DETAILED DESCRIPTION

[0021] This invention is a mixed-potential sensor for the detection of non-methane hydrocarbons (NMHCs). The sensor utilizes a stabilized-zirconia electrolyte, and platinum (Pt) and perovskite-type oxide electrodes. The Pt electrode acts as a pseudo-reference electrode while the oxide electrode gives the mixed potential in the presence of reducing-gases. The selectivity of the device is achieved by the proper selection of the oxide electrode, while the stability of the device is achieved by the precise control of the surface area (SA) of the electrode and the 3-phase interface region (3PA) (gas-electrolyte-electrode) of the sensor. In accordance with one aspect of the present invention, the oxide electrode is La_1-xA_xCrO₃, where A is Sr, Ca, or Mg, 0≦x≦0.5.

[0022] By controlling the ratio of the SA to the 3PA, the rates of the heterogeneous catalysis and electrochemical catalysis are controlled for any particular electrode used. Thus, by proper selection of the electrode material and electrode dimensions, the magnitude of sensor response to any particular gas species can be amplified (selectivity). Moreover the design of the sensor assures that the SA and 3PA are stable over time and can be easily reproduced from sensor to sensor which ensures the stable and reproducible nature of the sensor response.

[0023] The specific approach involves the following steps:

[0024] 1) Sintering the metal-oxide electrode to form a dense ceramic.

[0025] 2) Cutting the ceramic to size (area of one of the sides=SA).

[0026] 3) Co-pressing the ceramic pellet along with the electrolyte powder.

[0027] 4) Polishing off the excess electrolyte powder in order to expose the oxide electrode.

[0028] 5) Sintering or partial-sintering of the resultant electrode/electrolyte structure in order to give mechanical strength and sufficient ionic-conductivity to the electrolyte.

[0029] 6) Applying a current collector electrode on the ceramic electrode surface.

[0030] The stated approach produces one of the electrodes. The other electrode can either be made in a similar manner with another oxide of different composition or with a metal wire.

[0031] As shown in FIGS. 1A and 1B, specific test cells were fabricated in accordance with the above procedure using La_0.8Sr_0.2CrO₃ as the oxide electrode 12, yttria-stabilized zirconia (YSZ) as the electrolyte 10, and Pt as the reference electrode 16. A lead wire 14 was attached to oxide electrode 12. A La_0.8Sr_0.2CrO₃ powder was obtained from Praxair® and was pressed into a pellet and sintered at 1550-1650° C. for 10 hours in order to form a dense (>60% of theoretical density) ceramic. The ceramic pellet was cut into smaller pellets with dimensions 0.2 cm×0.3 cm×0.2 cm. The pellet was buried in YSZ powder (having a surface area of 33.2 m²/gm) and pressed uniaxially at 1500 lbs in a ½″-¾″ diameter die for 5 minutes. The excess YSZ was removed from the surface of the oxide electrode pellet using a razor blade. Once a clear electrode/electrolyte interface was exposed, the pellet was sintered at 1000-1100° C. for 10-24 hours. The resulting pellet was about 60% of theoretical density. A 0.004″ diameter Pt wire was fixed onto the surface of the oxide electrode using a small drop of Pt paint. The other electrode was a Pt (0.01″ diameter) wire buried into the YSZ-electrolyte. The sensor was fixed to Pt (0.01″ diameter) leads and was heated to 400-600° C. in various gas atmospheres in order to test the sensor response.

[0032] The first sensor was annealed at 800° C. for 30 minutes in order to stabilize the Pt electrodes and was then cooled to 500° C., where the sensor was tested. The sensor response to propylene concentration in a 1% O₂ test gas (as used herein, all % numbers are volume %) at T=500° C. is shown in FIG. 2. The sensor response was very reproducible and was stable with time for at least 10 days (maximum deviation was <10%). The sensor response to 500 ppm propylene was about 40 mV at T=500° C. and had a response time of about 50 sec. In order to improve the response time, the operating temperature of the sensor was raised to 550° C. The sensor response to propylene concentration in a 1% 0₂ test gas at T=550° C. is shown in FIG. 3. The response to 500 ppm propylene was about 15 mV with a response time of about 25 sec. The sensor response at T=550° C. was also found to be stable for a period of at least 3 days. The sensor response time could be further improved by using a testing system with a lower volume (t<10 sec) and/or higher space velocities of the test gas.

[0033] In order to test the sensor-sensor reproducibility, another identical sensor was made and tested at 500° C. The sensor response of this sensor (sample 2) was in good agreement with that of the first sensor (sample 1) as shown in FIG. 4. For example, the variation in response between the 2 sensors for 500 ppm propylene was about 1%.

[0034] In order to test for interference from other gases, the sensor was exposed to varying concentration of CO (primary interference gas) in a 1% O₂ background. The response to propylene and CO in 1% O₂ at T=550° C. is compared in FIG. 5. This sensor not only shows a linear response to varying concentration of the analyzed gas but also has very good selectivity to propylene relative to CO (14 times more sensitive to propylene). The sensor response to other gases commonly present in the exhaust (propane, H₂, and methane) was also tested. However the oxygen concentration was not kept constant during these tests and was allowed to vary from 1% O₂ (for 0 ppm of test gas) to 0.5% 0₂ (for 1250 ppm of test gas). The results (FIG. 6) indicate that the sensor responds primarily to NMHCs (propylene and propane) and shows little interference from other gases.

[0035] Another design was also tested in order to improve the current collection from the chromate electrode. A Pt wire (0.01″ diameter) was buried in the chromate powder before it was pressed into a pellet and was sintered along with the pellet in order to make a Pt wire current collector that was both stable and robust. This sensor had a very similar response to those described above indicating that the various methods of current collection could be employed and that the sensor response is primarily controlled by the perovskite-type oxide. Moreover this sensor also had a response time (to 90% of level) of ≈10 sec.

[0036] There are two significant improvements of the present sensor over existing technology: the sensor response is stable over time and is reproducible from sensor to sensor, and the sensor response is highly selective to non-methane hydrocarbons.

[0037] The mixed-potential sensor can be used to sense the amount of NMHC emitted by any combustion, and, more particularly, in the tailpipe emissions of an automobile. This would help the auto-manufacturers test the efficiency of the catalytic converter. The current technology utilizes a two oxygen sensor approach where one oxygen sensor is placed in the exhaust upstream of the catalytic converter and the other downstream. The difference in the signal between the two sensors when the engine is cycled through stoichiometry is a measure of the oxygen capacity in the catalytic converter. The above method to determine the health of the converter is indirect (does not measure the actual conversion efficiency) and moreover will work only when the engine is cycled through stoichiometry.

[0038] The sensor according to the present invention can provide a direct measure of the catalyst efficiency because the sensor can be placed downstream of the converter and the sensor provides a measure of the actual amount of NMHC coming out of the tailpipe. Moreover the sensor works in high oxygen concentrations (≈1%) which makes it ideal for operation in the exhaust of a lean-burn engine.

[0039] As described above, one aspect of the present invention is a solid electrode embedded in a porous YSZ electrolyte. In another aspect of this invention, the 3-phase interface is increased by increasing the porosity of the electrolyte. It is well known that the commonly used electrode materials, e.g., precious metals, perovskite or transition-metal oxides and spinels, have much greater heterogeneous catalysis rates than the YSZ electrolyte. By increasing the porosity of the electrolyte material, the gases meander through the YSZ electrolyte instead of the electrode with a concomitant decrease in the heterogeneous catalysis that occurs before the gases reach the 3-phase interface. Hence, the response time of the sensors is improved by increasing the 3-phase interface without losing the sensor response due to heterogeneous catalysis on the porous electrode.

[0040] The effect of varying porosity of the electrolyte on the sensor response was illustrated by preparing several sensors with differing electrolyte density. This was achieved by sintering the YSZ electrolyte at various temperatures. The starting YSZ powder was obtained from Praxair as agglomerated powder at 10.5 m²/gm surface area. The YSZ electrolyte was sintered at 1050° C., 1100° C. and 1200° C., producing electrolyte materials that were approximately 81%, 86% and 90% of theoretical density, respectively. Two different sensors (given in Table 1) were studied in order to illustrate the effect of porosity (density) on the response time of these sensors. 1

[0041] Sensors A and B were initially sintered at 1050° C. for 10 hours and their response to propylene was evaluated under various conditions. Then, these two sensors were sintered again at 1200° C. and their propylene response was reevaluated. FIG. 7 shows the response of sensor A to 500 ppm of propylene at 600° C. in 1% O₂. It is seen that the sensor response to 500 ppm of propylene is increased from 80 mV to ≈110 mV when the sintering temperature is increased from 1050° C. to 1200° C. However, the response time increases from 3 sec to ≈140 sec as the sintering temperature is increased. This shows that the response time can be dramatically improved by sintering the YSZ electrolyte at a lower temperature. The better response time is due to a sensor body with higher porosity.

[0042] FIG. 8 shows similar response curves obtained for Sensor B, which was operated at a temperature of 650° C. The higher operating temperature compared to that of Sensor A results in a lower sensor response due to greater heterogeneous catalysis at this higher temperature. For example, the sensor response to 500 ppm of propylene is only 45 mV as compared to a response of 80 mV for Sensor A. However, the trends in the response times observed are identical and the response time is increased from 5 sec to ≈160 sec as the sintering temperature is increased from 1050° C. to 1200° C.

[0043] Accordingly, the sensor response time is decreased by increasing the porosity (or decreasing the density) of the electrolyte. The more porous the electrolyte, the faster is the gas access to the YSZ/electrode interface resulting in faster kinetics of the reduction and oxidation reactions. Moreover, an increase in the operating temperature and the flow rate of the gas also resulted in faster gas access and hence faster sensor response. By optimizing the various parameters, a sensor response of approximately 1 sec was obtained.

[0044] It will be appreciated from the above description that the resulting theoretical density and concomitant porosity is a function of the beginning YSZ powder morphology and sintering temperature. That is, the un-agglomerated powders (e.g. a powder having 33.2 m²/gm surface area as the first group of sensors) will maintain a relatively low density (higher porosity) at a given sintering temperature than the spherical agglomerated Praxair powders (having a 10.5 m²/gm surface area as in samples A and B).

[0045] To obtain a fast response time, it is desired that the YSZ electrolyte be sintered at a temperature effective to produce a density less than about 81% of theoretical density. The minimum acceptable density is a density that maintains the mechanical integrity of the sintered material. In the first set of sensors, the YSZ electrolyte was sintered at 1000° C.-1100° C. to produce a density of about 60% of theoretical density. Sensors A and B were sintered at 1050° C. to produce a density of 81% theoretical density; at 1100° C. to produce a density of 86% of theoretical density; and at 1200° C. to produce a density of 90% of theoretical density.

[0046] The foregoing description of the invention has been presented for purposes of illustration and description and is not intended to be exhaustive or to limit the invention to the precise form disclosed, and obviously many modifications and variations are possible in light of the above teaching. The embodiments were chosen and described in order to best explain the principles of the invention and its practical application to thereby enable others skilled in the art to best utilize the invention in various embodiments and with various modifications as are suited to the particular use contemplated. It is intended that the scope of the invention be defined by the claims appended hereto.