Semester

Spring

Date of Graduation

2022

Document Type

Dissertation

Degree Type

PhD

College

Statler College of Engineering and Mineral Resources

Department

Mechanical and Aerospace Engineering

Committee Chair

Xingbo Liu

Committee Member

Edward M. Sabolsky

Committee Member

Konstantinos A. Sierros

Committee Member

Wenyuan Li

Committee Member

Harry Finklea

Abstract

An advanced smart sensor network is essential to a combustion system, which is in favor of in situ, locally placed, and low-cost gas sensors. However, most chemical/electrochemical sensors fail to work in a combustion boiler, due to the demanding working temperatures (> 1000 ℃). This work for the first time reports a well-functioning mixed-potential type CO sensor at 1000 ℃ - 1200 ℃ using nickel oxide (NiO) as the sensing material. The influence of feed gas flow rate, electrode thickness, and porosity on sensor behavior is investigated, delivering notable and fast responses to 1000 ppm in 3% O2: 109 mV @1000 ℃, 40 mV @1100 ℃, and 7 mV @1200 ℃.

It’s worth noting that the sensing mechanism, different from the common theory based on CO oxidation coupled with oxygen reduction, is attributed to the reversed reactions: CO reduction coupled with oxygen revolution. For the first time, an inversion temperature is found for the CO-NiO reaction. Below the inversion temperature, CO oxidation occurs as commonly presented. Above the inversion temperature, CO is electrochemically reduced over NiO into carbon and oxygen ions, evidenced by electrochemical and compositional characterizations. The results complement CO sensing mechanisms for mixed potential sensors in high temperatures.

Furthermore, we endeavored to develop a combined CO, O2 and temperature sensor device, named as CO/O2/T sensor, to be applied in a utility boiler of Longview (a coal-fired power plant). Three times installations and testing were complete, while the sensor probe was installed through the observation port of the cooling wall into the boiler until the temperature reading was around 1000 ℃. For the first 20-h trial testing, a reasonable data was obtained: -45 mV (oxygen sensor) meaning 2.5%-3.5% PO2. 20 mV (CO sensor) corresponding to 2058 ppm CO. For the second trial testing that took 25-days, the aggressiveness of boiler condition was first perceived since the dusts and ashes fully covered the sensor head and dust coverage was thick, hard and stiff when the probe was took out. For the third installation, porous refractory bricks were placed on the top of the chamber to prevent the sensor materials from contacting the extreme poisoning ashes and dusts. However, a large fluctuation of the signal was observed. This may be due to the violent atmosphere change in the sensor chamber as the flue ash started to build up on the sensor shield. Ashes and dusts blocked the porous structure of the porous refractory bricks eventually, preventing the gas exchange between the boiler and the sensor chamber. Thus, a further improvement of sensor design is needed that can filter ashes and dusts as well as maintain the sample gas flow.

Download

Share

COinS