Solid State Electrochemical Sensors for Nitrogen Oxide (NOx) Detection in Lean Exhaust Gases

Solid state electrochemical sensors that measure nitrogen oxides (NO x) in lean exhaust have been investigated in order to help meet future on-board diagnostic (OBD) regulations for diesel vehicles. This impedancemetric detection technology consists of a planar, single cell sensor design with various sensing electrode materials and yttria-stabilized zirconia (YSZ) as the electrolyte. No reference to ambient air is required. An impedance analysis method yields a signal that is proportional to the analyte gas concentration at a specific frequency. These sensors function by detecting the change in impedance caused by electron exchange in the redox reactions of NOx gases at the sensing electrodes. From the impedance data, the resulting shift in phase angle is calculated, which can be calibrated to yield to the NO x concentration at low parts per million (ppm) levels.;Three varieties of impedance-based, lean NOx sensors have been fabricated manually, tested with both NO and NO2 gases at concentrations typical of diesel exhaust, and analyzed under various conditions. All sensors consisted of a planar, single cell design. Sensing electrodes were either gold wire or prefired, gelcast lanthanum strontium manganate (LSM, La0.85Sr0.15MnO3). The LSM sensors were mounted on dense substrates consisting either of alumina (Al2O3) or of partially stabilized zirconia (PSZ, ZrO2 with Y2O 3). Electrochemical impedance spectroscopy (EIS) techniques were used to interrogate the sensors. At low frequency (10 Hz), a signal was obtained proportional to low analyte gas concentration. The effects of temperature, total gas flow rate, and cross sensitivity to oxygen were examined for all sensors.;In addition, a strong temperature dependence was observed for the sensors with gold wire electrodes. The phase angles correlated linearly with temperature at 105 Hz. Generally, lowering the sensing temperatures resulted in larger phase angle responses, possibly due to the slower kinetics of the oxygen reduction reaction at lower temperatures. The lowest temperature evaluated for sensors with gold wire electrodes, 600°C, exhibited the largest change in phase angle. Nevertheless, even the lowest operating temperature examined was several hundred degrees above the temperature of the exhaust in the designated location of the sensor, requiring the sensors described herein to be continuously heated by a separate power source.;Equivalent circuit modeling was performed for the sensors in order to better understand the processes underlying the sensing mechanism. Excellent agreement with gold sensor data was obtained with a R0-(R 1C1)-(R2C2) circuit. The subcircuit elements are associated with the following physical processes: (0) contact resistance, (1) charge transport through electrolyte bulk, and (2) adsorption and dissociation of O2. NOx exposure evoked changes in the parameter values of R2 and C2 only. Both varied linearly over the entire range of NO (0-100 ppm). This finding suggests that these parameters can be calibrated to determine NO concentration. The rate limiting step was likely a process with atomic oxygen such as dissociation or surface diffusion.;Although the sensor results showed promise, the technology based on this material system faces several challenges prior to commercialization. Signal drift and poor manufacturability are interrelated problems. Signal drift results from microstructural changes (aging) in the electrolyte during exposure to high temperature gases. Elevating the sintering temperature to 1500°C as is standard practice in the manufacturing of oxygen sensors using high temperature cofired ceramic (HTCC) methods would mitigate aging by completing the microstructural phase transformation, however, this temperature would degrade the electrodes. Typically the electrodes and electrolyte are cofired in order to achieve good contact, but at 1500°C the gold electrodes would melt, and the LSM electrodes would form nonconductive zirconate phases. Microfabrication methods that physically deposit the electrolyte might address the aging issue, but this approach would require significant cost reduction analysis and implementation in order to be successful in the marketplace. (Abstract shortened by UMI.).