Gas-Sensing Materials
- PDF / 4,368,148 Bytes
- 4 Pages / 604.8 x 806.4 pts Page_size
- 110 Downloads / 296 Views
strate upon which it is mounted to construct a sensor is equally critical, though less well recognized—even in this compendium. Thus some relevant comments are appropriate. So far, the most common Substrate material by far is alumina, a robust, electrically insulating ceramic, the properties and production techniques of which are well known. MOS gas-sensing materials operate at elevated temperatures, and alumina has also proved to be very convenient for the depositionor Containment of heater structures, 1 mostly metal but including deposited films of ruthenium dioxide, for example. However, alumina is also a poor conductor of heat, and for the larger Sub strates, this can lead to unwanted temp e r a t u r e g r a d i e n t s . Perhaps a better ceramic in this context is beryllia, which is a good insulator and a good heat con ductor, so that temperature gradients across it are minimal, irrespective of the heater geometry. 2,3 Perhaps inevitably, Silicon micromachined Substrates have also been investigated, 4,5 but with limited success at the time of this writing, although one device (the Motorola MGS 1100) did temporarily reach the market. The major problems encountered stem from the fact that the (comparatively) high operating tempera tures required by the sensor materials fall outside the acceptable temperature ränge of the Silicon, particularly when on-chip electronics is involved. Furthermore, although production runs are large, they do not compare with those common in the semiconductor industry and so may not merit the very high development costs typical in that industry. Nevertheless, it remains likeiy that such methods will eventually support volume produc tion of at least the cheaper domestic/ commercial sensors. For battery Operation, micropower de vices appear highly attractive, and here, heat loss becomes paramount. When the dimensions of "classical" sensors are very small, most of the heat loss occurs via conduction down the heater and inter-
rogation wires. For some experimental micromachined Silicon devices, the active material and the heater have been deposited onto a thin membrane (sometimes with a nitride layer) to minimize heat loss, which can lead to a lack of robustness and other problems. How ever, all very small sensors react (by changes in characteristics) to the rapid cooling that results from exposure to low-velocity drafts, and this necessitates more sophisticated feedback electronics for the heater drives. Such considerations clearly point to the need for multidisciplinary teams, including electronics, packaging, and materials specialists. In recent years, "electronic noses" have been featured prominently in the literature. 6,7 These involve arrays of partially selective sensors allied with sophisti cated signal processing electronics, in cluding neural nets. Although these have demonstrated great capabilities (e.g., in the odor-differentiating field), the general-purpose instruments available at the time of this writing tend to be expensive, with large arrays of sensors, both solid
Data Loading...
