Chemical Microsensors for Satellite Applications
- PDF / 1,201,096 Bytes
- 6 Pages / 414.72 x 648 pts Page_size
- 86 Downloads / 195 Views
hydride or reduction of an oxide in other circuit components such as resistors. 8 To confirm that H 2 is indeed responsible, we have developed and tested a Pd chemiresistor for in-situ detection of H 2 in these device packages. Furthermore, such sensors could be incorporated into the packages to serve as simple passive monitors for H2 levels to verify their health prior to launch. EXPERIMENTAL Figure 1 shows a photograph of the sensor developed. It consists of thin films of Pd/Ni alloy deposited using RF planar magnetron sputtering from separate Pd and Ni targets. The deposition rate was 10 Amin. and the film thickness was 500 A as determined by quartz crystal monitors and confirmed by stylus profilometry. Film composition was measured by energy dispersive x-ray spectroscopy (EDX) of witness samples and determined to be 10.1± 0.5 % Ni. The pattern of stripes for resistance and dots for capacitance measurements was obtained by photolithography and liftoff. Gold contact pads (2000 A thick) for wire bonding were sputtered onto the Pd/Ni using a separate mask. The substrate was a 2" diameter Si(100) wafer with 250 A of thermal oxide. Testing of sensor response was performed using two separate apparatus, a flow cell for testing atmospheric pressure gas mixtures (obtained using flow meters and dynamic mixing) and a static cell for exposure to sub-atmospheric pressures of pure gases. Both cells are standard NW-40 crosses and both apparatus use a multimeter (HP 3478A) interfaced to a Macintosh computer with LabView software for data acquisition and automated testing. Two resistors on separate devices were tested, resistor A is 0.15 x 4 mm and resistor B is 0. 15 x 2 mm.
Figure 1.Photograph of the Pd chemiresistor developed for hydrogen detection. RESULTS AND DISCUSSION Figure 2 shows the response of resistor A to pressures of pure H 2 over the range from 1.1 to 82.9 torr. The data was taken in the static cell and similar data have been obtained at pressures as high as 1000 torr. When the percent change in resistance as a function of H2 pressure is plotted, all of the points lie on a smooth curve showing that there is no hysteresis in the sensor response to varying levels of H 2 . As shown in the figure, the response times to these concentrations of H2 are quite fast, less than one minute. The temperature is ambient and at higher temperatures it should be possible to achieve faster response times. Figure 3 shows the response of resistor B to varying levels of H2 in the flow cell. Six hour time periods were used and the sequence of gas mixtures consisted of the following: 0, 1.00 x 103, 0, 1.00 x 104, 0, and 3.00 x 104 ppm H2 in N2 respectively. Note that the response to an increase
536
112 111-
192
S 188
110-
184 184-S••-•
U 109•
180_/ 1760
. I 5
V 18 . . I I 10 15 Time (min)
108107-
I
20
Figure 2. Response of the sensor to various pressures of pure hydrogen in the static cell. The pressures are as follows: 0.0, 1.1, 17.7, 38.9, 82.9, 13.8, 1.9, 0.0, 8.5, 23.5, 70.6 and 51.6 torr respectively,
I
0
I
5
Data Loading...
