Directly synthesizing individual CNTs onto a desired site is highly preferred in order to use the unique material properties of individual CNTs for various applications and prevent interactions between CNTs. An individual CNT was synthesized when the 40-nm-diameter aperture was used to pattern the iron catalyst, as shown in the SEM image in Figure 4e. The correlation

between the aperture diameter and the number of CNTs synthesized under the experimental conditions is summarized in Figure 4f. The number selleck chemical of CNTs obviously decreased with decreasing aperture diameter. For example, Trichostatin A price although 39.6% of the CNTs synthesized through the 40-nm-diameter aperture were individual CNTs, the yield for the growth of single CNTs decreased to 19.6% and 8.7% when the 80- and 140-nm-diameter apertures were used, respectively. Furthermore, the yield for the synthesis of two CNTs using the 80-nm-diameter aperture was more than twice compared to that for the synthesis of two CNTs using the other two apertures. Hence, there is a high chance of controlling the number of CNTs synthesized by adjusting the diameter of the aperture used in the nanostencil this website mask. More

results for the number of CNTs synthesized using various aperture diameters are shown in Additional file 1: Figure S3. The diameter of the synthesized CNTs was 10 to 30 Interleukin-2 receptor nm, which indicates that they exhibited a multiwalled structure. It also reveals that the iron catalyst was agglomerated into a size similar to the diameter of CNTs in CVD temperature of 700°C [40–42]. No CNTs were found on approximately 40% of the catalytic sites produced using the three different aperture sizes. It could possibly be from the size deviation in each catalyst pattern, and this would be improved by enhancing the mechanical stability of the stencil mask through the design of corrugated structures [43], by increasing the directionality and the nominal thickness

of the iron catalyst, or by introducing a buffer layer such as aluminum oxide between the catalyst and the silicon substrate to prevent the possible formation of iron silicide. Although our method is not perfect, it retains higher throughput, yield, and scalability than other serial processes used to integrate individual CNTs on specific sites, such as electron beam lithography on dispersed CNTs [10], pick-and-place manipulation [18], and localized synthesis on microheaters [44]. The integrity and throughput of our method are also superior to those of dielectrophoretic assembly [14–17], which is frequently used to integrate individual CNTs. CNTs should be immersed and sonicated in an aqueous solution for dielectrophoresis. This process usually contaminates the CNTs, deteriorating their unique material properties.