Date of Award

Winter 2007

Project Type

Thesis

Program or Major

Mechanical Engineering

Degree Name

Master of Science

First Advisor

Todd Gross

Abstract

Solid-state gas sensors consist of a solid material, usually a metal oxide semiconductor, whose conductivity changes when gases adsorb onto its surface. The goal of this project is to create an ozone detection system using these sensors for use in autonomous environmental monitoring stations (such as weather balloons). To meet EPA specifications (less than 5% error), the sensor will have to have 1ppb resolution and be able to detect ozone in the range of 0-100ppb.

A sensor was purchased from MicroChemical Sensors (MiCS) for adaptation to this application but the stated range of application (10-1000ppb) was 10 times the desired resolution. The bulk of the effort on this thesis focused on calibrating the sensor at its lower detection limit. The sensor was tested for its response to ozone, as well as for crosstalk with H2O, CO, NOx, SO2, and CO2. The sensor generated a measurable response to ozone levels in the proposed measurement range, but had an equally strong response to humidity (it was insensitive to the other gases). The sensor response vs. humidity and ozone concentration were measured over a range of 0ppb to 150ppb ozone and 2.5% to 85% relative humidity. The results were fitted to a modified multiple-gas Langmuir adsorption equation using an error minimization algorithm available in MATLAB. The data fit the modified Langmuir equation with an R2 value of 0.9958.

The combined humidity and ozone equation was rearranged to obtain an equation for in-situ calculation of the ozone concentration. This was input into a LabVIEW program and tested against an ozone analyzer. The calibration held to within +/-2.8ppb, or +/-5% (with the exception of one extraneous data point), meeting EPA specifications.

Download

Share

COinS