Electronic Noses are devices able to generate digital maps of complex odors. Their working principles are aimed at reproducing the human olfactive system. The sensitivity of these artificial sensors, in fact, is comparable to that of human olfactive receptors; their data processing is similar to the sensing events occurring inside the olfactive bulbs and the resulting odor classification is carried out by Neural Networks or Statistic Multi-Component Analysis which reproduce both the learning and identification mechanisms of the cortical neurons. In the Microelectronics and Microsystems Institute Laboratories of the National Research Council (Rome, Lecce and Bologna Departments) several different prototypes of artificial olfactive systems have been set up since few years. These systems are based on chemical sensors, which work as miniaturized transducers, reacting selectively and reversibly to volatile chemicals and generating electrical signals as a function of substance concentrations.
The devices set up in the Rome Department, in collaboration with the Electronic Engineering and Chemistry Departments of University of "Tor Vergata", were based on piezoelectrical sensors, i.e. mass sensors. They are placed inside the "nostril" (measuring chamber) of the nose (LibraNose and EnQbe). The sorption of volatile molecules on the sensing layer (Me-Tetraphenylporphyrins differently derivatized and deposited) determines a frequency shift in crystal oscillations related to mass increase of the quartz (Thickness Shear Mode Resonators). In the devices assembled in Lecce and Bologna Departments, chemical interactions induce electron transfers between the semiconductor surface and gas molecules adsorbed thereon, which results in variations of their conductivity (Metal Oxide Semiconductor).
Specifically, these variations are consequent to the volatile compound oxidation occurring at semiconductor surface and to reduction of Oxygen molecules previously adsorbed on the sensor surface (SnO2, In2O3, TiO2...).
The features common to all the prototypes are: a prompt response to chemical interactions and the absence of time consuming sample pretreatment. Research in the three Departments were differently focused on minimizing the size of the equipment (by using microtechnologies) and improving both electronics, softwares and chemical features of sensors.
These purposes are achieved by setting up new materials able to increase the resolution (ppb) towards analytes, to decrease the response time, to obtain a complete reversibility, accuracy and reproducibility in measurements, and finally to get a high output signal with respect to noise. The setting up of such machines may have several powerful applications. In foodstuff manufacturing they may be used to assess the food real quality (e.g. evaluation of foodstuff preservation, control of marks guaranteeing the quality and origin of food products, fraud and food adulteration preventions ...), in environment monitoring (exhausted gas; gas, aromatic idrocarbons and aerosols leakages ...), in medicines (diagnostic tool for skin, endocrin system, gastrointestinal and lung apparatus diseases, etc.