Search

We propose a reliable and accurate method that uses ultraviolet–visible–near infrared absorbance to measure the isotropic-cloud point (ICP) of carbon nanotube (CNT)/chlorosulfonic acid (CSA) solutions. This information is necessary to improve understanding of the isotropic-to-nematic transition of CNT/CSA solutions and to guide solution processing and characterization of CNTs. In this method, the ICP of CNT/CSA solutions is defined as the CNT concentration at which the absorbance at 600 nm is saturated. This wavelength was selected as suitable by considering the resonance ratio and mass absorption coefficient. The reliability of the method was tested by comparing the ICPs measured using this method to those measured by a method that uses polarized optical microscopy and by a method that uses centrifugation. The results obtained using the three methods were consistent. An inverse proportionality between the ICP of CNT/CSA solutions and the CNT aspect ratio (AR) was confirmed; this result further confirms the reliability of our method. The AR of unknown CNTs can be predicted from the ICP of a CNT/CSA solution. Hence, this method provides a new tool to estimate the AR of unknown CNTs.

The yield and length of carbon nanotube (CNT) forests were increased by using regenerated quartz tubes. Carbonaceous impurities remained in the walls of the regenerated quartz tube. When CNT forests were synthesized in regenerated quartz tubes, the carbonaceous impurities were released at temperatures between 400 and 450 °C, and by facilitating nucleation, caused increase in the number density (NCat) of Fe catalyst particles on the substrate. The substrate in the regenerated quartz tube had NCat = 6.0 × 1010/cm2, whereas a fresh quartz tube had NCat = 7.7 × 109/cm2. Furthermore, after the early stage (5 s) of synthesis process, the volumetric density of the CNT forest was higher in the regenerated quartz tube (0.016 g/cm3) than in the fresh quartz tube (0.011 g/cm3). When a regenerated quartz plate was placed in a fresh quartz tube, CNT forests were longer than when a fresh quartz tube was used alone. These results suggest that the carbonaceous impurity in the quartz tube establishes an environment that increases the reproducibility of synthesis of long CNT forests.

To increase the length and coherence of carbon nanotube (CNT) forests that are grown by chemical vapor deposition, we implant Fe ions into an Al2O3 supporting layer to weaken the gradient of Fe concentration when Fe catalyst layer is deposited on it. The weakened gradient slows loss of Fe catalyst nanoparticles (NPs) by sub-surface diffusion into the Al2O3 layer. Therefore, Fe catalyst NPs were well maintained on the Fe-implanted substrates after growth for up to 660 min, whereas significant loss of Fe NPs occurred due to sub-surface diffusion into the Al2O3 supporting layer on the non-implanted substrates. The lifetime of the catalyst was extended to 660 min (Fe-implanted substrate) from 540 min (non-implanted substrate), and the lengths of CNTs in the forest were increased to 20.3 ± 1.0 mm (Fe-implanted substrate, maximum 21.5 mm) from 15.8 ± 1.9 mm (non-implanted substrate). This study clearly demonstrates that sub-surface diffusion of Fe catalyst NPs into the supporting layer is the mechanism that limits CNT forest growth and may provide insights into development of methods to design an “immortal” catalyst to grow CNT forests.

The direct spinning process is a useful method for continuously synthesizing carbon nanotube (CNT) fibers. A reactor should be designed based on reactor design principles in order to study the synthesis mechanism and mass production of CNT fibers. The internal temperature in a continuous flow reactor depends on the length of the reactor, and because the temperatures required for the decomposition of various feedstocks differ, each feedstock is decomposed at a different point in the reactor. Here, we report that high-quality CNT fibers were continuously synthesized by water injection at temperatures above the catalyst-forming range during the direct spinning process, as determined by the decomposition temperatures of the feedstocks. The decomposition temperatures of the feedstock materials were measured. Thiophene and toluene decomposed at 500–600 °C and 600–700 °C, respectively, and these temperatures were lower than those reported in the literature. The quality of the CNT fibers was improved by injecting water into zones with temperatures that exceeded the decomposition temperatures of the feedstocks in the reactor. Oxygen containing functional groups formed less often on the CNT surfaces, further improving the quality of CNT fibers. Moreover, the CNT fibers were continuously synthesized, even if excess water were injected.

Carbon nanotube (CNT) fibers were strengthened by wet stretching with 1-Methyl-2-pyrrolidinone (NMP) or chlorosulfonic acid (CSA) as the wetting liquid, and the change of hierarchical structure of CNT fibers was observed at a nanometer level. Two kinds of voids were present in a CNT fiber: inter-bundle void and intra-bundle void. Wet stretching was a very effective method to aggregate CNT bundles and to remove the voids in a CNT fiber. CNT bundle thickness and the number of CNTs in a bundle increased and the bundle-to-bundle distance decreased as a result of the aggregation of the bundles. After infiltration of liquids, CNTs with elliptical and non-circular cross-sections appeared and the distance between CNTs in a bundle i.e., intra-bundle voids, decreased. It was confirmed that the stretch alone did not affect the intra-bundle void. CSA infiltrated CNT bundles better than NMP did, so CSA yielded better densification than NMP. Load at failure, specific strength (SS) and tensile strength of CNT fibers increased with wet-stretching. SS of the CNT fiber increased from 0.95 to 2.19 N/tex (CSA13%). Understanding structure and morphology of CNT fibers in nanoscale will be helpful to build better strengthening strategies.

We report the synthesis of carbon nanotube (CNT) forests with a narrow diameter distribution based on Fe ion implantation method. By annealing the Fe-implanted SiO 2 /Si wafer in an Ar atmosphere at 800 °C for 15 min, the Fe particles on the surface of SiO 2 layer are successfully formed by the diffusion of Fe atoms from the SiO 2 layer. Interestingly, the size distribution of Fe catalyst particles for Fe-implanted SiO 2 /Si wafers does not change with the prolonged annealing durations of up to 12 h. Using secondary ion mass spectroscopy and transmission electron microscopy (TEM), we confirmed that the implanted Fe atoms diffuse out of the SiO 2 layer and form Fe particles on both the SiO 2 surface and the interface between SiO 2 and Si. The cross-sectional TEM images indicate that the Fe catalyst particles are anchored in the SiO 2 layer, which limits the particles' mobility and results in an invariant catalyst size distribution for prolonged annealing durations. Therefore, we anticipate that implantation can be an efficient alternative catalyst preparation method for CNT forest growth which can solve various growth issues that are inherently caused by conventional physical vapor deposition method.

The optimum synthesis conditions for carbon nanotube (CNT) fibers were investigated using the Design of Experiment (DOE) technique. Direct spinning processes are governed by a variety of experimental factors: the methane flow rate, ferrocene flow rate, sulfur flow rate, hydrogen flow rate, water flow rate, and reaction temperature. The process was optimized in two stages that addressed first the Fractional Factorial Design (FFD) and then the Response Surface Methodology (RSM). Results from each experiment were classified according to a 6-step rating system: nothing(1), black gas(2), dust(3), ribbon or film(4), fiber(5), or continuous fiber(6). In the first step, three major factors (methane, sulfur, temperature) were identified as important among the six experimental factors tested using FFD. The effects of the major factors and the interactions were analyzed through the main effect plot and the interaction plot. In the second step, the experimental conditions were optimized using a model equation derived from Box-Behnken design experiments. Finally, the CNT fibers were continuously synthesized under the optimum conditions. The synthesized CNT fibers mainly consisted of single-walled CNTs (SWCNTs) 1.2–3.8 nm in diameter. The I G /I D ratio of the CNT fibers was 48. This work provides a useful methodology for synthesizing the CNT fibers.

We investigate the effects of a SiO2 sub-supporting layer on the growth of carbon nanotube (CNT) forests, especially on the spinnability of the CNT forest into a CNT yarn, which is one of the most promising application areas based on CNTs as a light-weight, strong, and electrically and thermally conductive macro-scale material. So far, most spinnable CNT forest growths have been performed using Fe/Al2O3 deposited on a Si wafer with a thermally grown SiO2 layer. However, only a few studies have focused on examining the effects of the SiO2 sub-supporting layer on the growth and spinnability of CNT forests by microscopic analyses. Herein, using atomic force microscopy (AFM), transmission electron microscopy (TEM), liquid contact angle, and ellipsometry measurements, we demonstrate that the presence of a SiO2 sub-supporting layer significantly affects the structure of Al2O3, adhesion between Al2O3 and Fe layers, and the number density of Fe catalyst particles, thereby strongly affecting the growth and spinnability of CNT forests. This study opens up new possibilities for accurately controlling the growth of CNT forests by proper designing of the sub-supporting layers.