Continuous Preconcentrator for Trace Gas Analysis
Yen-Lin Han and Marcus P. Young
Affiliation: Department of Aerospace and Mechanical Engineering, University of Southern California, Los Angeles, CA 90089, USA.
Keywords: Preconcentrator, gas sensors, MEMS, particle detection systems, &, sensor technologies
Patent records document a continuous development of trace gas/particle preconcentrators for use in particle detection systems. The most common application proposed for the inventions is the detection of biological agents, chemical agents, or explosives for security purposes although other applications are clearly possible. Existing particle detection systems such as gas chromatograph/mass spectrometer (GC/MS) systems do not have the required sensitivity for the characteristic application. Existing sensor technologies require pretreatment of the sample including precon-centration. There have been two classes of trace gas/particle preconcentrators: cycled and continuous. Preconcentrators that are based on adsorption membranes cannot operate on a continuous basis, because the gas must be stopped, a desired amount adsorbed, then released. These types of preconcentrators may therefore not be able to maintain the time fidelity of the analyte gas concentrations. Hence, preconcentrators that can operate continuously may be useful for detecting in substantial real time variations in the concentration of the trace gases that are being analyzed. A recent patent of a mesoscale, continuous flow-through, trace gas preconcentrator can enable this specific application with the capability of achieving significant trace-gas concentration increases using one or two simple meso-scale mass diffusion separation stages. The continuous trace-gas concentrator could increase the sensitivity for portable gas sensors/detectors by several orders of magnitude in a device - with a size less than several centimeters and with power consumption in the tens of milliwatts range. It can be operated continuously; instead of using adsorption-desorption cycles, and the response time is predicted to be more than one order of magnitude shorter than most currently available cycled techniques.
Rights & PermissionsPrintExport