Page 26 - Carbon Nanotube Fibres and Yarns

Yarn production from carbon nanotube forests 19

Fig. 2.4 SEM images of CNT forests with different levels of drawability: (A) forest of poor
spinnability showing tortuous CNTs, (B) forest with intermediate drawability, and (C)
highly spinnable forest showing high level CNT alignment [27]. (D)–(F) SEM images of a
CNT forest at different magnifications [28]. (Panels (A–C) reprinted with permission from L.
Zheng, G. Sun, Z. Zhan, Tuning array morphology for high-strength carbon-nanotube fibers,
Small 6 (2010),132–137. Panels (D–F) reprinted with permission from W. Cho, M. Schulz,
V. Shanov, Growth and characterization of vertically aligned centimeter long CNT arrays,
Carbon 72 (2014) 264–273.)

CNT alignment can be characterized either in terms of directionality using
a power spectrum [20] or small-angle X-ray scattering (SAXS) [30] or a
tortuosity factor calculated from the geometry of individual tubes [19, 20].
Narrow size distribution of catalysts, high nucleation density, and very
clean surfaces of CNTs are required for good drawability [31]. The drawabil-
ity of CNT forests is closely related to the morphology of the CNT arrays
and can be altered, for example, by adjusting catalyst pretreatment time
[32] or by introducing a small amount of hydrogen during CVD growth
[27]. The shortest catalyst pretreatment time led to CNT arrays with the
best drawability, while prolonged pretreatment resulted in coarsening of
catalyst particles and non-drawable CNT forests [32]. Well-aligned CNT
arrays were obtained from the hydrogen-assisted growth, and wave-like ar-
rays could be obtained from the oxygen-assisted growth [27]. Huynh and
Hawkins [10] and Kim et al. [25, 33] investigated the influences of cata-
lyst deposition and synthesis conditions on CNT forest drawability. For a
44-mm internal diameter reactor, the optimum results were obtained by