Fibers and Polymers 2014, Vol.15, No.4, 762-766

DOI 10.1007/s12221-014-0762-2

Enhanced Spinnability of Carbon Nanotube Fibers by Surfactant Addition Junyoung Song, Soyoung Kim, Sora Yoon, Daehwan Cho1*, and Youngjin Jeong* Department of Organic Materials and Fiber Engineering, Soongsil University, Seoul 156-743, Korea 1 School of Chemical and Biomolecular Engineering, Cornell University, Ithaca, NY 14853, USA (Received July 15, 2013; Revised September 9, 2013; Accepted September 18, 2013) Abstract: A surfactant is used to enhance spinnability of carbon nanotube (CNT) fibers during direct spinning via chemical vapor deposition (CVD). In this study, the non-ionic surfactant, polysorbate, is used due to its good solubility in the CNT synthesis solution. The addition of the surfactant increased the specific strength and electrical conductivity of CNT fibers. Due to these enhanced properties, CNT fibers can be spun at higher speeds which results in lower linear density. These enhancements are due to the reduced agglomeration of iron catalysts during the synthesis of CNT fibers via CVD. This simple approach may create new applications for CNT fibers, such as for artificial muscles and power cables. Keywords: Carbon nanotube, Fiber, Spinnability, Chemical vapor deposition, Surfactant

spinning, one of the wet CNT spinning techniques, is characterized by injecting CNT-dispersed liquid into a polymeric solution and then replacing the dispersant between CNTs by a polymer. This polymer serves to bond CNT particles in aggregate form [12-14]. This coagulation spinning has the advantage of achieving a high CNT content up to 60 wt% and this method allows simple fabrication of CNT assemblies. However, this technique may prevent the inherent properties of CNT from reaching their full potential, because the powdered CNTs are combined with polymers. The other wet assembly technique, liquid-crystalline spinning, can form a liquid crystalline structure using a CNT solution under certain conditions, which organizes CNTs into well aligned structure along a fiber axis [15]. This technique is favorable for producing CNT assemblies with good orientation, but it exhibits a very slow spinning rate. This technique also has difficulty forming liquid crystalline phases of CNTs. The best-known dry technique for fabricating CNT fibers is to draw and twist CNT forests that have been grown vertically on a silicon wafer [16]. With this technique, catalysts are deposited on the wafer and CNTs are grown vertically on the wafer in a high-temperature furnace. This spinning technology is not suitable for mass or continuous production. A preferable method is direction spinning as proposed by Prof. Windle. This enables the CNT assembly to be produced continuously [17]. This method injects a liquid carbon source and catalyst with a carrier gas into a vertical reactor for production of the CNT assembly. The assembly is wound up at the bottom of the reactor. The synthesis of the CNT assembly by direct spinning could dramatically expand the applications of CNTs [17]. However, his studies have focused on the properties of CNTs, such as diameter and number of walls, but not on the CNT fibers themselves [10,11,18]. Thus, the present study investigated a simple method for improving the physical and mechanical properties of the CNT fibers, thereby increasing their spinnability. The improvement is realized simply by adding a surfactant to the CNT synthesis solution.

Introduction CNT has been one of the most promising materials due to its exceptional thermal, electrical, and physical properties and its high aspect ratio [1-3]. To our knowledge, however, there have been few commercialized products developed using CNTs. This lack of success in application is because CNTs are mostly synthesized in powder form, which causes problems of dispersion and difficulties in aligning or placing the CNTs in the desired position. On the other hand, CNT assemblies such as CNT fibers and CNT film are free from these problems, because they rarely require chemical treatment for dispersion, which is commonly applied to powdered CNTs; CNT assemblies are also very easy to handle because of their large size. For example, it is difficult for the powdered CNTs to be utilized in a field emission electron microscope or field emission display, because the tips of the CNTs where electrons are emitted should be vertically aligned to the counter electrode to work effectively as field emitters. The process of vertically aligning nano-sized CNT particles requires a high level of technical skill [4,5], but it should be noted that CNT assemblies can be made into fibers. Therefore, if a carpet is made with the fibers, CNTs will be naturally aligned along the fiber axis and, thus, could be readily used as electron field emitters [6]. This is much easier and less uncertain than the process of aligning CNT particles in the vertical direction. Other applications of CNT fibers include their use in artificial muscles, power cables, high-strength/high-toughness fibers, pressure/damage sensors, and electrical elements [7-10]. CNTs in powder form, however, are not easy to use in those applications, because they exist as nano-sized particles, rather than in a continuous form [11]. CNTs can be incorporated into fiber morphology by two main methods: wet processing and dry processing. Coagulation *Corresponding author: [email protected] *Corresponding author: [email protected] 762

Spinnability of Carbon Nanotube Fibers

Experimental Materials Acetone (99.7 %) was used as a carbon source (Samchun Chemical, Korea). Ferrocene (≥98 %) was used as a catalyst precursor, thiophene (≥99 %) as an activating agent, and polysorbate_20 as surfactants, all of which were purchased from Sigma Aldrich. Synthesis of Carbon Nanotube Assemblies The solutions for CNT synthesis were prepared by mixing acetone 96.0-99.0 wt%, ferrocene 0.2 wt%, thiophene 0.8 wt% and surfactant 0.0-3.0 wt%. The prepared solution was injected with hydrogen gas into a vertical reactor heated to 1200 oC as shown in Figure 1. Upon injecting the solution, the CNT assembly was synthesized instantly within the reactor. The assembly became much denser while passing through a water bath, and the dense assembly was wound up easily as fibers. In this study, the synthesis solution was injected at a rate of 10 ml/hr with a hydrogen gas flow rate of 1000 sccm, and the synthesized CNT fibers were wound up in the range of 7.5 or 9.0 m/min at the bottom of the reactor. Although the principles of direct spinning were described in detail by Li et al. and Zhong et al. [16,17], brief explanation is given as following. When the injected solution reached a zone above decomposition temperature of ferrocene, iron particles were emitted. Iron-sulfide was also formed as sulfur was emitted from thiophene. Then, the iron-sulfide formed a liquid phase, and carbon nanotubes grew on the iron catalysts from when the carbon supplied by the decomposition of acetone was saturated by diffusing into the

Figure 1. Synthesis of CNTs and fabrication of CNT fibers by direct spinning.

Fibers and Polymers 2014, Vol.15, No.4

763

iron-sulfide. CNT assembly continued to form while the solution was continuously supplied into the reactor. Characterization The mechanical properties of the CNT fibers were measured with a universal tensile machine (Instron, model 4464); the gauge length and tension speed of the sample were set to 10 mm and 3 mm/min, respectively. Electrical conductivity was determined by a 4-point probes method using Hioki’s mΩ HiTESTER model. Wide-angle X-ray diffraction (WAXD) data from the CNT fibers was obtained by Bruker GADDS using a 1.54 Å X-ray source for 300 s exposure time in order to observe the orientation of the CNTs along the yarn axis.

Results and Discussion An anion surfactant, polysorbate, was selected as it showed a good solubility in a CNT solution in our previous study [19] where the surfactant reduced the agglomeration of the iron catalysts during the CNT synthesis. Therefore, the CNTs synthesized with surfactant had a better crystalline perfection and smaller diameter [19]. In general, those changes in CNTs are reflected on the properties of CNT fibers such as strength and conductivity. Those changes also affect spinnability since the winding speed is closely related to the quality of CNTs. A fast winding speed is desirable because it aligns the molecular chains along a fiber axis at a higher level of orientation and increases the cohesive forces among the CNTs. Spinnability During the spinning of CNT fibers, winding speed affects their structure and, as a result, it also affects their mechanical and physical properties. To investigate the effect of the addition of the surfactant, two different CNT solutions were prepared. One was prepared without the addition of the surfactant, with a composition of acetone 99.0 wt%, ferrocene 0.2 wt%, and thiophene 0.8 wt%. The other CNT solution was prepared with the addition of the surfactant, with a composition of acetone 98.0 wt%, ferrocene 0.2 wt%, thiophene 0.8 wt%, and polysorbate 1.0 wt%. When the polysorbate was added to the CNT solution, CNT fibers could be wound at a rate of 9.0 m/min. However, 7.5 m/min was the maximum winding speed when the surfactant was not added. This difference in spinnability seems to result from the variation in crystalline quality of the CNTs making up the fibers. As reported in our previous study [19], the addition of the surfactant increased crystalline perfection by reducing the agglomeration of iron catalysts during the CNT synthesis. Also, it is known that the degree of orientation of CNTs within the fiber increases as its winding speed increases as shown in the XRD graph of Figure 2 where CNT assemblies are oriented to the (002) plane [20]. In general, the physical and mechanical properties of synthetic fibers vary depending

764

Fibers and Polymers 2014, Vol.15, No.4

Junyoung Song et al.

Figure 3. Diameters and linear densities of CNT fibers at different polysorbate concentrations. Figure 2. XRD patterns of CNT fibers wound at 7.5 m/min without the addition of polysorbate (a) and fibers wound at 7.5 m/ min (b), and 9.0 m/min (c) with the addition of 1.0 wt% polysorbate, respectively.

on the orientation of their molecules; likewise, the physical and mechanical properties of CNT fibers are affected by the degree of orientation of nanotubes. As confirmed in Figure 2, an increase in the winding speed of CNT fibers led to an increase in the orientation of the fibers’ components, which may affect their mechanical and physical properties. Linear Density and Diameter Figure 3 shows the linear density and diameter measurements of CNT fibers synthesized at different concentrations of polysorbate addition. The samples used for measurements were prepared at the winding speed of 7.5 m/min. Tex is a unit of measure for the linear mass density of fibers and is defined as the mass in grams per 1000 meters. As can be seen in Figure 3, the linear density and diameter of CNT fibers exhibited similar patterns when polysorbate was added. These patterns of CNT fibers were similar to Raman measurements, as reported in our previous study [19]. That is, the crystalline perfection increased by adding polysorbate up to 1.0 wt% as revealed by the Raman analysis, but the crystalline perfection subsequently decreased by increasing polysorbate concentrations above 1.0 wt%. This is attributable to the production of amorphous carbon due to the excessive addition of polysorbate. Consequently, the formation of amorphous carbon increased the diameter of CNT assemblies that caused the increase in their linear density. Tensile Strength and Electrical Conductivity The addition of the surfactant affects both the linear density and the tensile strength of CNT fibers. Figure 4 shows how the tensile strength and conductivity of CNT fibers varied depending on the presence of polysorbate and

Figure 4. Effects of polysorbate addition and winding speed on the tensile behavior (a) and electrical conductivity (b) of CNT assemblies.

winding speed. The CNT fibers fabricated at the same winding speed showed increases of approximately 160 % in specific strength and about 185 % in electrical conductivity

Spinnability of Carbon Nanotube Fibers

Fibers and Polymers 2014, Vol.15, No.4

765

Conclusion

Figure 5. CNT fibers’ specific strength and electrical conductivity versus polysorbate concentration.

when polysorbate was added to the synthesis solution, compared to when no polysorbate was added. The specific strength and electrical conductivity of CNT fibers prepared at a high winding rate with the addition of polysorbate increased to 114 % and 174 %, respectively. These increases were likely caused by the decreased linear density of CNT fibers and the increased cohesive force between CNT, as a result of fast winding. The CNT fibers fabricated at the high winding speed were highly oriented and thus found to have a reduced breaking elongation and an increased modulus of elasticity. This indicated that the surfactant concentration and winding speed were factors that greatly influenced the properties of CNT fibers. To investigate the of the surfactant electrical conductivity in detail, it was measured for the CNT fibers synthesized by varying the surfactant amount from 0 to 3.0 wt%. Figure 5 shows the specific strength and electrical conductivity of CNT fibers wound at the constant winding rate of 7.5 m/ min. As shown in the figure, the two properties were found to have similar behaviors. Both properties increased with the addition of polysorbate up to a concentration of 1.0 wt% but subsequently decreased at higher concentrations. This appears to be closely related to the crystalline perfection and linear density observed above. Typically, the strength of a material is greatly affected by its crystals. Therefore, with the increase in the polysorbate concentration, the enhanced crystalline perfection strengthened the CNT fibers. The electrical conductivity of CNT fibers was also affected by crystalline perfection and the amount of amorphous carbon [21]. This was consistent with the pattern of crystalline perfection below and above a 1.0 wt% of polysorbate concentration, as described above. The formation of amorphous carbon due to the excessive addition of polysorbate probably increases contact resistance and reduced conductivity [22].

This study examined the effects of the addition of the surfactant on the spinnability and electrical conductivity of CNT fibers fabricated by direct spinning. CNT fibers were prepared with the addition of the nonionic surfactant polysorbate, while varying its concentration and winding speed. CNT fibers synthesized with the addition of polysorbate could be wound at a higher speed. Also, electrical conductivity and specific strength were increased by the addition of the surfactant. All of these improvements were due to the synthesis of CNTs with higher crystalline perfection. However, there was an increase of amorphous carbon when excessive surfactant was added, which was an undesirable effect. CNT fibers with the addition of an appropriate polysorbate concentration manifested high specific strength and conductivity, as they could be wound at higher speeds. The addition of the surfactant was a simple and easy way to improve the performance of CNT fibers. This result may contribute to successful attempts to promote the applicability of CNT fibers.

Acknowledgments This work was supported by the Human Resources Development Program (No. 20124010203160) of the Korea Institute of Energy Technology Evaluation and Planning (KETEP) grant funded by the Korea government Ministry of Trade, Industry and Energy.

References 1. S. Iijima, Nature, 354, 56 (1991). 2. P. Avouris, J. Appenzeller, R. Martel, and S. J. Wind, Proceedings of the IEEE, 91, 1772 (2003). 3. W. A. de Heer, A. Chatelain, and D. Ugarte, Science, 270, 1179 (1995). 4. A. Pandey, A. Prasad, J. P. Moscatello, and Y. K. Yap, ACS nano., 4, 6760 (2010). 5. L. Zhu, Y. Sun, D. W. Hess, and C. P. Wong, Nano Lett., 6, 243 (2006). 6. J. M. Bonard, H. Kind, T. Stockli, and L. O. Nilsson, SolidState Electronics, 45, 893 (2001). 7. J. Lee, Y. Jung, J. Song, J. S. Kim, G. W. Lee, H. J. Jeong, and Y. Jeong, Carbon, 50, 3889 (2012). 8. T. Chen, L. Qiu, Z. Yang, Z. Cai, J. Ren, H. Li, H. Lin, X. Sun, and H. Peng, Angew. Chem. Int. Edit., 51, 11977 (2012). 9. J. Foroughi, G. M. Spinks, G. G. Wallace, J. Oh, M. E. Kozlov, S. Fang, T. Mirfakhrai, J. D. W. Madden, M. K. Shin, S. J. Kim, and R. H. Baughman, Science, 334, 494 (2011). 10. W. Lu, M. Zu, J. H. Byun, B. S. Kim, and T. W. Chou, Adv. Mat., 24, 1805 (2012).

766

Fibers and Polymers 2014, Vol.15, No.4

11. M. Zu, Q. Li, Y. Zhu, M. Dey, G. Wang, W. Lu, J. M. Deitzel, J. W. Gillespie, J. H. Byun, and T. W. Chou, Carbon, 50, 1271 (2012). 12. B. Vigolo, A. Penicaud, C. Coulon, C. Sauder, R. Pailler, C. Journet, P. Bernier, and P. Poulin, Science, 290, 1331 (2000). 13. B. Vigolo, P. Poulin, M. Lucas, P. Launois, and P. Bernier, Appl. Phys. Lett., 81, 1210 (2002). 14. S. Zhang, K. K. Koziol, I. A. Kinloch, and A. H. Windle, Small, 4, 1217 (2008). 15. M. D. Lynch and D. L. Patrick, Nano Lett., 2, 1197 (2002). 16. M. Zhang, K. R. Atkinson, and R. H. Baughman, Science, 306, 1358 (2004). 17. Y. L. Li, I. A. Kinloch, and A. H. Windle, Science, 304,

Junyoung Song et al.

276 (2004). 18. R. M. Sundaram, K. K. Koziol, and A. H. Windle, Adv. Mat., 23, 5064 (2011). 19. J. Song, S. Yoon, S. Kim, D. Cho, and Y. Jeong, Chem. Eng. Sci., in Press (2013). 20. H. Zhao, J. H. Kwak, Z. C. Zhang, H. M. Brown, B. W. Arey, and J. E. Holladay, Carbohyd. Polym., 68, 235 (2007). 21. Q. W. Li, Y. Li, X. F. Zhang, S. B. Chikkannanavar, Y. H. Zhao, A. M. Dangelewicz, L. X. Zheng, S. K. Doorn, Z. X. Jia, D. E. Peterson, P. N. Arendt, and Y. T. Zhu, Adv. Mat., 19, 3358 (2007). 22. Q. Li, J. S. Wang, Y. L. Kang, Y. L. Li, Q. H. Qin, Z. L. Wang, and X. H. Zhong, Mat. Sci. Eng. A, 549, 118 (2012).