Titanium dioxide based high temperature carbon monoxide selective sensor Nancy O. Savage a , Sheikh A. Akbar b , Prabir K. Dutta a,* a Department of Chemistry, The Ohio State University, 100 West 18th Avenue, Columbus, OH 43210, USA b Department of Materials Science and Engineering, The Ohio State University, 2041 College Road, Columbus, OH 43210, USA Received 7 July 2000; received in revised form 10 October 2000; accepted 16 October 2000 Abstract The anatase form of TiO 2 has been examined for the sensing of CO and CH 4 at temperatures of 873 K. Though, there were differences in the sensitivity of the anatase sensor towards CO and CH 4 , both gases showed considerable resistance changes. However, in the presence of lanthanum oxide and copper oxide (labeled as ALC sensor), the sensor showed minimal response towards CH 4 , while still exhibiting sensitivity towards CO. The insensitivity towards CH 4 was also con®rmed by measuring the sensor response in the presence of both gases. In order to understand the basis for selective CO sensing, diffuse re¯ectance infrared spectroscopy was carried out on the sensor materials at elevated temperatures. Lanthanum oxide was used to inhibit the anatase to rutile transformation. Infrared spectroscopic data strongly suggest that there is a layer of lanthanum oxide on the titania surface, which acts as a trap for the oxidation products of CO and CH 4 . Upon oxidation of CO on ALC, carbonate species were detected, whereas the reaction of CH 4 produced negligible carbonate species. The insensitivity of the ALC sensor towards CH 4 is proposed to be due to its rapid oxidation by O 2 on the copper oxide. This ef®cient oxidation was responsible for lack of CH 4 reaction on the anatase surface, thus, producing minimal resistance change. CO oxidation also occurred partially on the CuO surface but signi®cant reaction also occurred on the anatase surface and produced a change in resistance. # 2001 Elsevier Science B.V. All rights reserved. Keywords: Semiconducting oxide; Hydrocarbon sensing; Anatase; Oxygen sensing; Infrared spectroscopy 1. Introduction In hydrocarbon-based combustion processes, the ef®- ciency of the process is determined by the completeness of the conversion of hydrocarbons to H 2 O and CO 2 . High temperature combustion of hydrocarbons for energy gen- eration is used in a wide variety of processes, including automobiles and power plants. An incomplete combustion process will lead to the formation of CO and unreacted hydrocarbons (HC). Monitoring these species should pro- vide information useful for feedback control of combustion processes. The development of sensors with speci®c selec- tivity towards CO and HC as well as operation under harsh environments is necessary in this regard. Commercial sensors for ambient monitoring of CO, dri- ven primarily by the unhealthy, and sometimes deadly effects of CO exposure, are available [1]. Many of these sensors use semi-conducting metal oxides, such as SnO 2 [1± 3]. Selectivity towards particular gases is usually accom- plished by incorporating catalysts on these oxides. For example, Pd/SnO 2 is reported to have good sensitivity and selectivity towards CO [2]. For most of these sensors, though, the working temperatures are below 673 K [3]. Our research group has been interested in examining gas sensors that work at higher temperatures, preferably above 773 K. Towards that goal, we have examined the behavior of TiO 2 and have reported several studies on this material [4,5]. Like SnO 2 , TiO 2 is non-stoichiometric at high temperatures and its resistance change upon gas exposure provides the basis for sensing. We have focused on metal oxides as catalysts for improving the selectivity of the sensor. The lower costs and selectivity that can be obtained with metal oxides make them attractive alternates for noble metal catalysts. Our earlier studies have included research on the sensing behavior of anatase and anatase doped with La 2 O 3 and CuO towards CO in an inert environment [4]. Reasons for focusing on the anatase phase in this study is because it has been reported to be an excellent support in catalytic studies and far less is known about its sensing properties, as compared to the rutile phase of TiO 2 [4]. The choice of CuO as a catalyst was based on a study by Larsson et al. that found CuO/TiO 2 to be a more active combustion Sensors and Actuators B 72 (2001) 239±248 * Corresponding author. 0925-4005/01/$ ± see front matter # 2001 Elsevier Science B.V. All rights reserved. PII:S0925-4005(00)00676-6