Development of an electronic nose to identify and quantify volatile hazardous compounds

Area of Science:

• Analytical Chemistry
• Sensor Technology
• Chemical Safety

Background:

• Hazardous chemical compound detection is crucial for safety protocols.
• Existing methods for identifying airborne chemicals can be complex and time-consuming.
• Developing rapid and accurate detection systems is essential for emergency response and risk assessment.

Purpose of the Study:

• To develop and evaluate a new electronic nose for identifying and quantifying hazardous chemical compounds.
• To assess the system's capability in a simulated hazardous release scenario.
• To determine the feasibility of using this technology for risk evaluation in store-rooms.

Main Methods:

• An electronic nose comprising an array of six coated piezoelectric quartz crystal sensors was designed.
• An artificial neural network (ANN) was employed for pattern recognition and compound identification.
• The system was tested with 10 different hazardous compounds (ammonia, propanone, hexane, acetic acid, toluene, methanol, tetrachloromethane, chloroform, ethanol, dichloromethane) released from a broken flask in a controlled environment.
• Both identification and quantification of the airborne compounds were performed.

Main Results:

• The electronic nose, utilizing an ANN, successfully identified and quantified airborne chemical vapors from a mixture of 10 hazardous compounds.
• Individual sensors lacked specificity, but the combined sensor array and ANN achieved reliable detection.
• The ANN could be simplified for identification-only tasks, with quantification achieved through individual sensor calibration.

Conclusions:

• The developed electronic nose is a viable tool for identifying and quantifying hazardous chemical releases.
• This technology offers a potential solution for rapid risk assessment in environments like store-rooms.
• Further optimization of the ANN and sensor calibration can enhance the system's performance for specific applications.