Effect of dynamic precursor gas pressure on growth behavior of amorphous Si–C–O nanorods by electron beam-induced deposition

Abstract

Electron beam-induced deposition (EBID) is a versatile maskless technology [1] to fabricate submicronor nanometerscale structures from various elements in scanning electron microscope [2] as well as transmission electron microscope [3] and scanning transmission electron microscope [4]. During EBID, adsorbed precursor gas molecules on a substrate surface are irradiated and dissociated by an electron beam. This induces a chemical reaction that results in the deposition of non-volatile materials. If the electron beam is not moved relative to the substrate, a nanodot can be formed; electron beam scanning can produce nanorod within a suitable precursor gas pressure. Nanofabrication by using EBID is hence exceeding in terms of controlling the position and morphology of low dimensional functional nanostructures, e.g., carbon nanotube and ferromagnetic FePt alloy nanorods [5, 6]. It is very necessary to probe the effects of various deposition parameters on the target deposits, e.g., electron beam scan speed, precursor gas pressure, and deposition time, in order to fully develop and improve EBID serving as nanotechnology. No much importance, however, has been attached to dynamic precursor gas pressure in EBID, although it is generally known that the higher the precursor gas pressure is, the more the decomposed gas molecules are, and the larger the size of the deposits should be during EBID [7–9]. Undoubtedly, it will be the most direct route to know about the transport of precursor gas molecules by real-time monitoring (dynamic) precursor gas pressure, which helps to unravel the microscopic effect of gas pressure on the successive nanofabrication. In this paper, we will investigate the effects of dynamic precursor gas pressure and electron beam scan speed on the EBID process. Electron beam-induced deposition experiment was carried out in a 30 kV field emission gun scanning electron microscope (JEOL JSM-7800UHV). The electron beam current was about 0.8 nA with a beam diameter of 4 nm. The irradiation position and time of electron beam were controlled by an external voltage input to the beam deflectors using a computer with digital–analog converters. Tetraethoxy-silane SiOC2H5 was used as a precursor. A gas introduction system consists of a nozzle (its tip was about 1 mm away from the electron beam position, as shown in Fig. 1), gas pipeline, variable leak valve, and gas source reservoirs. The base gas pressure in the specimen chamber was 2 9 10 Pa and the maximum allowable gas pressure is 2 9 10 Pa. A holey carbon film was used as substrates. A 300 kV JEM-3000F field-emission gun transmission electron microscope, attached with a post-column Gatan imaging filter and a 1 k 9 1 k Ultrascan CCD camera, was employed to characterize the deposits by a high-resolution transmission electron microscopy (HRTEM) and energyfiltered transmission electron microscopy (EFTEM). To obtain elemental maps, the 3-window method [10] was applied, where the intensities from the first two windows (pre-edges 1 and 2) and third window (post-edge) are used for the background subtraction and the map, respectively. The sample was stabilized for 60 min before recording, in order to minimize the possible mechanical drift of the sample stage. When the focused electron beam position was moved into space at a certain speed (here 1.2–3 nm/s) along the edge of holey carbon film, freestanding rod-like nanostructures W. Zhang (&) K. Furuya National Institute for Materials Science, 3-13 Sakura, Tsukuba 305-0003, Japan e-mail: phdweizhang@gmail.com