Nanochannels’ fabrication using Kirkendall effect Aurelian Marcu a, * , Takeshi Yanagida b , Tomaji Kawai b a National Institute for Laser Plasma and Radiation Physics, Laser Department, Atomistilor 409, Bucharest-Magurele, Romania b Institute of Scientific and Industrial Research, Division of Advanced Materials Science and Technology, Osaka University, 8-1 Mihogaoka, Ibaraki, Osaka 567-0047, Japan article info Article history: Received 23 January 2009 Received in revised form 10 April 2009 Accepted 24 April 2009 Available online 8 May 2009 Keywords: Nanowire Nanochannel Kirkendall effect Core–shell structures abstract Evidences of nanochannel formation based on Kirkendall effect have been previously reported for oxide nanowires covered with a thin alumina shell layer. Here we will investigate the nanochannel formation on an in situ pulsed laser deposition (PLD) fabricated structure of iron oxide shell layer over ZnO and MgO nanowire core and will compare with the alumina shell layer results. In all (four) cases a chemical reaction takes place on the interface producing a spinel buffer layer. Nanochannel formation process could be understood based on material diffusion coefficients through the spinel buffer layer but shell layer crystal structure seems to play a significant role. Ó 2009 Elsevier Masson SAS. All rights reserved. 1. Introduction Progresses in nanotechnology include devices’ miniaturisation down to micrometers and even to nanometer scale while recent advance of the nanobiotechnology [1,2] increases the interest in nanopipes and nanofluidic channels as tools for DNA molecules’ manipulation and investigation. Top-down technologies such as e-beam lithography [2,3], focused ion beam [4] or transmission electron microscopy (TEM) techniques [5] were continuously developed to fabricate nanostructures down to several nanome- ters. However, for nanopipes of sizes below several nanometers, bottom-up approach seems to be a much more promising option from both a technological and an economical point of view. In our case, the bottom-up approach is based on the asymmetric diffu- sion process in some material interfaces, also known as the Kirkendall effect. Thus, due to the fact that the materials are not going to diffuse symmetrically into one another, we obtain an effective material movement and void spaces remain behind. By controlling this movement we can control structure geometry and in our case, inside the cylindrical structure of the nanowire, a nanochannel formation. Using this technique nanochannel fabrication inside ZnO nanowires covered with an Al 2 O 3 shell layer has been reported, even if some undiffused areas remained at the bottom of the structures [6]. Attempts have been made for MgO core wires with Al 2 O 3 shell layer [7] but a wet etching had to be used to remove the remaining MgO core and, even so, some parts of the channels were still not continuous. The aim of the present study is to investigate the nanochannel formation for a Fe 2 O 3X shell layer using both ZnO and MgO core wires and to compare the results and diffusion process with the Al 2 O 3 shell layer case. 2. Experiment MgO and ZnO nanowires were grown by vapor–liquid–solid (VLS) method using laser ablation technique. We used a Lamda Physics ArF pulsed excimer laser for ablation of MgO and ZnO targets and we grew nanowires on MgO and alumina substrates in an oxygen atmosphere. We used Au as liquid catalyst. More details about growth techniques have been given elsewhere [8]. Furthermore, we covered these structures with a 5–50 nm thick shell layer using in situ PLD method. By changing the target ablation and deposition conditions we optimized shell layer morphology and crystal structure and properties, case by case [9]. The next step was to warm up the samples at various temper- atures and time intervals in order to observe the diffusion process and nanochannels’ formation based on the differences in the diffusion process known as the Kirkendall effect. We investigated temperatures up to 1500  C and heating intervals up to 48 h. The morphology results were investigated by SEM (EFSEM JEOL JSM- 6330FT) and TEM (HRTEM JEOL JEM-3000F). For a better inside * Corresponding author. Tel./fax: þ40 21 457 4027. E-mail address: marcu@ifin.nipne.ro (A. Marcu). Contents lists available at ScienceDirect Solid State Sciences journal homepage: www.elsevier.com/locate/ssscie 1293-2558/$ – see front matter Ó 2009 Elsevier Masson SAS. All rights reserved. doi:10.1016/j.solidstatesciences.2009.04.034 Solid State Sciences 12 (2010) 978–981