Abstract

In spite of much research progress in the science and synthesis of carbon nanotubes (CNTs), control over the location and orientation of CNTs on substrates remains a major challenge. A key breakthrough was the synthesis of vertically aligned CNT (VA-CNT) arrays using thermal chemical vapor deposition (CVD) and plasma-enhanced CVD, where the CNTs self-orient perpendicular to the substrate surface due to initial crowding and continue to grow upward in this direction. These CNT arrays have wide-ranging applications, including membranes, heat dissipation, electrical interconnects, and nanoelectronics. The catalysts for synthesis of VACNTs are commonly prepared by sputtering or evaporating a thin metal film onto a substrate, which dewets to form catalyst nanoparticles at an elevated temperature prior to growth. While these catalysts are easily prepared and patterned by shadow masking or lithography, these approaches are not easily able to create nanocluster catalysts that have monodisperse diameters and quantifiable areal densities. In thin metal films, both the nanocluster size and areal density are coupled to the film thickness, and the annealing procedure affects the size, density, and the chemical state of the nanoclusters. Recently, Huh et al. demonstrated a route for controlling the density of CNT growth using varying densities of colloidal cobalt nanoparticles; however, due to nanoparticle coalescence their route does not enable precise quantification of nanocluster areal density and leads to a broad distribution of CNT diameters. Zhang et al. also recently demonstrated the control of CNT growth by varying the density of Co–Mo nanoparticles, although their route is unable to independently vary the diameter and the areal density of nanoparticles. We employ a methodology for synthesizing iron oxide nanoclusters that utilizes micelles formed by the amphiphilic block copolymer, polystyrene-b-poly(acrylic acid) (PS-bPAA). This catalyst system has significant value because it enables the creation of nanocluster arrays of a chosen metal species, with independent control of the nanocluster diameter and areal density. In previous work, nanocluster diameters were varied between 5 and 16 nm and the areal density was varied from 6.0 × 10 to 1 × 10 cm, although variation outside of these ranges is easily accessible. At higher areal densities the nanoclusters are hexagonally ordered. Further, as we have presented separately, the nanocluster arrays can be patterned on the micrometer length scale using microcontact printing. Here, we utilize this system to create arrays of uniform-diameter iron oxide nanoclusters, with quantifiable areal densities that can be varied over more than an order of magnitude. We achieve vertical CNT growth from our catalyst system through appropriate selection of the substrate, catalyst preparation procedure, and reaction conditions. To the best of our knowledge, our work is the first example of vertical growth of CNTs from a block-copolymer-based catalyst. Because this catalyst system allows for precise quantification of the nanocluster areal density, we can estimate the percentage of nanoclusters that nucleate the growth of a CNT. By uniformly varying the areal density of iron oxide nanoclusters on the substrate surface, we manipulate the morphology of the CNT film from a tangled and sparse arrangement of individual CNTs, through a transition region with locally bunched and self-aligned CNTs, to rapid growth of thick vertical CNT films. The procedure for synthesizing the iron oxide nanocluster arrays utilized spherical micelles formed by PS-b-PAA in toluene, which were loaded with FeCl3, and then spin-cast onto a substrate (see Experimental Section). The polymer thin film was then treated in oxygen plasma to remove the organic components, leaving an iron oxide nanocluster array, as shown in Figure 1A. Previous studies have demonstrated that the iron oxide nanocluster structure is Fe2O3. [22,25] The molecular weight and metal-loading ratios of Figure 1A led to iron oxide nanocluster arrays with diameters of 16 ± 1.6 nm that were arranged in a quasihexagonal array with an areal density of 6.0 × 10 cm. The root-mean-square roughness of the nanocluster arrays was 2.2 nm, as determined from using atomic force microscopy (AFM). C O M M U N IC A TI O N S

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call