Multi metal oxide sensors system for some chemical functional group classification: The study through pattern classifier using correlation model in Electronic Nose Technology

Abstract

Four metal oxide sensors commercially available namely MQ2, MQ3, MQ6 and MQ8 were used as array sensors to specify methanol, ethanol, acetone and acetaldehyde. The voltage divider circuit was created to archive voltage signal with 10kW (??5%) reference resistor. Four-channel signal analyzer was carried out using ATMEGA328p microprocessor. An internal 10-bit ADC was programming with BASCOMAVR compiler to archive digital chart recorder. The signal was then digitalized and recorded using Microsoft Visual Basic??? via a serial interface at baud 19200 bps, 8 data bit, 1 stop bit and none parity with the sampling rate of 1 sample per second on four channels simultaneously couple with graphical plots on four different sensor elements. A reaction chamber with small DC diaphragm dispensing pump was designed to archive gas array sensor detection system. Four aliquots of each volatile (100%, 50%, 25%, 12.5% v/v) were diluted by distilled water and the head space was aspirated into the reaction chamber using 50mL disposable syringes. The injection volume to the chamber was optimized to 30 mL. The recorded signals were analyzed using correlation model. The purposed system was accurately specifying each volatile organic with the correlation factor of similarity > 0.995 with accuracy less than 5% RSD. The sampling rate of speciation each volatile was 15 samples per hour.

??