Printer Friendly

Processing, structure, and properties of gel spun PAN and PAN/CNT fibers and gel spun PAN based carbon fibers.

INTRODUCTION

Gel spinning has been used for the production of polyethylene [1], poly(vinyl alcohol) [2], and PAN [3] fibers, as well as their nanocomposite counterparts [4-8]. Advantages associated with gel spinning include the ability to produce highly oriented fibers, yielding high strength and high modulus [9, 10]. Currently, PAN serves as the predominant precursor fiber for the production of high strength carbon fiber [11]. Commercially available high performance PAN based carbon fibers possess excellent tensile strength (3-7 GPa) and modulus (230-588 GPa), as well as good electrical (~[10.sup.4]-[10.sup.5] S/m) and thermal (~10 W/m K) conductivity [11], while having a circular or nearly circular cross-sectional shape. Recently, PAN and polyacrylonitrile/carbon nanotube PAN/CNT gel spun precursor fibers have been carbonized, resulting in fibers possessing high tensile strength and modulus [9, 10, 12-14], without a particular focus on the cross-sectional shape of the fiber.

Efforts have been made since the 1960's to understand the coagulation process of wet-spun acrylic fibers [15-17], and more recently for dry-jet wet-spun acrylic fibers [18-20]. Cross-sectional shapes of the acrylic fibers are diffusion controlled, and are dependent upon the fiber spinning parameters such as solution temperature at spinneret, coagulation bath temperature, coagulation bath composition, solvent, nonsolvent, and polymer composition. Controlling the outward diffusion of the solvent and the inward diffusion of the nonsolvent is not always a straightforward task due to the dynamic nature of the fiber formation process. Fiber formation is controlled by the thermodynamics of the system, while the kinetics dictate the rate at which the final fiber reaches a state of equilibrium. Thermodynamically, fiber formation in wet spinning or dry-jet wet spinning occurs through the process of solidifying the polymer solution extrudate through the outward diffusion of the solvent and the inward diffusion of the non-solvent. However, the kinetics of fiber formation can alter the final fiber cross-section due to the evidence that nonuniform circumferential stresses arise that can collapse the fiber structure, therefore altering the cross-sectional shape of the fiber. Such a collapse is the result of a dense, thin, outer skin forming at the surface of the extruded solution prior to complete solidification of the fiber. As the fiber continues to solidify, the thin outer skin of the fiber collapses, creating a noncircular cross-section. While the cross-sectional shape of the fiber is controlled by the kinetics of the fiber formation, the presence of internal voids within the fiber is controlled by the thermodynamics. For this reason, production of PAN fibers is typically performed at relatively low coagulation bath temperatures (~10 to 15[degrees]C) in the presence of a solvent/nonsolvent coagulation bath [21]. Lower coagulation temperatures and the presence of solvent in the coagulation bath reduces the number of voids in the fiber [18].

Work by Paul [15] has focused on the diffusion processes associated with acrylic fiber formation where dimethylacetamide (DMAc) was used as the polymer solvent, a PAN copolymer consisting of polyacrylonitrile-co-vinyl acetate (PAN-co-VA) at a 92.3/7.7 composition by weight, and water was used as the nonsolvent. To obtain diffusion coefficients of the solvent and nonsolvent, gelled PAN-co-VA rods were prepared in DMAc at a polymer concentration of 26 wt% overnight in test tubes of a predetermined geometry. Rods were then coagulated in a water/ DMAc bath of varying concentrations at either 30 or 50[degrees]C. By monitoring the refractive index of the coagulation medium as a function of time, the outward diffusion rate of the DMAc solvent and inward diffusion rate of the water nonsolvent could be determined. It was concluded that when the coagulation bath temperature is increased from 30 to 50[degrees]C, the outward diffusion rate of the solvent and inward diffusion rate of the nonsolvent are increased. Likewise, as the DMAc content in the coagulation bath was increased, the outward diffusion rate of the solvent and the inward diffusion rate of the nonsolvent decreased. Finally, it was also reported that the ratio of the DMAc/water diffusion rate increased as the DMAc content in the coagulation bath increased, confirming that the outward mass diffusion of DMAc is faster than the inward mass diffusion of water.

In a comparison of the coagulation during the wet-spinning and dry-jet wet-spinning of acrylic fibers, Baojun et al. [18] monitored the diffusion rates of the solvent dimethylformamide (DMF) and nonsolvent (water) with coagulation bath temperatures ranging from 10 to 70[degrees]C and coagulation bath compositions containing up to 51 wt% solvent. It was found that the outward diffusion rate of the solvent and inward diffusion rate of the nonsolvent increase as the coagulation bath temperature is increased for both wet spinning and dry-jet wet spinning of the acrylic fibers. Similarly, as the polymer concentration is increased, the diffusion rates of the solvent and nonsolvent decrease. Diffusion rate ratios of the solvent to nonsolvent were also determined for dry-jet wet spun acrylic fibers. As the coagulation bath temperature was increased, the ratio of the solvent and nonsolvent diffusion rates decreased. Using a coagulation bath of 45-49% solvent and coagulation bath temperatures ranging from 10 to 70[degrees]C, the fiber cross-sectional shape is best described as nonuniform or unsymmetrical at coagulation bath temperatures of 10[degrees]C, oval at 45[degrees]C, and nearly circular at 70[degrees]C. It can then be summarized that as the coagulation bath temperature is increased, the ratio of the solvent to nonsolvent diffusion rate approaches a value of 1, and the cross-sectional shape becomes more symmetric and circular in shape.

Zeng et al. [22] prepared acrylic fibers using dimethylsulfoxide (DMSO) as a solvent, water as a nonsolvent, and a polyacrylonitrile terpolymer (polyacrylonitrile-co-methylacrylate-coitaconic acid, 96/2.5/1.5) using both wet spinning and dry-jet wet spinning techniques. The diffusion rate of both the solvent and nonsolvent increased as the extrusion velocity, bath coagulation temperature, and solvent concentration in the coagulation bath increased, for both wet spinning and dry-jet wet spinning. As the solvent and nonsolvent diffusion rates increased, the ratio of the solvent diffusion rate to the nonsolvent diffusion rate approaches a value of 1. Similar conclusions were drawn that as the ratio of the solvent diffusion rate to the non-solvent diffusion rate approached a value of 1, the fiber cross-sectional shape tends towards a circular shape.

Bajaj et al. [19] studied the structure and property development of three dry-jet wet spun fibers; polyacrylonitrile-comethyl acrylate (PAN-co-MA), polyacrylonitrile-co-methacrylic acid (PAN-co-MAA), and polyacrylonitrile-co-itaconic acid (PAN-co-IA). PAN-co-MAA fibers were spun into a water/DMF coagulation bath (60/40 v/v) maintained at 10 to 40[degrees]C, followed by a second water/DMF coagulation bath (90/10 v/v) maintained at 40[degrees]C, and a third water/DMF (100/0 v/v) coagulation bath maintained at 100[degrees]C. Fibers were then drawn across a heater plate at 130[degrees]C. Density of the PAN-co-MAA fibers collected prior to entering the second coagulation bath increased from 1.169 to 1.177 g/[cm.sup.3], as the temperature of the first coagulation bath was reduced from 40 to 10[degrees]C. The same fiber exhibited a reduction in crystal size from 3.65 to 2.29 nm as the temperature of the first coagulation bath was reduced from 40 to 10[degrees]C. However, the tenacity of the PAN-co-MAA drawn fibers (draw ratio = 6) remained nearly unchanged regardless of the temperature of the first coagulation bath, with tenacity values ranging from 2.23 g/den (10[degrees]C bath) to 2.30 g/den (40[degrees]C bath). It can therefore be concluded, that the internal structure and density of PAN-co-MAA fibers is dependent upon the initial fiber spinning conditions (including the coagulation bath temperature), despite similar tensile properties of the fully drawn fibers. The structure of the drawn PAN-co-MAA fibers exhibited differences in % crystallinity and crystal size, with the 10[degrees]C spun fibers exhibiting a crystallinity of 61% and a crystal size of 4.34 nm, and the 40[degrees]C spun fibers exhibiting a crystallinity of 55% and a crystal size of 5.01 nm, both after being drawn to a draw ratio of 6.

In this study, structure-property relationships of dry-jet wet spun and gel spun PAN and PAN/CNT based precursor fibers are investigated. Specifically, the effect of CNT content, polymer solvent, coagulation bath temperature, coagulation bath composition, and polymer molecular weight are explored. Metrics used include the calculated as-spun fiber circularity, fiber surface roughness using scanning electron microscopy (SEM), structure of as-spun and drawn fibers by wide angle X-ray diffraction (WAXD), and tensile properties of drawn precursor fibers. High resolution transmission electron microscopy (HR-TEM) and WAXD were used to observe the carbon fiber structure and single filament tensile testing provided mechanical property measurements of the PAN based carbon fibers. Finally, the relationship between as-spun and drawn PAN fiber structure and properties to that of the carbon fiber are discussed.

EXPERIMENTAL

Solution Preparation

Polyacrylonitrile-co-methacrylic acid (PAN, 96/4 mol%) polymers with a viscosity average molecular weight ([M.sub.v]) of 247,000 g/mol, 453,000 g/mol, and 513,000 g/mol were obtained from Japan Exlan Co. and dried in vacuum at 90[degrees]C for 2 days. Few-walled carbon nanotubes (FWNT) were obtained from CCNI (Lot # XOC231U) with 1.1 wt% impurity as determined by thermogravimetric analysis (TGA). N-N-dimethylformamide (DMF) and N-N-dimethylacetamide (DMAc) were purchased from Sigma Aldrich and were distilled for further purification. Polymer solutions were prepared by dissolving PAN in the solvent at a concentration of 10.5 g/dL for PAN polymers with [M.sub.v] = 453 and 513 kg/mol, 15 g/dl for PAN polymer with [M.sub.v] = 247 kg/mol. Typical solution preparation conditions included PAN copolymer powder slurry preparation in DMF or DMAc at about 4[degrees]C for about 1 h while stirring. This slurry was then gradually heated over a period of about 2 h to 70[degrees]C and then maintained at that temperature for another 3 h, while being stirred continuously. FWNTs were dispersed in distilled DMF or DMAc using 10 to 30 min of homogenization, followed by either probe sonication for 30 min or 5 cycles in a microfluidizer (Microfluidics), and 24 h bath sonication. The concentration of the CNT dispersion was 5 mg/1 and the dispersions could be stabilized through the addition of dilute PAN solutions. CNT dispersions were then added to the polymer solution where the excess solvent was removed by vacuum distillation to obtain the desired solution concentration. CNT concentrations with respect to the polymer weight were between 0.1 and 1.0 wt%.

Fiber Spinning and Drawing

Multifilament gel spinning was conducted on a spinning system manufactured by Hills Inc. (Melbourne, FL). Various processing conditions were used and are reported at appropriate locations in this article. A 40 or 100-hole spinneret with diameter = 200 pm was used for fiber spinning. A one or two-step gelation condition was used with a solvent/nonsolvent mixture in both gelation baths. Coagulation bath temperatures were controlled between -50 and 25[degrees]C. Table 1 provides the fiber gelation conditions of the as-spun fibers, which were collected directly after coagulation on a fiber spool using a spin draw ratio (SDR) between 2.0 and 3.2. as-spun fibers were kept immersed in a methanol container maintained at -40[degrees]C prior to fiber drawing. Fiber drawing of the 40 or 100 filament tow was conducted in multiple stages, beginning with drawing at room temperature, followed by a two-step hot drawing process on heated godet rollers maintained at 110 and 185[degrees]C, respectively. Post spin draw ratios (PSDR) of the precursor fibers is defined by these three draw ratios. The total draw ratio (TDR) is defined as the SDR multiplied by the PSDR. SDR is the ratio of the fiber take-up speed to the jetting speed of the polymer (from the spinneret), whereas PSDR is the product of the draw ratios at various stages of fiber drawing. Fiber samples collected immediately after fiber spinning and prior to fiber drawing will be designated as "as-spun sample A," "as-spun sample B," etc. Drawn fiber samples collected after postspin fiber drawing will be designated as "precursor A," "precursor B," etc. Table 2 provides the SDR and total draw ratios of the precursor fibers. Each as-spun fiber was drawn to its maximum extent under the condition such that no breakages in the 40 or 100 filament tow occurred throughout the drawing process.

Stabilization and Carbonization

Multifilament precursor fiber tows were converted into carbon fiber using small-scale continuous stabilization/carbonization equipment manufactured by Harper International. Stabilization of the precursor fiber using multiple oxidation ovens precedes carbonization using low-temperature (LT) and high-temperature (HT) furnaces. Carbonized fiber samples collected will be designated as "carbon fiber A," "carbon fiber B," etc. As an example, "as-spun sample A" will be drawn into "precursor A" and carbonized into "carbon fiber A." Stabilization temperatures increased monotonically from 200 to 290[degrees]C for carbon fibers A and E. Residence time of these fibers in the oxidation ovens was in the range of 135 to 155 min. For all other fibers (carbon fibers [A.sub.1], [A.sub.2], H, I, K, L, and M), stabilization temperature increased monotonically from 180 to 250[degrees]C and the residence times were in the range of 240 to 280 min. LT furnace temperatures ranged from 600[degrees]C in the first zone to 1000[degrees]C in the final zone for carbon fibers A and E, and the residence times were in the range of 7 to 10 min. All other carbon fibers passed through the LT furnace with the first zone maintained at 500[degrees]C and the final zone maintained at 675[degrees]C, with the residence times in the range of 10 to 14 min. Subsequently, all fibers went through HT furnace, where the maximum carbonization temperature was 1450[degrees]C, and the residence times were in the range of 9 to 16 min. Typical apparent tension values in various stabilization and carbonization zones were in the range of 20 to 50 MPa. This tension is based on the diameter of the precursor fiber. However, the diameter of the fiber decreases as it progresses through stabilization and carbonization, therefore actual tension values will be higher than the apparent tension values reported here. Carbon fibers Ai and A2 were manufactured from precursor fibers prepared using the same fiber spinning and drawing protocols as carbon fiber A.

Characterization

As-spun fiber cross-sections were imaged using a Leica DM 2500 optical microscope by mounting the as-spun fibers in epoxy (Epofix), and preparing 10 pm thick sections using a Leica DM 2550 microtome. Image analysis was performed using ImageJ (NIH) to determine the as-spun fiber equivalent diameter, circularity index, and roundedness. Circularity index as determined by ImageJ, was calculated using the following equation:

Circularity = 4[pi]A/[(Perimeter).sup.2] (1)

where A is the cross-sectional area of the filament. A perfectly circular filament will have a circularity index value of 1, and less circular filaments will have values less than 1.

Roundness as determined by Image J was calculated using the following equation:

Roundness = 4A/[pi] x [MajorAxis.sup.2] = [d.sup.2.sub.equiv]/[MajorAxis.sup.2] (2)

where A is the cross-sectional area, MajorAxis is the length of the major axis of the ellipse used to approximate the cross-sectional fiber shape, and [d.sub.equiv] is the equivalent diameter of a circular fiber with the same cross-sectional area.

SEM was performed on a Zeiss Ultra 60 FE-SEM at a working distance of 4 mm and an accelerating voltage of 2 keV. as-spun fiber surfaces were gold sputtered prior to imaging in a Hummer 6 gold/palladium sputtering system. Single filament tensile testing was performed on a Favimat tensile testing instrument equipped with a vibroscope for on-line linear density measurement. Precursor and carbon fibers were tested using methods previously reported [23], at a gauge length of 25.4 mm, at a strain rate of 1%/s and 0.1 %/s, respectively. WAXD was conducted on 100 filament as-spun fiber bundles, 200 filament precursor fiber bundles, and 200 filament carbon fiber bundles. HR-TEM was performed on a Tecnai F30 (FEI) at an accelerating voltage of 80 keV. TEM samples were prepared through focused ion-beam (FIB) milling (FEI, Nova Nanolab 200 FIB/SEM) of individual fibers mounted on 3-post copper grids (Electron Microscopy Sciences, Omniprobe lift-out grid) using Gatan G-2 Epoxy. TEM samples were thinned using 30 keV ion-beam voltage prior to final thinning at 5 keV.

RESULTS AND DISCUSSION

Effect of CNT Loading on Cross-Sectional Shape and Precursor Fiber Structure and Properties

Figure 1 provides the optical microscopy cross-sections of all as-spun samples. As-spun samples A, B, and C are provided in Figs, la-c, respectively. As a result of the methanol gelation temperature of -50[degrees]C, the cross-sectional shape of as-spun samples A, B, and C is irregular. Circularity indices of as-spun samples A, B, and C (Table 1) were calculated to be 0.65, 0.66, and 0.72, respectively. Table 3 contains the structural data obtained from WAXD of the as-spun fibers. As-spun samples A, B, and C all exhibit similar levels of crystallinity despite distinctly different crystal sizes. There appears to be no correlation between crystal size and CNT content in as-spun samples A, B, and C.

Drawing of as-spun samples A, B, and C resulted in total draw ratios of 24.6, 25.8, and 27.2, respectively. The tensile strength of precursors A, B, and C were measured to be greater than 1 GPa, while the tensile modulus ranged from 20 to 22 GPa. The structural parameters of precursor A, B, and C are listed in Table 4. Fiber crystallinity and crystal size are independent of CNT loading in precursors A, B, and C. CNT containing fibers (precursors B and C) both exhibit higher [f.sub.PAN] as compared to precursor A as well as a lower 20-meridional peak position. The lower 20-meridional peak position in precursors B and C suggests that the addition of CNT contributes to conformational changes of the PAN chains, with a higher propensity for extended chain PAN as compared to a helical conformation.

As compared to the corresponding as-spun samples, all three precursor fibers exhibit higher crystallinity, larger crystal size, higher [f.sub.PAN], and a lowering of the 20-meridional peak position. The ~10 to 12% increase in crystallinity is quite low as compared to the large increase in crystal size during fiber drawing, suggesting that neighboring crystals merge together to form larger crystals with only a 10 to 12% increase in overall fiber crystallinity during fiber drawing. As the fiber is drawn, the PAN crystallites achieve a higher orientation and the lowering of the 20-meridional peak position.

Effect of Gelation Bath Temperature on Cross-Sectional Shape and Precursor Fiber Structure and Properties

The effect of gelation bath temperature on the fiber structure and cross-sectional shape was investigated in PAN and PAN/CNT as-spun samples. To do so, fibers were gelated in a methanol bath maintained at either -50[degrees]C or at room temperature. Optical microscopy cross-sections of as-spun sample D (25[degrees]C methanol coagulation) in Fig. Id exhibited a kidney bean shape. Increasing the methanol gelation bath temperature from -50[degrees]C in as-spun samples A, B, and C, to 25[degrees]C in as-spun sample D, creates the change in cross-sectional shape from irregular to kidney bean. It is interesting to note that the circularity index of as-spun sample D does not reflect the observed change in cross-sectional shape in Fig. 1.

As-spun sample D (precursor D, TDR = 39) resulted in the highest tensile strength amongst all precursor fibers studied, with a measured tensile strength of 1.16 GPa, a crystallinity of 61% and [f.sub.PAN] of 0.89 were determined using WAXD.

Figures le and f provide the optical microscopy images of as-spun samples E and F, respectively. As-spun sample E (-50[degrees]C methanol gelation) exhibits an irregular cross-sectional shape, similar to that of as-spun sample A, while as-spun sample F (25[degrees]C methanol coagulation) exhibits a kidney bean cross-sectional shape, similar to that of as-spun sample D. Comparison of Fig. 1a and e and Fig. 1d and f, indicates that the higher solid content in as-spun samples E and F have a slight influence on the fiber cross-sectional shape.

Figures 2a and b provide the photographs of the as-spun sample E and as-spun sample F wound spools collected immediately after fiber spinning, respectively. It can be observed that the appearance of the two as-spun fiber spools is distinctly different, with as-spun sample E being transparent and yellowish in color, while as-spun sample F is opaque and whitish in color. Differences in appearances of fibers as a function of coagulation conditions have been attributed to the internal void structure in acrylic fibers [21], As the size/amount of internal voids is decreased, fibers became more transparent. The higher luster of as-spun sample E as compared to as-spun sample F, indicates as-spun sample E possesses fewer internal voids. However, it has been observed that after fiber drawing the internal voids within the as-spun fiber collapse and do not have a significant effect on the drawn precursor fiber appearance and tensile properties [21].

Surface morphology of as-spun fibers observed by SEM is shown in Fig. 2. Figures 2c and d show the fiber surface of as-spun sample E and as-spun sample F, respectively. Differences in the fiber surface are evident between the two fibers, with as-spun sample F exhibiting larger dimension striations as compared to as-spun sample E.

Table 3 provides the structural analysis of as-spun samples E and F as determined by WAXD, and these two samples exhibit distinctly different crystal sizes. Despite both fibers possessing 57% crystallinity, as-spun sample E possesses a 233% larger crystal size as compared to as-spun sample F, as well as a higher [f.sub.PAN] and a smaller [d.sub.2[theta]-17[degrees]] spacing of the (200,110) crystallographic plane. As-spun sample E also exhibits a downshift in the meridional-2[theta] peak position as compared to as-spun sample F.

Figure 3 presents the 2-D WAXD patterns of the as-spun fiber, drawn precursor fiber, and the carbon fiber, as well as their corresponding integrated, equatorial, and azimuthal intensity profiles. The crystal structure of PAN can be either hexagonal or orthorhombic. On the basis of the shape of 2[theta] ~17[degrees] peak in the as-spun fiber integrated intensity profile (Fig. 3d), as well as the equatorial d-spacing of the 2[theta] ~ 17[degrees] and 2[theta] ~ 30[degrees] peak positions, the as-spun fiber crystal structure is hexagonal or pseudo-hexagonal. Figure 4a provides the hexagonal packing structure of PAN. Using the assumption of a perfect hexagonal packing structure and the same crystal length, the differences in crystal structure of as-spun sample E and as-spun sample F are depicted in Fig. 4b and c, respectively. Considering the crystal size of as-spun sample E (5.6 nm) and as-spun sample F (2.4 nm) and their equal levels of crystallinity, the distribution of crystals throughout the fiber is strikingly different. As can be seen in Fig. 4b and c, the size of the PAN crystals in as-spun sample E are larger than in as-spun sample F, but the estimated total crystallinity of the two schematics is the same, due to the total volume occupied by the crystals. The total number of crystals in as-spun sample F are ~5.4 times greater than the number of crystals in as-spun sample E. With the knowledge that as-spun samples E and F were prepared from the same PAN/DMF solution and the only difference was the gelation bath temperature of the two fibers, the lower gelation bath temperature in as-spun sample E results in distinctly different PAN crystal sizes as determined by WAXD (Table 4, Fig. 4), as well as fiber surface morphology and fiber luster.

Figure 5 presents the DMA data for as-spun samples E and F, which also shows interesting behavior. Figure 5a plots the storage (E') and loss (E") modulus of the two as-spun fibers, and Fig. 5b plots the tan [delta] behavior of the as-spun fibers at a frequency of 10 Hz. E' of as-spun sample F is higher than the E' of as-spun sample E from room temperature to ~130[degrees]C, at which point E' of as-spun sample E begins to have a higher E' than as-spun sample F. Differences in the tan [delta] behavior are clearly noticeable, with as-spun sample E possessing a sharp tan [delta] peak at ~85[degrees]C, while as-spun sample F shows an increased tan 8 peak temperature of ~104[degrees]C, as well as a reduced tan [delta] peak maximum and a broadening of the tan [delta] peak. Therefore, not only are the crystalline regions of the two as-spun fibers different as determined from WAXD, the amorphous regions of the two fibers are also markedly different. As-spun sample F tan 8 behavior suggests greater elastic behavior below the glass transition temperature ([T.sub.g]) and a broader [T.sub.g] as compared to as-spun sample E.

Post spin fiber drawing of precursors E and F resulted in moderately higher fiber crystallinity, as well as increases in crystal size. Both precursors E and F had an average tensile strength values above 0.8 GPa. The average tensile modulus was 18.4 and 18.0 GPa for precursors E and F, respectively. However, precursor E (12.8%) exhibited a higher elongation at break as compared to precursor F (10.8%).

Effect of Gelation Bath Composition on Cross-Sectional Shape and Precursor Fiber Structure and Properties

From the results presented above on as-spun samples A-F, the as-spun fiber cross-sectional shape, crystal structure, and fiber luster are greatly influenced by the gelation bath temperature. Addition of solvent to the gelation bath while maintaining an elevated bath temperature (room temperature or higher) has been shown to alter the fiber cross-sectional shape as compared to fibers, which are coagulated in a coagulation bath with a lower volume fraction of solvent or in a solvent free coagulation bath [18]. In the solvent free bath, the fiber cross-sectional shape was kidney bean, and as the solvent content increased in the coagulation bath, the fiber cross-sectional shape could be described as oval or circular [18]. Figure 1 provides the optical microscopy cross-sectional images of as-spun fibers D, G, H, and 1, all of which were gelated in a mixture of MeOH and DMF at room temperature (Table 1). Comparison of Fig. 1d and g provide visual evidence of a change in the as-spun fiber cross-sectional shape as the gelation bath composition is changed from a 100:0 to 80:20 MeOH:DMF bath composition in as-spun sample D and as-spun sample G, respectively. The circularity index increases from 0.65 to 0.88 as the DMF content in the gelation bath is raised from 0 to 20 vol% in as-spun samples D and G. Further increases in the DMF content to 30 vol% in as-spun sample H (Fig. 1h) improves the cross-sectional shape, resulting in the circularity index of 0.93.

Figure 6 provides the SEM surface images of as-spun samples D, G, and H, respectively. From the images, it is possible to see differences in striations along the surface of the fiber, with smaller striations presenting as the DMF content in the gelation bath is increased. Based upon the results in Figs. 1 and 6, as the circularity index of the as-spun fibers is increased with increased in DMF content, the surface morphology exhibits an increased surface smoothness along with smaller striations on the fiber surface.

Postspin fiber drawing of precursors G and H increased PAN crystallinity and crystal size, as compared to their corresponding as-spun samples. The meridional 2[theta] peak position decreased from 2[theta] = 40.1[degrees] to 2[theta] = 39.4[degrees] after fiber drawing, indicative of a propensity for planar zigzag conformation of the PAN molecules and reduced proclivity of a helical conformation. Correspondingly, the tensile properties of precursors G and H resulted in tensile strengths of 0.84 and 0.81 GPa, and tensile modulus values of 18.1 and 18.4 GPa, respectively.

In contrast to as-spun sample H, as-spun sample I (Fig. 1i) was manufactured under the same gelation conditions, with a spin draw ratio 2.0, as compared to the typical spin draw ratio applied in this study of 3.0 to 3.2. As a result of the lower spin draw ratio in as-spun sample I, the PAN crystallinity, crystal size, and [f.sub.PAN] were all decreased and the 2[theta] meridional peak position was increased as compared to as-spun sample H. Because of the lower spin draw ratio, as-spun sample I could be drawn to a higher post spin draw ratio as compared to as-spun sample H. As a result of the higher PSDR applied to as-spun sample I, the tensile strength of precursor I increased to 1.03 GPa with a measured tensile modulus of 18.7 GPa. In comparison with precursor H, precursor I exhibits a 27% increase in tensile strength and 22% increase in elongation at break.

Effect of Organic Solvent on Cross-Sectional Shape and Precursor Fiber Structure and Properties

Polar aprotic solvents such as DMF, DMAc, and dimethylsulfoxide (DMSO) are typical spinning solvents for the production of PAN fibers. Because of the ternary nature of the phase diagram, which characterizes the thermodynamic relationship between polymer, solvent, and the nonsolvent, the choice of organic solvent will alter the fiber formation during the solidification process. By altering the choice of solvent from DMF to DMAc or DMSO used for fiber spinning, the interaction parameters between the solvent and polymer and the solvent and nonsolvent are modified. Changes in the interaction parameters will result in a modification of the cross-sectional shape of the fiber. Figure 1 contains the optical microscopy cross sections of as-spun samples J-M, which utilized DMAc as the solvent. Comparison of Fig. 1a and j depict the difference in cross-sectional shape when using DMF and DMAc as solvent and using a methanol gelation bath maintained at -50[degrees]C. The circularity index of as-spun sample J is improved to 0.85, as compared to as-spun sample A, which has a circularity index of 0.65. Changes in the as-spun fiber cross-sectional shape when a room temperature methanol bath is used can be observed by comparing as-spun sample D (Fig. Id) with as-spun sample K (Fig. 1k). An improvement in circularity index from 0.65 (as-spun sample D) to 0.90 (as-spun sample K) occurs when DMAc is used as the polymer solvent as opposed to DMF. Alteration of the gelation bath composition from 100:0 MeOH:DMAc to 90:10

MeOH:DMAc (Fig. 1) results in an increased circularity index from 0.90 to 0.96. Increasing the DMAc concentration in the gelation bath to 70:30 MeOH:pDMAc with a bath temperature of 25[degrees]C (Fig. lm) results in a circularity index of 0.96. This suggests that the optimal processing conditions to produce a circular cross-sectional shape is the use of a room temperature 90:10 MeOH:DMAc gelation bath for the PAN used in this study.

Postspin drawing of as-spun samples J-M, resulted in precursors J-M with tensile strengths ranging from 0.76 to 0.94 GPa and tensile modulus values ranging from 18.3 to 19.1 GPa. Precursor J, with 0.25 wt% FWNT loading and -50[degrees]C gelation bath temperature, exhibits the highest tensile strength (0.94 GPa), while precursor K exhibits the lowest tensile strength, 0.76 GPa, from fibers prepared with DMAc as the solvent. Precursor K could not be drawn to the same PSDR as the other fibers, which provides an explanation for the lower tensile strength values as compared to precursors J (0.94 GPa), L (0.86 GPa), and M (0.81 GPa). WAXD structural analysis of precursor M results in a fiber crystallinity of 55%, and crystal size of 15.4 nm. Such results suggest that as the gelation bath composition contains a larger volume fraction of solvent at an increased gelation bath temperature, fiber circularity improves while the drawn fiber tensile properties and crystallinity are reduced.

Carbonization of Gel-Spun PAN Based Carbon Fibers

Table 5 provides the tensile properties of the PAN based carbon fibers. Carbon fiber tensile strength and modulus values range from 2.6 to 5.8 GPa and 264 to 375 GPa. respectively. Table 6 provides the carbon fiber structural parameters as determined by WAXD, while Fig. 7 provides the cross-sectional and surface SEM images of the PAN based carbon fibers. Carbon fibers A and E underwent similar stabilization temperatures, as noted in experimental section, while the remaining carbon fibers underwent stabilization at lower temperatures that resulted in improved tensile properties. This improvement is shown in the comparison of carbon fiber A and carbon fiber [A.sub.1], where an 32% improvement in tensile strength is observed for carbon fiber [A.sub.1].

Figure 7a 1 provides the SEM cross-sectional fracture surface of carbon fiber A, which maintains the irregular shape observed in as-spun sample A (Fig. la). In comparison to the irregular cross-sectional shape observed in carbon fiber A, carbon fiber H possesses the highest circularity index from fibers manufactured with DMF as the solvent. Carbon fiber H (Fig. 7c 1), manufactured using a room temperature gelation bath with 30 vol% DMF, possesses a tensile strength of 3.9 GPa and a tensile modulus of 315 GPa. Differences in the carbon fiber surface morphology also exist in carbon fibers A and H. Carbon fiber A (Fig. 7a3 and 7a4) and carbon fiber H (Fig. 7c3 and 7c4) retain similar fiber surface features that were observed in the corresponding as-spun samples. Carbon fiber A exhibits a smooth carbon fiber surface whereas carbon fiber H shows striations along the fiber axis. HR-TEM images of carbon fiber A center in Fig. 8al and edge in Fig. 8a2 exhibit a distinct turbostratic crystallite structure in both the carbon fiber center and edge. In contrast, the turbostratic crystallite structure in carbon fiber H center (Fig. 8b 1) and edge (Fig. 8b2) is not as readily distinguishable as in carbon fiber A, indicative of a reduced sheath-core effect in the carbon fiber structure.

Carbon fiber E has the lowest tensile strength (2.6 GPa) and tensile modulus (282 GPa) of the carbon fibers manufactured from the gel spinning process in this study. It is interesting to note that precursor E has an ~11 % lower tensile strength and tensile modulus as compared to precursor A, while carbon fiber E has an ~69% and 12% lower tensile strength and tensile modulus as compared to carbon fiber A, respectively. The significantly lower tensile strength of carbon fiber E can partially be attributed to the lower molecular weight PAN utilized in the manufacture of precursor E as compared to carbon fiber A. During the thermal conversion of the precursor fiber in carbon fiber, the higher likelihood of PAN chain ends in precursor E will act as defect sites in carbon fiber E, and as a result is expected to produce a carbon fiber of lower tensile strength as compared to carbon fiber A [24],

Figure 7e1, f1, and g1 provide the cross-sectional SEM images of carbon fibers K-M, manufactured with DMAc as the solvent. Carbon fibers K and L both possess a tensile strength of 4.0 GPa, while carbon fiber M has a tensile strength of 3.6 GPa. It is interesting to note that despite the improved circularity of carbon fibers K-M, as compared to carbon fiber A, the tensile strength of the more circular fibers is lower than that of the irregularly shaped carbon fiber A. Unlike carbon fibers A and H, which were manufactured from DMF and show differences in fiber surface roughness, carbon fibers K-M do not show striations along the fiber surface. The HR-TEM transverse cross-sectional images of carbon fibers K and M are provided in Fig. 8c1 and c2 and Fig. 8d1 and d2, respectively. Carbon fiber K shows the similar random orientation of turbostratic crystallites throughout the cross-section that are observed in carbon fibers A and H. On the other hand carbon fiber M exhibits preferentially oriented turbostratic crystallites in the transverse HR-TEM cross-section. The oriented turbostratic crystallites observed in the transverse cross-section of carbon fiber M is distinct from the turbostratic structure in carbon fibers A, H, and K.

CONCLUSIONS

PAN and PAN/CNT fibers were manufactured using a gel spinning technique using DMF or DMAc as a solvent and a solvent-free or solvent/nonsolvent gelation bath maintained at -50 to 25[degrees]C. As-spun PAN and PAN/CNT fibers exhibited similar irregular cross-sectional shapes when gelated at -50[degrees]C in methanol, with increasing as-spun fiber circularity as the gelation bath temperature and solvent concentration was increased. Increased gelation bath temperature and solvent content in the gelation bath resulted in smaller striations on the as-spun fiber surface, indicative of higher surface roughness. As-spun fibers manufactured with DMAc as the solvent resulted in higher circularity fibers as compared to similarly manufactured fibers where DMF was used as the solvent. Despite the differences in gelation bath conditions, drawn precursor fibers exhibited tensile strengths and tensile modulus values from 0.76 to 1.16 GPa and 18.0 to 22.1 GPa, respectively. Carbonization of the PAN precursor fibers resulted in fibers with tensile strengths as high as 5.8 GPa and tensile modulus as high as 375 GPa. Reduction of the molecular weight of the PAN copolymer during precursor fiber manufacture resulted in a carbon fiber tensile strength and modulus of 2.6 and 282 GPa, respectively. These results are a point of reference for the mechanical properties of gel spun PAN based carbon fibers manufactured from a PAN precursor fiber prepared from the PAN copolymer with [M.sub.v] = 247,000 g/ mol. Further process optimization is required for an adequate comparison with the gel spun PAN based carbon fibers manufactured from the PAN copolymer with [M.sub.v] = 513,000 g/mol. The cross-sectional shape of the PAN based carbon fibers could be altered from irregular to circular cross-section, through the use of different spinning solvents, increased coagulation bath temperatures, and increased solvent concentration in the coagulation bath. Adjustment of the fiber spinning parameters listed, not only impacts the cross-sectional shape of the resulting car bon fiber, but the internal structure development as well. It will be essential to manufacture PAN based carbon fibers with the internal structure present in the high strength and high modulus gel spun PAN based carbon fibers that possess a circular cross-sectional shape.

REFERENCES

[1.] P. Smith, P.J. Lemstra, B. Kalb, and A.J. Pennings, Polym. Bull., 1, 733 (1979).

[2.] W.I. Cha, S.H. Hyon, and Y. Ikada, J. Polym. Sci. Part II: Polym. Pliys., 32, 297 (1994).

[3.] D. Sawai, Y. Fujii, and T. Kanamoto, Polymer, 47, 4445 (2006).

[4.] S.L. Ruan, P. Gao, and T.X. Yu, Polymer, 47, 1604 (2006).

[5.] M.L. Minus, H.G. Chae, and S. Kumar, Macromol. Chem. Phys., 210, 1799 (2009).

[6.] H.G. Chae, M.L. Minus, and S. Kumar, Polymer, 47, 3494 (2006).

[7.] B.A. Newcomb, P.V. Gulgunje, K. Gupta, Y. Liu, H.G. Chae, D. Tsentalovich, M. Pasquali, and S. Kumar, Polymer, 55, 2734 (2014).

[8.] R. Jain, M.L. Minus, H.G. Chae, and S. Kumar, Macromol. Mater. Eng., 295, 742 (2010).

[9.] H.G. Chae, B.A. Newcomb, P.V. Gulgunje, Y. Liu, K.K. Gupta, M.G. Kamath, K.M. Lyons, S. Ghoshal, C. Pramanik, L. Giannuzzi, K. Sahin, I. Chasiotis, and S. Kumar, Carbon, 93. 81 (2015).

[10.] B.A. Newcomb, L.A. Giannuzzi, K.M. Lyons, P.V. Gulgunje, K. Gupta, Y. Liu, M. Kamath, K. McDonald, J. Moon, B. Feng, G.P. Peterson, H.G. Chae, and S. Kumar, Carbon, 93, 502 (2015).

[11.] Y. Liu and S. Kumar, Polym. Rev. (Philadelphia. PA), 52, 234 (2012).

[12.] H.G. Chae, M.L. Minus, A. Rasheed, and S. Kumar, Polymer, 48, 3781 (2007).

[13.] Y.D. Liu, H.G. Chae, and S. Kumar, Carbon, 49, 4487 (2011).

[14.] K. Sahin, N.A. Fasanella, I. Chasiotis, K.M. Lyons, B.A. Newcomb, M.G. Kamath, H.G. Chae, and S. Kumar, Carbon, 77, 442 (2014).

[15.] D.R. Paul, J. Appl. Polym. Sci., 12, 383 (1968).

[16.] D.R. Paul, J. Appl. Polym. Sci., 12. 2273 (1968).

[17.] D.R. Paul, J. Appl. Polym. Sci., 13, 817 (1969).

[18.] Q. Baojun, P. Ding, and W. Zhenqiou, Adv. Polym. Techno!., 6, 20 (1986).

[19.] P. Bajaj, T.V. Sreekumar, and K. Sen, J. Appl. Polym. Sci., 86, 773 (2002).

[20.] S.H. Bahrami, P. Bajaj, and K. Sen, J. Polym. Sci., 89, 12 (2003).

[21.] B.G. Frushour, in Acrylic Fiber Technology and Applications, J.C. Masson, Ed., Marcel Dekker, New York, Chapter 7, 197 (1995)

[22.] X.M. Zeng, G. Zhang, Y.W. Zhang, J.X. Zhao, and D. Pan, J. Macromol. Sci. Part A: Pure Appl. Chem., 43, 1711 (2006).

[23.] K.M. Lyons, B.A. Newcomb, K.J. McDonald, H.G. Chae, and S. Kumar, J. Compos. Mater., 49, 2231 (2015).

[24.] H.G. Chae and S. Kumar, Science, 319, 908 (2008).

Bradley A. Newcomb, [1] Prabhakar V. Gulgunje, [1] Kishor Gupta, [1] Manjeshwar G. Kamath, [1] Yaodong Liu, [1] Lucille A. Giannuzzi, [2] Han Gi Chae, [1] Satish Kumar [1]

[1] School of Materials Science and Engineering, Georgia Institute of Technology, Atlanta, Georgia 30332

[2] L.A. Giannuzzi & Associates LLC, 12580 Walden Run Dr., Fort Myers, Florida 33913

Correspondence to: S. Kumar; e-mail: satish.kumar@mse.gatech.edu Contract grant sponsor: DARPA, the Army Research Office; contract grant number: W9I INF-10-1-0098; contract grant sponsor: Air Force Office of Scientific Research; contract grant number: FA9550-14-1-0194.

DOI 10.1002/pen.24153

Published online in Wiley Online Library (wileyonlinelibrary.com).

TABLE 1. Fiber spinning conditions and image analysis of as-spun
fiber cross-sections.

                     Polymer        CNT       Viscosity at
                   (Mv,kg/mol)/   content   1 rad [s.sup.-1]
Trial                solvent       (wt%)         (Pa-s)

As-spun Sample A     513/DMF       None            83
As-spun Sample B     513/DMF        0.1           100
As-spun Sample C     513/DMF        1.0            63
As-spun Sample D     453/DMF        0.5            92
As-spun Sample E     247/DMF       None            45
As-spun Sample F     247/DMF       None            45
As-spun Sample G     513/DMF       None            88
As-spun Sample H     513/DMF       None            88
As-spun Sample I     513/DMF       None            88
As-spun Sample J     513/DMAc      0.25            36
As-spun Sample K     513/DMAc      None           219
As-spun Sample L     513/DMAc      None           181
As-spun Sample M     513/DMAc      None           113

                            Coagulation conditions

                      Bath 1                       Bath 2
                   temperature      Bath 1      temperature
Trial              ([degrees]C)   composition   ([degrees]C)

As-spun Sample A     N/A (a)          N/A           -50
As-spun Sample B     N/A (a)          N/A           -50
As-spun Sample C     N/A (a)          N/A           -50
As-spun Sample D     N/A (a)          N/A            25
As-spun Sample E     N/A (a)          N/A           -50
As-spun Sample F        25           100:0          -50
As-spun Sample G        25           80:20          -50
As-spun Sample H        25           70:30          -50
As-spun Sample I        25           70:30          -50
As-spun Sample J     N/A (a)          N/A           -50
As-spun Sample K     N/A (a)          N/A            25
As-spun Sample L        25          90 : 10          25
As-spun Sample M        25          70 : 30         -50

                      Equivalent
                       diameter           Circularity
Trial                 ([micro]m)             index

As-spun Sample A   32.0 [+ or -] 2.5   0.65 [+ or -] 0.09
As-spun Sample B   30.7 [+ or -] 2.8   0.66 [+ or -] 0.08
As-spun Sample C   31.7 [+ or -] 2.1   0.72 [+ or -] 0.08
As-spun Sample D   33.4 [+ or -] 2.1   0.65 [+ or -] 0.03
As-spun Sample E   34.8 [+ or -] 2.0   0.61 [+ or -] 0.07
As-spun Sample F   34.9 [+ or -] 2.2   0.67 [+ or -] 0.05
As-spun Sample G   31.7 [+ or -] 2.0   0.88 [+ or -] 0.05
As-spun Sample H   32.7 [+ or -] 2.5   0.93 [+ or -] 0.03
As-spun Sample I   38.9 [+ or -] 2.9   0.91 [+ or -] 0.04
As-spun Sample J   31.0 [+ or -] 2.0   0.85 [+ or -] 0.03
As-spun Sample K   31.4 [+ or -] 2.6   0.90 [+ or -] 0.04
As-spun Sample L   31.8 [+ or -] 2.4   0.96 [+ or -] 0.02
As-spun Sample M   31.7 [+ or -] 2.5   0.96 [+ or -] 0.02

                                           Cross-sectional
Trial                 Roundedness               shape

As-spun Sample A   0.66 [+ or -] 0.07         Irregular
As-spun Sample B   0.66 [+ or -] 0.09         Irregular
As-spun Sample C   0.66 [+ or -] 0.07         Irregular
As-spun Sample D   0.64 [+ or -] 0.14        Kidney Bean
As-spun Sample E   0.62 [+ or -] 0.06   Irregular/Kidney Bean
As-spun Sample F   0.71 [+ or -] 0.12   Kidney Bean/Dog bone
As-spun Sample G   0.69 [+ or -] 0.09   Kidney Bean/Dog bone
As-spun Sample H   0.77 [+ or -] 0.08       Dog bone/oval
As-spun Sample I   0.73 [+ or -] 0.09       Dog bone/oval
As-spun Sample J   0.87 [+ or -] 0.05   Irregular/Kidney Bean
As-spun Sample K   0.69 [+ or -] 0.09           Oval
As-spun Sample L   0.86 [+ or -] 0.06       Oval/Circular
As-spun Sample M   0.87 [+ or -] 0.06       Oval/Circular

(a) Bath number 1 was not used in these trials. Composition of bath
number 2 was 100% methanol in all cases.

TABLE 2. Draw ratios and mechanical properties of precursor fibers.

                                                         Tensile
Trial         SDR (a)/TDR (b)   Diameter ([micro]m)   strength (GPa)

Precursor A      3.0/24.6           11.2 (7.1)         1.05 (12.4)
Precursor B      3.0/25.8           11.5 (7.8)         1.00 (11)
Precursor C      3.2/27.2           11.1 (7.2)         1.06 (10.4)
Precursor D      3.0/39              9.5 (5.3)         1.16 (12.9)
Precursor E      3.0/33.3           12.4 (4.8)         0.95 (15.8)
Precursor F      3.0/33.3           12.3 (4.9)         0.82 (13.4)
Precursor G      3.1/26.4           11.3 (7.1)         0.84 (10.7)
Precursor H      3.2/26.6           11.4 (6.1)         0.81 (14.8)
Precursor I      2.0/24.8           11.4 (6.1)         1.03 (8.7)
Precursor J      3.0/24.9           11.1 (6.3)         0.94 (12.8)
Precursor K      2.5/19.2           12.7 (7.1)         0.76 (23.7)
Precursor L      3.0/24.6           12.2 (7.4)         0.86 (12.8)
Precursor M      3.0/24.6           11.1 (8.1)         0.81 (12.3)

Trial         Tensile modulus (GPa)   Elongation at break (%)

Precursor A        20.4 (3.4)               12.1 (14.9)
Precursor B        20.2 (4.0)               10.3 (7.8)
Precursor C        22.1 (3.2)               11.4 (10.5)
Precursor D        20.4 (1.5)               10.0 (12.0)
Precursor E        18.4 (3.3)               12.8 (16.4)
Precursor F        18.0 (3.3)               10.8 (18.5)
Precursor G        18.1 (4.4)               11.7 (12.8)
Precursor H        18.4 (3.3)               10.5 (16.2)
Precursor I        18.7 (3.7)               12.8 (13.3)
Precursor J        18.3 (3.8)               13.8 (14.5)
Precursor K        18.3 (4.9)               12.4 (25.8)
Precursor L        18.9 (4.2)               11.8 (18.6)
Precursor M        19.1 (4.2)               10.1 (9.9)

Values in parentheses are coefficient of variation (%).

Individual fibers were tested at 25.4 mm gauge length.

(a) SDR is the spin draw ratio of the fiber (ratio of the take-up
speed of the fiber to the jetting speed of the polymer).

(b) TDR is the total draw ratio of the fiber (SDR multiplied by the
draw ratio at each stage of fiber drawing).

TABLE 3. WAXD structural analysis of as-spun fibers.

                                   Crystal
                   Crystallinity    size
Trial                   (%)         (nm)     [f.sub.PAN]

As-spun Sample A        47           3.4        0.15
As-spun Sample B        49           5.3        0.28
As-spun Sample C        46           2.7        0.13
As-spun Sample D        --           --          --
As-spun Sample E        57           5.6        0.28
As-spun Sample F        57           2.4        0.06
As-spun Sample G        54           3.1        0.10
As-spun Sample H        54           3.2        0.10
As-spun Sample I        52           3.1        0.07
As-spun Sample J        49           5.3        0.25
As-spun Sample K        50           3.4        0.15
As-spun Sample L        54           3.2        0.13
As-spun Sample M        52           3.6        0.11

                   2[theta]-meridional   [d.sub.2[theta] ~
Trial                 peak position        17[degrees]]

As-spun Sample A          40.2                 0.532
As-spun Sample B          40.0                 0.530
As-spun Sample C          40.7                 0.528
As-spun Sample D           --                   --
As-spun Sample E          40.0                 0.526
As-spun Sample F          40.8                 0.534
As-spun Sample G          40.1                 0.526
As-spun Sample H          40.1                 0.525
As-spun Sample I          40.2                 0.525
As-spun Sample J          40.4                 0.525
As-spun Sample K          40.0                 0.527
As-spun Sample L          40.2                 0.527
As-spun Sample M          39.9                 0.525

TABLE 4. WAXD structural analysis of drawn precursor fibers.

                              Crystal
              Crystallinity    size
Trial              (%)         (nm)     [f.sub.PAN]


Precursor A        60          15.2        0.88
Precursor B        61          16.4        0.90
Precursor C        60          15.4        0.90
Precursor D        61          15.7        0.89
Precursor E        59          16.0        0.90
Precursor F        59          16.2        0.89
Precursor G        58          15.2        0.87
Precursor H        59          14.1        0.88
Precursor I        57          14.3        0.88
Precursor J        62          16.3        0.91
Precursor K        58          16.3        0.92
Precursor L        62          16.0        0.90
Precursor M        55          15.4        0.90

              2[theta]-meridional   [d.sub.2[theta] ~
Trial            peak position        17[degrees]]

Precursor A          39.4                 0.527
Precursor B          39.1                 0.525
Precursor C          39.1                 0.526
Precursor D          39.2                 0.524
Precursor E          39.3                 0.529
Precursor F          39.6                 0.527
Precursor G          39.4                 0.527
Precursor H          39.4                 0.527
Precursor I          39.4                 0.531
Precursor J          39.2                 0.525
Precursor K          39.3                 0.526
Precursor L          39.3                 0.526
Precursor M          39.3                 0.525

TABLE 5. Mechanical properties of gel spun PAN based carbon fibers.

                                    Tensile       Tensile
                       Diameter     strength    modulus (a)
Trial                 ([micro]m)     (GPa)         (GPa)

Carbon Fiber A (b)    5.1 (7.8)    4.4 (25.0)       316
Carbon Fiber A1 (c)   5.2 (5.8)    5.8 (24.1)       357
Carbon Fiber A2 (c)   5.0 (6.0)    5.6 (23.2)       375
Carbon Fiber E (b)    6.7 (7.5)    2.6 (26.9)       282
Carbon Fiber H (c)    5.4 (5.6)    3.9 (28.2)       337
Carbon Fiber I (c)    5.9 (5.1)    3.6 (33.3)       345
Carbon Fiber K (c)    6.9 (5.8)    4.0 (22.5)       332
Carbon Fiber L (c)    5.3 (9.4)    4.0 (32.5)       304
Carbon Fiber M (c)    5.8 (5.2)    3.6 (38.9)       342

                       Strain to    Weibull
Trial                 failure (%)   modulus

Carbon Fiber A (b)    1.49 (20.8)    4.48
Carbon Fiber A1 (c)   1.70 (22.9)    4.37
Carbon Fiber A2 (c)   1.64 (22.6)    4.76
Carbon Fiber E (b)    1.00 (23.0)    3.94
Carbon Fiber H (c)    1.25 (26.4)    3.67
Carbon Fiber I (c)    1.15 (30.4)    2.91
Carbon Fiber K (c)    1.31 (19.8)    5.00
Carbon Fiber L (c)    1.49 (35.6)    3.08
Carbon Fiber M (c)    1.13 (37.2)    2.82

Values in parentheses are coefficient of variation (%).

Individual fibers were tested at 25.4 mm gauge length.

(a) Compliance corrected tensile modulus.

(b) Carbon fibers A and E underwent similar stabilization
temperatures, ranging from 200 to 290[degrees]C.

(c) Carbon Fibers A1, A2, H, I, K, L, and M underwent similar
stabilization temperatures, ranging from 180 to 250[degrees]C.

TABLE 6. WAXD structural analysis of PAN and PAN/CNT based carbon
fibers.

                      [d.sub.002]   [L.sub.002]   [d.sub.10]
Trial                    (nm)          (nm)          (nm)

Carbon Fiber A (a)       0.347         1.64         0.210
Carbon Fiber A1 (b)      0.345         1.86         0.210
Carbon Fiber A2 (b)      0.344         1.82         0.210
Carbon Fiber E (a)       0.346         1.72         0.210
Carbon Fiber H (b)       0.345         1.84         0.210
Carbon Fiber I (b)       0.346         1.84         0.210
Carbon Fiber K (b)       0.346         1.91         0.210
Carbon Fiber L (b)       0.346         1.90         0.210
Carbon Fiber M (b)       0.345         1.80         0.210

                      [L.sub.10]
Trial                    (nm)      [f.sub.002]   [FWHM.sub.002]

Carbon Fiber A (a)       2.46         0.86            28.1
Carbon Fiber A1 (b)      2.30         0.87            25.1
Carbon Fiber A2 (b)      2.49         0.87            25.7
Carbon Fiber E (a)       2.16         0.84            29.9
Carbon Fiber H (b)       2.35         0.88            26.3
Carbon Fiber I (b)       2.27         0.88            25.8
Carbon Fiber K (b)       2.37         0.85            28.7
Carbon Fiber L (b)       2.54         0.86            27.1
Carbon Fiber M (b)       2.23         0.87            25.6

(a) Carbon fibers A and E underwent similar stabilization
temperatures, ranging from 200 to 290[degrees]C.

(b) Carbon fibers A1, H, I, K, L, and M underwent similar
stabilization temperatures, ranging from 180 to 250[degrees]C.
COPYRIGHT 2015 Society of Plastics Engineers, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2015 Gale, Cengage Learning. All rights reserved.

 
Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Polyacrylonitrile; carbon nanotube
Author:Newcomb, Bradley A.; Gulgunje, Prabhakar V.; Gupta, Kishor; Kamath, Manjeshwar G.; Liu, Yaodong; Gia
Publication:Polymer Engineering and Science
Article Type:Report
Geographic Code:1USA
Date:Nov 1, 2015
Words:8961
Previous Article:Study on phenolphthalein poly(ether sulfone)-modified cyanate ester resin and epoxy resin blends.
Next Article:High-performance fibers based on copolyimides containing benzimidazole and ether moieties: molecular packing, morphology, hydrogen-bonding...
Topics:

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters