Porous Indium Oxide Nanotubes: Layer‐by‐Layer Assembly on Carbon‐Nanotube Templates and Application for Room‐Temperature NH3 Gas Sensors

Abstract

Indium oxide (In2O3), as an n-type and wide-bandgap semiconductor, is of great interest for use in toxic-gas detectors, solar cells, and light-emitting diodes because of its high electrical conductivity and high transparency. In particular, In2O3 has been extensively applied in film-based chemical sensors for a long time. However, In2O3-film-based sensing devices possess several critical limitations such as a limited maximum sensitivity and high operation temperatures (200–600 °C). In2O3 nanostructured materials, possessing ultrahigh surface-to-volume ratios, are expected to be superior gas-sensor candidates that may overcome the fundamental limitations as mentioned above. Therefore, considerable efforts have been devoted to synthesizing In2O3 nanostructures such as nanoparticles, nanowires, nanotubes, and nanobelts. Among them, nanotubes are believed to be one of the most promising structures for chemical sensors because of their higher surface-to-volume ratios and, moreover, they do not aggregate as easily as nanoparticles. Up to now, template-assisted approaches have been widely used to synthesize metal oxide nanotubes. Metal oxide nanotubes prepared by template-assisted approaches possess higher surface-to-volume ratios than those prepared by template-free approaches because of their polycrystalline and porous structure, and, therefore, may display a more superior gas-sensor performance. As a result, owing to the simplicity in the synthesis of nanotubes and their availability, quite a few metal oxide nanotubes have been fabricated by nanoporous alumina template assisted approaches such as Ga2O3, In2O3, TiO2, and Fe2O3. [8] Nevertheless, there are some disadvantages in using nanoporous alumina as a template to synthesize metal oxide nanotubes. Firstly, mass production of metal oxide nanotubes by such an approach is impractical, which is one of the bottlenecks for their wide application. Secondly, it is very difficult to completely remove the nanoporous alumina template. Thirdly, the diameters of the prepared metal oxide nanotubes by such an approach are usually larger than 100 nm. Recently, carbon nanotubes (CNTs) have been considered to be an ideal template for the synthesis of metal oxide nanotubes, which can circumvent the disadvantages of nanoporous alumina as mentioned above. For example, Rao and co-workers first fabricated ZrO2, Al2O3, V2O5, SiO2, and MoO3 nanotubes by a metal-alkoxide-based sol–gel process using CNTs as templates in combination with subsequent calcination. However, the deliquescence, toxicity, and high cost of metal alkoxides, as well as the long reaction time, restrict the practical applications of this approach. Liu and co-workers reported the synthesis of Fe2O3/CNT core–shell nanostructures and polycrystalline Fe2O3 nanotubes by a supercritical-fluid-approach using CNTs as templates. Unfortunately, this approach needed to be carried out at high temperature and pressure. In addition, metal oxide/CNT core–shell nanostructures and metal oxide nanotubes have been obtained by CNT-template-assisted chemical vapor deposition (CVD), which was also carried out at high temperature and, moreover, only resulted in the deposition of oxides on the top surface of the CNTs. Metal oxide/CNT core–shell nanostructures were also fabricated by a chemical precipitation method. However, in this route, the formation of metal oxide nanoparticles in the solution or metal oxides with a very large grain size on the surface of the CNTs was inevitable, which made it difficult to form metal oxide nanotubes after oxidation of the CNTs. We report a novel and versatile approach to synthesize metal oxide nanotubes using layer-by-layer (LBL) assembly on the CNT templates in combination with subsequent calcination. LBL assembly is based on the electrostatic attraction between charged species and it has been widely used to synthesize polymeric multicomposites, inorganic and hybrid hollow spheres, polymer nanotubes, and core–shell nanostructures. We now present its use, for the first time, to synthesize metal oxide nanotubes including In2O3, NiO, SnO2, Fe2O3, and CuO. Of these, In2O3 nanotubes are used to illustrate the basic idea underlying the approach presented in this work. The as-synthesized In2O3 nanotubes were applied in an NH3 gas sensor operated at room temperature, which exhibits improved performance and thus promising applications. C O M M U N IC A IO N