Hexagonal boron nitride (h-BN) is an electrical insulator with a large band gap of 5 eV and a good thermal conductor of which melting point reaches about 3000 °C. Due to these properties, much attention was given to the thermal stability rather than the electrical properties of h-BN experimentally and theoretically. In this study, we report calculations that the electronic structure of monolayer h-BN can be influenced by the presence of a vacancy defect which leads to a geometric deformation in the hexagonal lattice structure. The vacancy was varied from mono- to tri-vacancy in a supercell, and different defective structures under the same vacancy density were considered in the case of an odd number of vacancies. Consequently, all cases of vacancy defects resulted in a geometric distortion in monolayer h-BN, and new energy states were created between valence and conduction band with the Fermi level shift. Notably, B atoms around vacancies attracted one another while repulsion happened between N atoms around vacancies, irrespective of vacancy density. The calculation of formation energy revealed that multi-vacancy including more B-vacancies has much lower formation energy than vacancies with more N-vacancies. This work suggests that multi-vacancy created in monolayer h-BN will have more B-vacancies and that the presence of multi-vacancy can make monolayer h-BN electrically conductive by the new energy states and the Fermi level shift.
28 January 2015 in Physical Chemistry Chemical Physics
With the advances in the separation and purification of carbon nanotubes (CNTs), the use of highly pure metallic or semiconducting CNTs has practical merit in electronics applications. When highly pure CNTs are applied in various fields, CNT networks are preferred to individual CNTs. In such cases, the presence of an electrical path becomes crucial in the network. In this study, we report on the electrical percolation thresholds of semiconducting single-walled carbon nanotube (s-SWCNT) networks, and their electrical characteristics in field-effect transistors (FET). Using the Monte Carlo method, s-SWCNT networks were randomly generated in the channels defined by the source–drain electrodes of the FET. On the basis of percolation theory, the percolation thresholds of s-SWCNT networks were obtained at different channel lengths (2, 6, and 10 μm) by generating random s-SWCNT networks 100 times. The network density corresponding to the electrical percolation threshold was theoretically gained at each channel length. As a result, the network densities at the percolation thresholds for the channel lengths of 2, 6, and 10 μm were 6.8, 9.0, and 9.9 tube μm−2, respectively. In addition, SPICE calculations were performed for each s-SWCNT network, constituting an electrical path between the source and the drain electrodes of the FET. In all channel lengths, the on/off ratio of the s-SWCNT networks was enhanced with increasing network density. Finally, we found a power law relationship between the on/off ratio of the s-SWCNT networks and the network density at the percolation threshold.
17 July 2015 in Physical Chemistry Chemical Physics
In this work, an easy method to etch monolayer graphene is shown by catalytic oxidation in the presence of ZnO nanoparticles (NPs). The catalytic etching of monolayer graphene, which was transferred to the channel of field-effect transistors (FETs), was performed at low temperature by heating the FETs several times under an inert gas atmosphere (ZnO + C → Zn + CO or CO2). As the etching process proceeded, diverse etched structures in the shape of nano-channels and pits were observed under microscopic observation. To confirm the evolution of etching, current–voltage characteristics of monolayer graphene were measured after every step of etching by catalytic oxidation. As a result, the conductance of monolayer graphene decreased with the development of etched structures. This decrease in conductance was analyzed by percolation theory in a honeycomb structure. Finally, well-patterned graphene was obtained by oxidizing graphene under air in the presence of NPs, where Al was deposited on graphene as a mask for designed patterns. This method can substitute graphene etching via carbon hydrogenation using H2 at high temperature.
We report that the decoration of metallic single-walled carbon nanotube (m-SWCNT) networks with cobalt(II) oxide (CoO) can improve the electrical conductivity of the networks. To measure the electrical conductivity, we prepared m-SWCNT networks between the source and drain electrodes of field-effect transistors (FETs). Then, the amount of CoO nanoparticles (NPs) used for decoration was controlled by treating the FETs with different volumes of a solution containing Co(NO3)2·6H2O. Atomic force microscopy imaging showed that CoO NPs were intensively deposited on the intertubular junction of the m-SWCNT networks. X-ray photoelectron spectroscopy confirmed that the oxidation state of the Co element on m-SWCNT was CoO. Raman spectra revealed that heavy decoration of CoO increased the D-band intensity of the m-SWCNT, indicating that the CoO NPs disordered the sp2 hybridized carbon atoms of the m-SWCNT via decoration. The electrical conductivity of the m-SWCNT networks was enhanced up to 28 times after decoration, and this was attributed to the CoO NPs connecting the m-SWCNTs at junctions of the networks.