Fabrication of Trans-1,4-polyisoprene Nanofibers by Electrospinning and Its Crystallization Behavior and Mechanism

INTRODUCTION

Trans-1,4-polyisoprene (TPI), as the isomer of natural rubber (NR, cis-1,4-polyisoprene), can be naturally obtained from Balata tree and Gutta-percha tree. However, TPI differs from amorphous NR in that it is a semi-crystallized polymer at room temperature. TPI can be crystallized in monoclinic (α) or orthorhombic (β) forms depending on the crystallization conditions.

Crystallization of TPI from solutions was reported, and normally, mixtures of the α-form and β-form were obtained^{[5, 11, 12]}. Pure α-form was obtained by the precooling method which involves dissolution at an elevated temperature, precipitation, redissolution at a moderate temperature and crystallization at a constant temperature^[13-15]. Flanagan got the β-TPI from agitation butyl acetate solution at 49 ℃^[16]. Direct crystallization from solution, consisting of dissolution at an elevated temperature followed by isothermal crystallization, led to the formation of β-form at lower crystallization temperatures and α-form at higher crystallization temperatures^{[15, 17]}.

Electrospinning technique, as a simple and effective method for the production of ultrathin fibers with diameter in the range of 100-1000 nm, has drawn great attentions. Electrospinning technology could provide ultra-high stretching ratio and high voltage on the polymer dopes. Therefore we are interested in the crystallization behavior of the TPI fibers crystallized from solution under high stretching ratio and high electric voltage during the electrospinning process. Till now, no reports focus on this area. In this study, electrospinning parameters were adjusted to fabricate TPI nanofibers and fibrous membranes from solutions. FTIR was used to in situ follow the crystallization behavior of the TPI nanofibers right after finishing electrospinning. The crystal form and crystallinity of the fully dried and crystallized TPI fibers were obtained by X-ray diffraction.

Regarding the question that whether the α-TPI or β-TPI can be transformed into each other in a solid state, Lovering and Boochathum gave opposite conclusions. Lovering believed that the β-crystals to α-crystals transformation only occurred in melt recrystallization process, the crystal type would not be changed in a solid state^[6]. However, Boochathum found that β to α transformation occurred during crystallization by using DSC. Cyclic transformation between the two forms was observed when samples crystallized at 0 ℃. The inverse transformation from α to β was observed during storage at room temperature for TPI crystals grown from amyl acetate solutions^[7]. It was also reported that the transition between α and β crystals could be induced during stretching. For example, Nikitin and Volchek investigated the ductility of TPI by using infrared spectrum analysis, and found that stretching of the α-TPI produced a more elongated rotational isomer, i.e., β-TPI; and the β-TPI was produced by recrystallization of the α-TPI^[8]. Recently, Weng and Chen investigated the real time structural evolution of α-TPI during tensile deformation, and found the melt and recrystallization process of α-TPI led to the formation of β-TPI, and the total crystallinity increased during deformation^[9]. Under stretching part of the monoclinic α-phase transformed into highly oriented orthorhombic β phase at strain ε = 0.4. The β-phase had rather high orientational degree with polymer chains parallel to the stretching direction while the orientational degree of α-phase was much lower^[10].

Some researchers found the crystallization temperature affected the polymorphism of TPI greatly. In general, it was reported that α-crystals could be obtained when TPI crystallized at higher temperatures, while β-crystals formed at lower crystallization temperatures^[1]. Ratri made highly oriented pure β-form samples by quenching the TPI melt into liquid N₂ followed by stretching about four times of the original length at room temperature, as well, the highly pure and highly-oriented α-form was obtained by the cold draw of the melt-quenched sample at T_g followed by annealing at 60 ℃ for 30 min under tension to eliminate the small amount of the β-form^{[2, 3]}. Fisher and Henderson observed the high rate of primary nuclei formation of β-crystals at lower crystallization temperatures and attributed this temperature dependence to a lower end surface free energy of β-crystals of TPI comparing to that of the α-crystals^[4]. In addition, the cooling rate also influences the crystal form of TPI crystallized from melt. The α-crystals could be prepared by slow cooling of the molten sample, while the β-crystals could be obtained by quenching the molten sample in dry ice-methanol^[5].

Some researchers also studied the crystal-crystal transformation between α-TPI and β-TPI in solution. Mandelkern reported the irreversible transformation from β to α form and the transformation conditions were not clarified^[1]. Leeper found that, β-TPI to α-TPI transformation occurred in benzene solution at temperatures in the range of 64 ℃ to 72 ℃^[18]. The Woodward research group reported a β-TPI to α-TPI transformation for synthetic unfractionated TPI when it was swollen with amyl acetate at 35 ℃ for 17 h. Swelling of TPI in n-butyl acetate for one day at 25 ℃ or 17 h at 35 ℃ also led to this transformation^{[15, 17, 19]}.

EXPERIMENTAL

Electrospinning of TPI

TPI solutions were prepared by dissolving a measured amount of TPI in mixed solvent (volume ratio of CHCl₃ to DMF is 9/1) at 45 ℃ and stirring gently to form transparent solutions. The electrospinning solutions were placed into a 10 mL syringe with a capillary tip having an inner diameter of 0.5 mm. The electrospinning equipment including high voltage power, syringe pump and temperature controller, was purchased from Kaiweixin Technology Company of China. The syringe pump was used to feed polymer solution into capillary tip and the feeding rate of syringe pump can be adjusted. The dope concentration (c) was 3-10 g/mL, the applied voltage (V) was 15-30 kV, the tip-to-collector distance (d) was 10-20 cm, the feeding rate (v) was 20-80 μL/min, and the environmental temperature (T) during electrospinning process was controlled in the range of 25-45 ℃. Aluminum foil was used as collector which was connected to the ground. The electrospun fibrous nonwoven membranes, which composited by lots of TPI nanofibers, were named as ETPI-A to ETPI-O and the electrospinning parameters are shown in Table 1. The electrospun TPI fibrous membranes were dried in vacuum oven at the same temperature with the corresponding electrospinning temperature for 24 h to dry off residue solvents and then stored at room temperature prior to characterization.

Table1. The β-TPI content and total crystallinity of the electrospun TPI nanofibers under different electrospinning conditions

表选项

导出Excel

下载全尺寸表图像

Sample	c (g/100mL)	V (kV)	d (cm)	v (μL/min)	T (℃)	Average fiber diameter (nm)	Hβ/(Hα+Hβ) a	Crystallinity b (%)
ETPI-A	3	30	15	20	25	619	0.79	31.84
ETPI-B	5	30	15	20	25	921	0.75	39.88
ETPI-C	8	30	15	20	25	1611	0.74	30.06
ETPI-D	10	30	15	20	25	1660	0.76	40.55
ETPI-E	3	15	15	80	25	634	0.80	28.40
ETPI-F	3	20	15	80	25	617	0.80	36.15
ETPI-G	3	25	15	80	25	368	0.81	34.57
ETPI-H	3	30	15	80	25	696	0.85	42.66
ETPI-I	3	30	15	80	35	334	0.80	30.41
ETPI-J	3	30	15	80	40	332	0.74	32.23
ETPI-K	3	30	15	80	45	432	0.56	41.01
ETPI-L	3	15	10	80	25	453	0.83	39.24
ETPI-M	3	22.5	15	80	25	366	0.82	32.95
ETPI-N	3	30	20	80	25	490	0.76	31.97
ETPI-O	3	30	15	40	25	727	0.83	41.55
Raw TPI	-	-	-	-	-	-	0	20.62
Casting TPI	3	-	-	-	25	-	0	23.04
a Hα - The peak height of α-TPI (0.49 nm) in XRD; Hβ - The peak height of β-TPI (0.47 nm) in XRD; b Calculated by XRD

Table1. The β-TPI content and total crystallinity of the electrospun TPI nanofibers under different electrospinning conditions

Instrumentation and Characterization

The morphologies of the electrospun TPI fibrous membranes were observed by a scanning electron microscope (SEM, JEOL JSM-7500F, Electronics Corporation, Japan) at an accelerating voltage of 2 kV or 3 kV. Each sample for SEM measurement was stored at room temperature for at least 6 days and then sputter-coated with platinum for observation. Fourier transform infrared (FTIR) spectra were recorded with a Magna-IR Spectrometer 750 instrument using single attenuated total reflectance model. A series of FTIR spectra of the fresh electrospun TPI fibrous membranes were obtained with aging. Wide-angle X-ray diffraction analysis (XRD, D-MAX2500/PC Rigaku Corporation, Japan) was performed with Cu Kα radiation (λ = 0.15406 nm) at a generator voltage of 40 kV and generator current of 100 mA. The scanned 2θ range was from 3° to 45° with a scanning rate of 5 (°)×min⁻¹. Each sample for XRD measurement was stored at room temperature for at least 6 days. The software of Jade 5.0 was used for integral of the peak area and the crystallinity of the electrospun TPI fibrous membranes was calculated according to the Eq. (1)^[20].

where X_c - the crystallinity; A_c - the sum of crystallization peak area; A_a - the sum of amorphous peak area.

Materials

Trans-1,4- polyisoprene (TPI, Trans-1,4- unit ≥ 95 mol%, Mooney viscosity(100 ℃, 1 + 4) of 25) was supplied by KeDa Fangtai Materials Engineering Co., Ltd., Qingdao, China. Chloroform (purity ≥ 99%, Yantai Sanhe Chem. Co., Ltd.) and N, N-dimethyl formamide (DMF, purity ≥ 99.5%, Tianjin Bodi Chem. Co., Ltd.) were used as received.

RESULTS AND DISCUSSION

Figure 3. XRD patterns of the electrospun TPI fibrous membranes and TPI raw material

图选项

下载全尺寸图像

下载幻灯片

TPI can be dissolved in organic solvent such as chloroform. Electrospinning of TPI solutions was studied in this work. Considering the low conductivity of chloroform, mixed solvents of CHCl₃ and DMF were used. By adjusting the electrospinning parameters, perfect TPI fibrous membranes composed of TPI nanofibers were fabricated. The SEM images (Fig. 1) showed that smooth TPI nanofibers with diameters in the range of 300 nm to 1600 nm were obtained. Two factors including dope concentration and electrospinning temperature influenced the fiber morphologies greatly^[21]. The increased dope concentrations led to much thicker fibers, and the increased electrospinning temperature (35 ℃ to 45 ℃) caused fiber adhesion due to the difficulty in fiber solidification. The average fiber diameters are also given in Table 1.

Figure 4. Schematic diagram of β-TPI formation and decrease in crystallinity for electrospun TPI fibers during aging

图选项

下载全尺寸图像

下载幻灯片

Figure 1. SEM images of electrospun TPI fibrous membranes from mixed solvent (volume ratio of CHCl₃ to DMF = 9/1) with different electrospinning parameters as shown in Table 1

图选项

下载全尺寸图像

下载幻灯片

Figure 2. FTIR spectra (a and b) and the crystal content of TPI fibrous membranes during aging (volume ratio of CHCl₃ to DMF = 9/1): (a) and (c) ETPI-A (v = 20 µL/min and T = 25 ℃); (b) and (d) ETPI-K (v = 80 µL/min and T = 45 ℃)

图选项

下载全尺寸图像

下载幻灯片

FTIR was used to characterize the crystallization behaviors of the fresh electrospun TPI fibers with aging, as shown in Fig. 2. Figure 2(a) shows a strong absorption peak at 878 cm^-1 representing β-TPI when the aging time is 0.08 h for ETPI-A. An absorption peak at 668 cm^-1 representing solvent was also observed^{[22, 23]}. This means the fresh electrospun TPI fibers contain solvent crystallized into β-TPI. With the increase in aging time, especially after aging for 24 h, fully dried TPI fibers were still mainly composed of β-TPI. Absorption peak at 1150 cm^-1 attributed to C―C vibration was not affected by TPI crystallization and was used as the internal peak^[24]. Therefore, c_β (peak height at 878 cm^-1/peak height at 1150 cm^-1) representing the relative content of β-TPI, c_α (peak height at 863 cm^-1/peak height at 1150 cm^-1) representing the relative content of α-TPI, c_α+β (c_α + c_β) representing the content of the two crystals and c_β/(c_α + c_β) representing the relative content of β-TPI in the two crystals were calculated and listed in Fig. 2(c). It was found that the c_α+β of the TPI fibers decreased with aging and the rapid decrease occurred in the first 70 h aging. The c_β and c_α curves showed the similar decrease tendency, while c_β/(c_α + c_β) kept nearly unchanged. It seemed that the total crystallinity of the ETPI-A achieved the highest level during electrospinning process and decreased greatly with aging. That means both the α-TPI and β-TPI decreased in the same tendency while the ratio between α-TPI and β-TPI kept unchanged. For ETPI-K, a strong absorption peak at 878 cm^-1 representing β-TPI was observed when the aging time is 0.08 h in Fig. 2(b). The difference happened for ETPI-K was the appearance of obvious absorption peak at 863 cm^-1 representing α-TPI with aging. Based on Fig. 2(d), an obvious decrease in c_β and an increase in c_α were observed, and c_β/(c_α + c_β) decreased with aging. This might indicates the transformation from β to α happened during aging. And the β to α transformation happened in the first 50 h during aging. Therefore the in situ observations of the transformation of fresh electrospun TPI fibers help us to understand the crystallization behaviors of the TPI fibers. It seemed that the fresh ETPI-K aged at 45 ℃ and the fresh ETPI-A aged at 25 ℃ were composed mainly of β-TPI. The ultra-high ratio stretching during electrospinning process may induce the β-TPI formation and is beneficial to the highest crystallinity of the TPI fibers. With the increase in aging time, the crystallinity of the TPI fibers decreased. While for ETPI-K aged at 45 ℃ for 24 h, a transformation from β to α was observed and this transformation probably happened in the first 50 h during aging. What should be pointed out is that the β to α transformation happened in both wet state (containing solvent, 0-24 h aging) and solid state (without solvent, 24-50 h aging).

XRD was used to characterize the crystal form and the crystallinity of the electrospun TPI fibrous membranes, as shown in Fig. 3. The raw TPI material was mainly composed by α-TPI with characteristic 2θ appeared at 18.1° (110), 19.3° (111), 22.8° (200) and 27.0° (210). While for electrospun TPI fibrous membranes, strong characteristic diffraction peaks with 2θ at 18.9° (120) and 22.8° (200) assigning to β-TPI were observed^{[9, 25]}. In order to study the influence of solvents on the crystalline form of TPI, a solution casting membrane of TPI as control was prepared from the same mixed solvent of electrospinning. The XRD pattern of the casting TPI membrane indicated that it was mainly composed of α-TPI. Therefore, we can conclude that the electrospinning process promotes the formation of β-TPI, and solvent will not induce the β-TPI formation. The corresponding β-TPI content and total crystallinity of the electrospun TPI fibrous membranes under different electrospinning conditions are shown in Table 1. The data in Table 1 showed that the increase in voltage and feeding rate could result in an increase in β-TPI content and crystallinity. That indicated the high voltage and high feeding rate would enhance the chain orientation during electrospinning process^[26]. While an increase in electrospinning temperature and distance from tip to collector led to an obvious decrease in the β-TPI content. All the electrospun TPI fibrous membranes possessed higher crystallinity (30%-40%) compared to TPI raw material and casting TPI membrane (20%-23%).

The possible mechanism was proposed to explain the super high crystallinity and high β-TPI content for electrospun TPI nanofibers, as shown in Fig. 4. The TPI dope containing TPI chains in random coil conformation in the solvent (stage 1) was fed to the capillary tip. A high voltage applied on the dope resulted in the whipping movement of polymer jet, which made polymer chains highly stretched and oriented under the high electric-field (stage 2). Once the fibers were collected with most solvent evaporated, the polymer chains in the as-obtained fresh TPI fibers were in highly ordered alignment state. That’s why the electrospun TPI fibers have a higher crystallinity than the raw and casting TPI films (stage 3). With the increase of aging time, the left solvent in fiber caused the movement of TPI chains and facilitated the disorientation of the highly aligned TPI chains. As a result, some crystal phases were broken gradually with the relaxation of TPI chains, and turned to be in amorphous state. Therefore, the crystallinity decreased with aging. Notably, when the fresh TPI fibers aged at 45 ℃, part of the disoriented molecular chains aligned orderly again to a more stable thermodynamic α-TPI state, thus β to α transformation led to the formation of more α-TPI. However, since not all of the disoriented TPI chains coming from the β-crystal domain can align again, the total crystallinity of the samples decreased with the aging time at 45 ℃.

CONCLUSIONS AND OUTLOOK

In this study, electrospinning of TPI solutions was performed and TPI nanofibers with different morphology and fiber diameters were fabricated successfully through changing the electrospinning parameters. The diameter of TPI fibers can be regulated from 300 nm to 1600 nm. In an electrospinning process, an increase in electrospinning voltage and feeding rate, which enhance the chain orientation during electrospinning process, could predominately result in an increase in β-TPI content and crystallinity. An increase in electrospinning temperature and distance led to an obvious decrease in the β-TPI content and crystallinity. All the electrospun TPI fibrous membranes possessed higher crystallinity (30%-40%) compared to TPI raw material and casting TPI membrane (20%-23%).

The crystallinity of the electrospun TPI nanofibers decreased with aging, and the β-TPI ratio changed little with aging at 25 ℃. While for TPI nanofibers aged at 45 ℃, a decrease in crystallinity and β to α transformation were observed with aging and this changing happened in the first 50 h. A possible mechanism was proposed to explain the super high crystallinity and high β-TPI content for electrospun TPI nanofibers. During the electrospinning processing, the highly electrostatic stretching by high voltage combined with solvent volatilization lead to the highly oriented arrangement of TPI chains in β-TPI form.

References

1. [1]

   Mandelkern, L., Quinn, J.F. and Roberts, A.D., J. Am. Chem. Soc., 1956, 78: 926
   

2. [2]

   Ratri, P.J., Tashiro, K. and Iguchi, M., Polymer, 2012, 53: 3548
   

3. [3]

   Ratri, P.J. and Tashiro, K., Polym. J., 2013, 45: 1107
   

4. [4]

   Fischer, E. and Henderson, J.F., J. Polym. Sci. A, 1967, 5: 377
   

5. [5]

   Takahashi, Y., Sato, T. and Tadokoro, H., J. Polym. Sci. Polym. Phys. Ed., 1973, 11: 233
   

6. [6]

   Lovering, E.G. and Wooden, D.C., J. Polym. Sci., Part A: Polym. Chem., 1969, 7: 1639
   

7. [7]

   Boochathum, P., Tanaka, Y. and Okuyama, K., Polymer, 1993, 34: 3694
   

8. [8]

   Nikitin, V.N. and Volchek, B.Z., Zhurnal Prikladnoi Spektroskopii, 1966, 4: 546
   

9. [9]

   Weng, G.S., Wu, J.R. and Xu, Y.C., J. Polym. Res., 2014, 21: 576
   

10. [10]

    Weng, G.S., Bao, J.B., Xu, Y.C. and Chen, Z.R., J. Polym. Res., 2013, 20: 1
    

11. [11]

    Fisher, D., Proc. Roy. Soc. London Set A, 1953, 66: 7
    

12. [12]

    Bunn, C.W., Proc. Roy. Soc. London Set A, 1942, 180: 40
    

13. [13]

    Keller, A. and Martuscelli, E., Makromol. Chem. Phys., 1972, 151: 189
    

14. [14]

    Anandakumaran, K., Herman, W. and Woodward, A.E., Macromolecules, 1983, 16: 563
    

15. [15]

    Anandakumaran, K., Kuo, C.C., Mukherji, S. and Woodward, A.E., J. Polym. Sci., Part B: Polym. Phys., 1982, 20: 1669
    

16. [16]

    Flanagan, R.D. and Rijke, A.M., J. Polym. Sci., Part B: Polym. Phys., 1972, 10: 1207
    

17. [17]

    Kuo, C. and Woodward, A.E., Macromolecules, 1984, 17: 1034
    

18. [18]

    Leeper, H.M. and Schlesinger, W., J. Polym. Sci., 1953, 11: 307
    

19. [19]

    Xu, J.R. and Woodward, A.E., Macromolecules, 1986, 19: 1114
    

20. [20]

    Balta, C.F. and Vonk, C.G., “X-ray scattering of synthetic polymers”, Elsevier, Amsterdam, Holland, 1989, p.175

21. [21]

    Ali, A., Mohamed, M.N., Matsuura, T. and Reza, F., Chinese J. Polym. Sci., 2014, 32(6): 793
    

22. [22]

    Gavish, M., Corrigan, J. and Woodward, A.E., Macromolecules, 1988, 21: 2079
    

23. [23]

    Cao, Y.L., Hao, F.L., Li, J.Y. and Han, S.H., China Elastomer, 1993, 3: 34
    

24. [24]

    Su, F.M., Li, X.Y., Zhou, W.M. and Chen, W., Polymer, 2013, 54: 3408
    

25. [25]

    Sun, X.H. and Ni, X.Y., J. Appl. Polym. Sci., 2004, 94: 2286
    

26. [26]

    Cao, L., Su, D.F., Su, Z.Q. and Chen, X.N., Chinese J. Polym. Sci., 2014, 32(9): 1167