Date of Award

8-2003

Level of Access Assigned by Author

Open-Access Thesis

Degree Name

Doctor of Philosophy (PhD)

Department

Chemistry

Advisor

Carl P. Tripp

Second Committee Member

Bruce L. Jensen

Third Committee Member

Raymond Fort

Abstract

Semiconducting metal oxide sensors are limited in their usage because of their poor detection selectivity. The current approach to achieve better selectivity in SMO detection uses prefiltering/preconcentration schemes to reduce the number of gases in contact with the sensor in combination with array-based detection. In this thesis we have investigated different materials and approaches for use as elements in an array based detection system.

One approach we have investigated involves the use of porous monoclinic WO3 to obtain size selectivity in detection within the sensing element itself. In chapter 3 we describe the synthetic protocol used to generate high surface area porous monoclinic tungsten oxide. Mesoporous oxides are produced by a sol-gel polymerization in the presence of a self-assembled surfactant structure. This approach has not been applied to the synthesis of WO3 based oxides because the presence of salts leads to mixtures of WO3 and tungstates. By minimizing the presence of Na+ ions, it is shown that ordered porous monoclinic WO3 can be prepared. The sodium tungstate is first passed through an ion exchange resin to remove the sodium and tungstic acid thus formed is then added to solution containing a cationic surfactant, n-cetyltrimethylammonium bromide (CTAB) to template the structure. While a salt is formed with the CTAB cation, it does not lead to stable tungstates because these salts are easily decomposed during the calcination step. It is also shown that the need for ion-exchange can be avoided by using ammonium tungstate as a precursor in place of sodium tungstate. As with CTAB cations, the NH4+ ions are easily decomposed during the calcination step. While the surfactant template collapses during the calcination step, the morphology and properties of the product is controlled by the initial template structure. Using these cationoic surfactant based receipies unique high surface area and porous monoclinic WO3 powders are prepared.

In Chapter 4, we examine the sensor properties of the various porous W03 powders. The sensors were tested to a series of alcohols of various size as well as dimethyl methyl phosphonate (DMMP, a nerve agent stimulant) and it was found that there was a size dependent response signal on the porous WO3 relative to sensors fabricated with nonporous WO3 powders. IR spectroscopic measurements shows that the difference in sensor responses on porous material was due to a size dependent control over the amount of alcohol absorbed on the surface. A key aspect of this approach is to operate the sensors in a difference mode in which a gas pulse is simultaneously exposed to several sensors composed of both porous and nonporous powders. By comparing the response on a porous sensor to that of a nonporous sensor it is possible to separately distinguish the signal of DMMP from methanol. The ability to distinguish the response of DMMP from methanol has been a longstanding goal to demonstrate selectivity in nerve agent detection.

In chapter 5 we examined a different approach to achieve selectivity in an array based SMO sensor. Specifically, the approach involves the use of UV illumination to selectively decompose adsorbed molecules from the surface of WO3. In infrared studies, it is found that adsorbed DMMP decomposes under UV illumination at room temperature to form a stable methyl phosphate species on the surface. However, the decomposition under UV does not occur with the lattice oxygen but rather with the ozone or surface 0' sites oxygen radicals produced in the gas phase and this is and this is unlikely to lead to a change in sensor response. In addition, it is found that the sensor base conductivity is also very sensitive to UV illumination at room temperature. The UV generates electron-hole pairs that decompose surface water and these results in the intercalation of H+ into the material to produce tungsten bronzes and a resulting change in base conductivity.

Download

Included in

Chemistry Commons

Share