Prototype microsensor arrays connected to an artificial neural network (ANN) running on a PC have been shown to reliably identify trace amounts of toxic gases in seconds, well before concentration levels become lethal.
Laboratory experiments at the US-based National Institute of Standards and Technology (NIST) showed that the sensor arrays can detect compounds such as sulphur-mustard gas and nerve agents (tabun and sarin) at levels below 1 part per million.
The sensors themselves use NIST-patented ‘microheaters’ that are coated with metal oxide films and then programmed to cycle through a range of temperatures.
Airborne chemicals attach to the films in characteristic ways depending on factors such as temperature and film material, causing changes in the flow of current through the sensors. These changes serve as a ‘signature’ for identifying both the type and concentration of the gas in the ambient air.
Signals from the sensors are then fed to ANN software running on a PC. The software processes signals from the sensor arrays, rapidly distinguishing between the gases and predicting their concentration in ambient air.
Further research is ongoing at NIST to more fully assess the impact of background interference as a means of avoiding false positives, and to enhance the robustness of the sensors after repeated use.