Letter pubs.acs.org/NanoLett
Meter-Long Spiral Carbon Nanotube Fibers Show Ultrauniformity and Flexibility Yuanyuan Shang,*,† Chunfei Hua,† Wenjing Xu,‡ Xiaoyang Hu,§ Ying Wang,† Yu Zhou,† Yingjiu Zhang,† Xinjian Li,† and Anyuan Cao*,‡ †
School of Physical Engineering, Zhengzhou University, Zhengzhou, Henan 450052, China Department of Materials Science and Engineering, College of Engineering, Peking University, Beijing 100871, People’s Repubic of China § Collage of Science, Henan Institute of Engineering, Zhengzhou, Henan 451191, China ‡
S Supporting Information *
ABSTRACT: Conventional straight fibers spun from carbon nanotubes have rather limited deformability; creating a spiral structure holds the promise to break this shape restriction and enhance structural flexibility. Here, we report up to one meterlength threads containing purely single-walled nanotubes twisted into spiral loops (about 1.3 × 105 loops per meter) with tunable fiber diameters and electrical conductivity. Because of significant increase of the loop number and long-range homogeneity, the fibers display many unique properties (e.g., self-shrinking and forming extremely entangled structure, fast stretching with great resilience, large-degree axial and lateral deflection, and excellent fatigue resistance) that are difficult to achieve in straight yarns or short helical segments. They also have potential applications as macroscopic fiber-shaped temperature sensors and deformable gas sensors. Our long spiral fibers may be configured into versatile structures such as nanotextiles for developing wearable electronics and multifunctional fabrics. KEYWORDS: Spiral fiber, carbon nanotube, nanotextile, fiber-shaped device, gas sensor arbon nanotube (CNT) fibers are emerging highperformance nanotextiles with superior properties than traditional polymeric or carbon fibers.1−6 Fabrication of macroscopic, continuous CNT fibers and yarns has promoted many exciting applications including twist-yarn based artificial muscles working under various modes (coupling electrical, optical, chemical, and mechanical effects),7 robust electrical wires sustaining large current densities,8 as well as a variety of fiber-shaped energy devices such as solar cells, supercapacitors, and batteries.9−14 One of the most desired applications would be wearable electronics and smart nanotextiles, in which the fiber flexibility is an important consideration. In general, thin CNT fibers can be bent to large angles, knit into simple knots, or woven into fabrics without much degradation in mechanical properties. However, the deformability of most straight fibers is rather limited. Considering a simple model in which a linearly elastic beam is fixed at its two ends and subjected to a concentrated load (F) applied on the middle, the maximum deflection (Δ) within the elastic range can be described as
C
FL3 Δ= 48EI
,where L is the beam length, E is the Young’s modulus, and I is the moment of inertia.15 The intrinsic fiber (filament) flexibility, as defined by Δ3 , is dependent on the material FL
properties (modulus) as well as the structural size (e.g., diameter of a fiber with circular cross-section). For straight CNT yarns, it does not matter if they are made from solution extrusion or dry-spinning methods, their deformation and intrinsic flexibility are still restricted by the above formula. Too large deformation in both axial and lateral directions would cause slippage between CNT components and fracture of the fiber. Recently, our group has reported helical (or double-helix) single-walled nanotube (SWNT) yarns containing close arranged microloops, and applications as rotational actuators and stretchable supercapacitors.16−19 Such structure holds the promise to break the restriction of fiber deformation depicted in eq 1, because the helical loops lead to axial stretchability and reduced rigidity as the loop separation could accommodate Received: November 23, 2015 Revised: February 1, 2016
(1) © XXXX American Chemical Society
A
DOI: 10.1021/acs.nanolett.5b04773 Nano Lett. XXXX, XXX, XXX−XXX
Letter
Nano Letters
Figure 1. Meter-long spiral SWNT fibers and unique properties. (a) Photos showing the spinning process from a SWNT film (3 m in length and uniform width) to a spiral fiber (final length of 94 cm) by overtwisting. Camera photos on the left end, middle, and right end of the fiber only show a black line and cannot reveal the very thin spiral structure. (b) Self-shrinking of a 80 cm-long spiral fiber into smaller lengths (20 and 10 cm, respectively) and finally forming an entanglement. (c) A 12 cm-long spiral fiber was fixed on a balloon and elongated into a length of 20 cm accompanying the balloon expansion. (d) A 52 cm-long spiral fiber fixed at two ends and deflected by one or multiple metal cylinders laterally to different configurations (e.g., zigzag). (e) A 4 cm fiber segment producing light emission under a dc voltage of 14 V.
fibers. Furthermore, we demonstrate macroscopic temperature and deformable gas sensors based on those spiral fibers, indicating potential applications toward functional nanotextiles. Results and discussion. We adopted a dry-spinning process on as-grown SWNT films and involved overtwisting to encode helical loops into the resulting yarn, as described in our previous report.16 Here, we show that it is possible to scale the fabrication process from centimeter-long yarns to meterscale while keeping an ultrauniform microstructure. The key point is the preparation of very long SWNT films (>2 m) with uniform width (1−3 cm) as the raw material through a modified chemical vapor deposition (CVD) process (see Experimental Section for details). As shown in Figure 1a, a suspended narrow film has an initial length of 3 m and uniform width of 1.5 cm, which is important to ensure the formation of constant size microloops upon spinning. At last, a 94 cm-long
large tensile strains and other deformation. Therefore, the helical structure adds a further variable for tailoring and improving the intrinsic flexibility of CNT yarns. The question is whether such helical structures can be synthesized at a much greater scale in order to explore novel properties and facilitate practical applications. Here, we successfully spin meter-long purely SWNT fibers containing as much as 1.3 × 105 (per meter) spiral loops, compared with previous short helical segments (200 μm (Figure 3c). The linear and volumetric densities of our spiral fibers are 1.06 × 10−5 g/cm and 1.51 g/cm3 (for d = 14 μm), 1.14 × 10−4 g/cm and 1.45 g/cm3 (d = 50 μm), and 3.15 × 10−4 g/cm and 1.27 g/cm3 (d = 100 μm), respectively. Larger-diameter fibers usually have lower volumetric density, indicating that thicker fibers contain more porous space per unit volume (being less dense). As a result, the electrical conductivity is enhanced in thinner fibers. For mechanical properties, long spiral fibers cut into segments with lengths of 9−200 mm generally reach tensile strains (ε) of about 75% in uniaxial tension and ultimate stresses (σ) in the range of 82 to 142 MPa (calculation based on d = 14 μm and a linear fiber density of 1.06 × 10−5 g/cm) (Figure 4a). Elastic recovery and reversible resistance change are observed during cyclic stretching to 40% strain (Figure S2), as also reported in our previous work.16,17 Dynamic test was performed by oscillating a prestretched fiber at a constant displacement (set as 2.6 mm, corresponding to a tensile strain of 5%) (illustrated in Figure 4b). At a lower frequency of 0.5 Hz, the load−displacement curves exhibited linear elastic behavior with small hysteresis (Figure 4c). After 1 × 105 cycles, a modest residual deformation of about 0.3 mm was produced as seen from the curves of last 100 cycles, without much decrease in maximum load (Figure S3a,b). Fatigue tests E
DOI: 10.1021/acs.nanolett.5b04773 Nano Lett. XXXX, XXX, XXX−XXX
Letter
Nano Letters
Figure 5. Application as temperature and gas sensors. (a) Recorded resistance when the fiber was heated from 35 to 100 °C (maintained for 6 min) and then cooled down. (b) Relative resistance change (ΔR/R0, R0 is original resistance) when the fiber was heated to different temperatures (75− 150 °C). Each data point was collected and averaged from four heating/cooling cycles. Inset shows four cycles for a heating temperature of 100 °C. (c) Sensing response of the fiber to three gases (NH3, NO, NO2) at different concentrations (10 to 200 ppm). (d) Comparison of sensing response to NO2 at 10 to 200 ppm when the spiral fiber was placed in original, stretched (ε = 100%) and entangled state, respectively. Inset shows different deformation states.
value when the fiber was cooled down. There are both semiconducting and metallic tubes in the SWNT fiber, and the result indicates that metallic tubes (in which electrical conductivity drops upon increasing temperature) may collectively dominate the charge transport through the intertube junctions. Such resistance change was reproducible and reversible at different temperatures (from 75 to 150 °C) and over many cycles (Figure S4a). In particular, the relatively resistance change showed a linear relationship within this temperature range, and could reach ∼14% at T = 150 °C (Figure 5b). Pure SWNTs are stable up to 500 °C in air, thus promising in making high temperature sensors. On the basis of this preliminary result, textiles made from those SWNT fibers might work as smart skin that is sensitive to environmental temperature variation. Carbon nanotubes are promising materials for gas sensing given their large surface area and tunable electrical conductivity upon molecular adsorption. Here, in a two-probe setup with a gas injection system we construct fiber-based gas sensors that also work under highly stretched or deformed conditions. Among several gases that have been tested, the SWNT fibers show rapid response to three of them (NH3, NO, and NO2), which are either toxic to the human body or cause the green house effect. The gas-sensing response (S) is defined as the relative resistance change, S = (Rg − Ra)/Ra, where Ra is the original resistance of the sensor in clean air, Rg is the stabilized resistance upon introducing target gas at a particular concentration (10−200 ppm). The positive S value from NH3 means that the fiber resistance increased upon gas introduction, while NO and NO2 showed negative S with
at other frequencies (0.25−1 Hz) continuously for 1 × 105 cycles showed similar behavior with stable maximum stresses and small residual deformations (Figure 4d). For even higher frequency (5 Hz) and larger displacement (4 mm, ε = 20%), the vibration of the fiber became severe, and the fiber morphology switched between a stretched band and a curved arc-shape rapidly. The structural evolution after cyclic tests was characterized by SEM, and we found that the loops were slightly separated from each other (forming gaps) (Figure S3c,d). Here, the spiral fiber behaves like a tension spring with the nanotube-loops being separated or closed under external loading, which is important for preventing irreversible deformation and thus maintaining the elasticity and mechanical strength. If the yarns had a straight shape, slippage between CNTs would occur during fast and repeated stretching, causing plastic deformation and rapid degradation in stress. Helical CNT fibers have been subjected to static cyclic tests (3000 rpm to spin the SWNT film into a straight yarn and then a spiral fiber by continuous overtwisting. To spin a spiral fiber with final length of about 1 m, the original film should be about 3 m-long and the spinning process takes about 20 min. Shorter fibers take much less time to spin (typically a few minutes). Structural Characterization, Mechanical, and Electrical Measurements. The morphology and microstructure of the spiral fibers were characterized by SEM (Hitachi S4800) and Raman spectroscopy (inVia Reflex). Samples for SEM include
■
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.nanolett.5b04773. Snapshots during spinning of spiral fibers; static mechanical tests coupling electrical measurement; more data and images on fatigue test; reversible temperature sensing and gas sensing of stretched fibers; movies showing the spinning process and manual stretching of the spiral fiber. (ZIP)
■
AUTHOR INFORMATION
Corresponding Authors
*E-mail:
[email protected]. *E-mail:
[email protected]. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS The authors gratefully acknowledge financial support from the National Natural Science Foundation of China under Grant NSFC 51325202, 51502267 and China Postdoctoral Science Foundation funded project (2015M582200).
■
REFERENCES
(1) Vigolo, B.; Pénicaud, A.; Coulon, C.; Sauder, C.; Pailler, R.; Journet, C.; Bernier, P.; Poulin, P. Science 2000, 290, 1331−1334.
G
DOI: 10.1021/acs.nanolett.5b04773 Nano Lett. XXXX, XXX, XXX−XXX
Letter
Nano Letters (2) Jiang, K. L.; Li, Q.; Fan, S. S. Nature 2002, 419, 801. (3) Baughman, R. H.; Zakhidov, A. A.; de Heer, W. A. Science 2002, 297, 787−792. (4) Ericson, L. M.; Fan, H.; Peng, H.; Davis, V. A.; Zhou, W.; Sulpizio, J.; Wang, Y. H.; Booker, R.; Vavro, J.; Guthy, C.; et al. Science 2004, 305, 1447−1450. (5) Li, Y. L.; Kinloch, I. A.; Windle, A. H. Science 2004, 304, 276− 278. (6) Zhang, M.; Atkinson, K. R.; Baughman, R. H. Science 2004, 306, 1358−1361. (7) Lima, M. D.; Li, N.; Andrade, M. J.; Fang, S. L.; Oh, J. Y.; Spinks, G. M.; Kozlov, M. E.; Haines, C. S.; Suh, D. S.; Foroughi, J.; et al. Science 2012, 338, 928−932. (8) Behabtu, N.; Young, C. C.; Tsentalovich, D. E.; Kleinerman, O.; Wang, X.; Ma, A. W. K.; Bengio, E. A.; ter Waarbeek, R. F.; de Jong, J. J.; Hoogerwerf, R. E.; et al. Science 2013, 339, 182−186. (9) Liu, D.; Zhao, M. Y.; Li, Y.; Bian, Z. Q.; Zhang, L. H.; Shang, Y. Y.; Xia, X. Y.; Zhang, S.; Yun, D. Q.; Liu, Z. W.; et al. ACS Nano 2012, 6, 11027−11034. (10) Chen, T.; Qiu, L.; Cai, Z.; Gong, F.; Yang, Z.; Wang, Z.; Peng, H. Nano Lett. 2012, 12, 2568−2572. (11) Zhang, S.; Ji, C. Y.; Bian, Z. Q.; Yu, P. R.; Zhang, L. H.; Liu, D. Y.; Shi, E. Z.; Shang, Y. Y.; Peng, H. T.; Cheng, Q.; et al. ACS Nano 2012, 6, 7191−7198. (12) Ren, J.; Li, L.; Chen, C.; Chen, X. L.; Cai, Z. B.; Qiu, L. B.; Wang, Y. G.; Zhu, X. R.; Peng, H. S. Adv. Mater. 2013, 25, 1155−1159. (13) Chen, X. L.; Lin, H. J.; Deng, J.; Zhang, Y.; Sun, X. M.; Chen, P. N.; Fang, X.; Zhang, Z. T.; Guan, G. Z.; Peng, H. S. Adv. Mater. 2014, 26, 8126−8132. (14) Fang, X.; Weng, W.; Ren, J.; Peng, H. S. Adv. Mater. 2016, 28, 491−496. (15) Finlayson, D. Journal of the Textile Institute Proceedings 1946, 37, P168−P180. (16) Shang, Y. Y.; He, X. D.; Li, Y. B.; Zhang, L. H.; Li, Z.; Ji, C. Y.; Shi, E. Z.; Li, P. X.; Zhu, K.; Peng, Q. Y.; et al. Adv. Mater. 2012, 24, 2896−2900. (17) Shang, Y. Y.; Li, Y. B.; He, X. D.; Du, S. Y.; Zhang, L. H.; Shi, E. Z.; Wu, S. T.; Li, Z.; Li, P. X.; Wei, J. Q.; et al. ACS Nano 2013, 7, 1446−1453. (18) Li, Y. B.; Shang, Y. Y.; He, X. D.; Peng, Q. Y.; Du, S. Y.; Shi, E. Z.; Wu, S. T.; Li, Z.; Li, P. X.; Cao, A. Y. ACS Nano 2013, 7, 8128− 8135. (19) Shang, Y. Y.; Wang, C. H.; He, X. D.; Li, J. J.; Peng, Q. Y.; Shi, E. Z.; Wang, R. G.; Du, S. Y.; Cao, A. Y.; Li, Y. B. Nano Energy 2015, 12, 401−409. (20) Miao, M. H. Carbon 2011, 49, 3755−3761. (21) Yu, D. S.; Goh, K. L.; Wang, H.; Wei, L.; Jiang, W. C.; Zhang, Q.; Dai, L. M.; Chen, Y. Nat. Nanotechnol. 2014, 9, 555−562. (22) Xin, G. Q.; Yao, T. K.; Sun, H. T.; Scott, S. M.; Shao, D. L.; Wang, G. K.; Lian, J. Science 2015, 349, 1083−1087. (23) Wang, J. N.; Luo, X. G.; Wu, T.; Chen, T. Nat. Commun. 2014, 5, 3848. (24) Xu, F. B.; Sadrzadeh, A.; Xu, Z. P.; Yakobson, B. I. J. Appl. Phys. 2013, 114, 063714. (25) Kong, J.; Franklin, N. R.; Zhou, C. W.; Chapline, M. G.; Peng, S.; Cho, K.; Dai, H. J. Science 2000, 287, 622−625. (26) Lu, G.; Ocola, L. E.; Chen, J. H. Adv. Mater. 2009, 21, 2487− 2491. (27) Sadrzadeh, A.; Farajian, A. A.; Yakobson, B. I. Appl. Phys. Lett. 2008, 92, 022103.
H
DOI: 10.1021/acs.nanolett.5b04773 Nano Lett. XXXX, XXX, XXX−XXX
