitrile polymers and copolymers. J Appl Polym Sci 1979; 23: 717–725.

29.

Wang

. One-dimensional nanostructures: Electrospinning technique and unique nanofibers, Heidelberg: Springer, 2013.

30.

Andrady

. Science and technology of polymer nanofibers, Hoboken: John Wiley & Sons, 2008.

31.

Fennessey

Farris

. Fabrication of aligned and molecularly oriented electrospun polyacrylonitrile nanofibers and the mechanical behavior of their twisted yarns. Polymer 2004; 45: 4217–4225.

32.

Zussman

Chen

Ding

et al.

Mechanical and structural characterization of electrospun PAN-derived carbon nanofibers. Carbon 2005; 43: 2175–2185.

33.

Jalili

Morshed

Hosseini Ravandi

. Fundamental parameters affecting electrospinning of PAN nanofibers as uniaxially aligned fibers. J Appl Polym Sci 2006; 101: 4350–4357.

34.

Lai

Zhong

Yue

et al.

Investigation of post-spinning stretching process on morphological, structural, and mechanical properties of electrospun polyacrylonitrile copolymer nanofibers. Polymer 2011; 52: 519–528.

35.

Liu

Chen

Gao

et al.

Structure and thermo-chemical properties of continuous bundles of aligned and stretched electrospun polyacrylonitrile precursor nanofibers collected in a flowing water bath. Carbon 2012; 50: 1262–1270.

36.

Qin

. Effects of the stabilization temperature on the structure and properties of polyacrylonitrile-based stabilized electrospun nanofiber microyarns. J Therm Anal Calorimetr 2013; 116: 303–308.

37.

Haghighat

Hosseini Ravandi

. Mechanical properties and in vitro degradation of PLGA suture manufactured via electrospinning. Fiber Polym 2014; 15: 71–77.

38.

Bashir

. Co-crystallization of solvents with polymers: The x-ray diffraction behavior of solvent-containing and solvent-free polyacrylonitrile. J Polym Sci Part B Polym Phys 1994; 32: 1115–1128.

39.

Gupta

Singhal

. Effect of copolymerization and heat treatment on the structure and x-ray diffraction of polyacrylonitrile. J Polym Sci Polym Phys Ed 1983; 21: 2243–2262.

40.

Grobelny

Sokół

Turska

. Nuclear magnetic resonance and wide angle X-ray scattering in poly (acrylonitrile-co-methyl acrylate): 1. The influence of comonomer on structural characteristics and annealing behaviour. Polymer 1989; 30: 1187–1196.

41.

Gupta

Kothari

. Manufactured fibre technology, London: Springer, 1997.

42.

Papkov

Zou

Andalib

et al.

Simultaneously strong and tough ultrafine continuous nanofibers. ACS Nano 2013; 7: 3324–3331.

43.

Ero-Phillips

Jenkins

Stamboulis

. Tailoring crystallinity of electrospun PLLA fibres by control of electrospinning parameters. Polymers 2012; 4: 1331–1331.

44.

Ramakrishna

Fujihara

Teo

et al.

An introduction to electrospinning and nanofibers, London: World Scientific, 2005.

45.

Hosseini Ravandi

Pan

. Morphological characterization of individual polyacrylonitrile nanofibers. Curr Nanosci 2011; 7: 415–419.

46.

Naraghi

Arshad

Chasiotis

. Molecular orientation and mechanical property size effects in electrospun polyacrylonitrile nanofibers. Polymer 2011; 52: 1612–1618.

47.

Zhao

Wang

et al.

Electrospinning of ethyl–cyanoethyl cellulose/tetrahydrofuran solutions. J Appl Polym Sci 2004; 91: 242–246.

48.

Uddin

Xiong

et al.

Process, structure, and properties of electrospun carbon nanotube-reinforced nanocomposite yarns. Adv Mater Sci Eng 2009; 2009: 1–5.

49.

Hosseini Ravandi

Hassanabadi

Tavanai

et al.

Mechanical properties and morphology of hot drawn polyacrylonitrile nanofibrous yarn. J Appl Polym Sci 2012; 124: 5002–5009.

50.

Sadrjahani

Hoseini

Mottaghitalab

et al.

Development and characterization of highly oriented PAN nanofiber. Brazil J Chem Eng 2010; 27: 583–589.

51.

et al.

Orientation structure and mechanical properties of polyacrylonitrile precursors. Adv Mater Res 2006; 11–12: 383–386.

52.

Morton

Hearle

JWS

. Physical properties of textile fibers, Boca Raton: Textile Institute, 1986.

53.

Sreekumar

Liu

Min

et al.

Polyacrylonitrile single-walled carbon nanotube composite fibers. Adv Mater 2004; 16: 58–61.

54.

Hearle

Grosberg

Backer

. Structural mechanics of fibers, yarns, and fabrics, New York: John Wiley & Sons Inc., 1969.