The effect of the field emission on CNTs for N 2 detection Chien-Sheng Huang , Bohr-Ran Huang, Meng-Hsien Tsai, Yu-Min Chuang, Chih-Hong Hsiao Department and Institute of Electronic Engineering, National Yunlin University of Science and Technology, 64002, Taiwan Available online 4 October 2006 Abstract Carbon nanotubes (CNTs) were used as a field emitter and N 2 detector under high vacuum measurement at room temperature. First, the CNTs detected different N 2 concentrations by playing as a resistance sensor. It was found that the sensitivity was about 11% when exposed to 500 Torr N 2 concentration. Then, the CNTs underwent some field emission manipulations for the following times: 1 h, 2 h and 3 h. After such field emission treatment, the CNTs went through the same procedure for N 2 detection and showed a sensitivity of nearly 30% under 500 Torr N 2 concentration. It was suggested that due to the field emission, the CNTs provided more adsorbing sites for N 2 , as shown by FE-SEM and Raman. © 2006 Elsevier B.V. All rights reserved. Keywords: Defects; Field emission; Sensors; Nanotubes 1. Introduction Since their discovery in 1991 [1], carbon nanotubes (CNTs) have been the most promising material due to their high specific surface area and hollow geometry [24]. Existing electrical sensor materials include semiconducting metal oxides [57], silicon-based materials [8,9], organic materials [10,11], and carbon blackpolymer composites [12]. These traditional gas sensors made by using semiconducting oxides are inexpensive, safe and sensitive. For example, they have been widely used for NO 2 and NH 3 detection [1315]. However, they have to be operated at high temperature to enhance chemical reactivity between the material and the gas molecule. This is a major weakness and limits traditional gas sensors for future applica- tions. On the other hand, CNTs acted as a gas sensor with many advantages, and these include fast response, low temperature operation, and high sensitivity. Recently, CNTs have been used to detect various gases, including N 2 , Ar, NH 3 , NO 2 ,O 2 , CO 2 , CH 4 ,H 2 O, etc. Generally, the gas molecule was adsorbed on CNTs at four possible sites: on the external grooves, interstitial channel-site, pore and external surface [16,17]. The reaction mechanism of gas molecule adsorption on CNTs involved two types of phenomena, physisorption and chemisorption. The behavior of physisorption between CNTs and gas molecules was attributed to the Van Der Waals force, and the chemisorption between CNTs and gases to electron transfer, a weak chemical bond. In the literature, a small charge transfer (0.010.035 electron per molecule ) and weak binding (0.2) was reported while the CNTs were exposed to NH 3 and O 2 [18]. Different sites may have preference to physisorption or chemisorption. In this work, we investigated a CNT resistance sensor for N 2 detection. Successful improvement of the sensitivity of CNTs for N 2 detection was achieved by means of field emission manipulation. The sensitivity of CNTs increased from 11% to 30% of such treatment. 2. Experiment The growth of carbon nanotubes was carried out by using thermal chemical vapor deposition (thermal CVD) system. Prior to CNTs growth, a 12 nm Fe thin film acting as catalyst was deposited on (100) p-type silicon substrate by Radio Frequency (RF) sputtering system. Subsequently, the substrate was transferred to the thermal CVD chamber. The Fe-coated Si substrates were loaded on a alumina boat then put inside the thermal CVD quartz tube. The ultimate pressure of the chamber was 10 3 Torr by mechanical pump. The N 2 gas was then introduced into the reaction tube at a flow rate of 100 sccm with 6.5 Torr for 20 min without the introduction of C 2 H 2 . The pre- treated process of the catalyst was maintained at 600 °C. The quartz tube was heated up to the growth temperature of 700 °C, Diamond & Related Materials 15 (2006) 2015 2018 www.elsevier.com/locate/diamond Corresponding author. E-mail address: huangchs@yuntech.edu.tw (C.-S. Huang). 0925-9635/$ - see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.diamond.2006.08.028