Valerio Vignoli

Tin oxide gas sensing: comparison among different measurement techniques for gas mixture classification

In this paper, a study is presented aimed at the selection of the most appropriate measurement technique for wine classification. In particular, the problem of detecting typical wine aroma components in mixtures where ethanol is present is taken into account. The literature proposes different solutions in order to enhance metal-oxide sensor selectivity. An interesting approach concerns the application of different measurement techniques. In this work, three methods based on chemical transient, AC measurements, and temperature modulation have been investigated.

Surface State Model for Conductance Responses During Thermal-Modulation of SnO

Metal oxide gas sensors (MOXs) are widely used in olfactory electronic systems for their high sensitivity and low-cost. These sensors modify their conductivity in presence of oxidizing and reducing gases, and their performance is strictly dependent on the measurement technique adopted. In particular, it was already established by many works that a noticeable improvement in selectivity can be obtained by operating MOXs with a variable temperature. The temperature profile, however, must be tailored to the specific application, and the shape of the optimum profile for a given application depends both on the specific sensor and on the tested chemicals. In this context, there exists a strong interest in developing simplified models able to predict the sensor response, and aiming at a better comprehension of the mechanisms involved in sensing operations. In this paper, three simple gray-box models able to predict the behavior of some commercial thick film SnO2-based sensors in presence of oxygen and a reducing gas (CO) are proposed and discussed, whereas in a second paper the experimental validation of the model is presented

_{2}

In this paper, an innovative measurement system for odor classification, based on quartz crystal microbalances (QCMs), is presented. The application proposed in this paper is the detection of typical wine aroma compounds in mixtures containing ethanol. In QCM sensors, the sensitive layer is, e.g., a polymeric layer deposited on a quartz surface. Chemical mixtures are sorbed in the sensitive layer, inducing a change in the polymer mass and, therefore, in the quartz resonance frequency. In this paper, the frequency shift is measured by a dedicated, fully digital front-end hardware implementing a technique that allows reducing the measurement time while maintaining a high-frequency resolution . The developed system allows, therefore, measuring variations of the QCM resonance frequency shifts during chemical transients obtained with abrupt changes in odor concentration. Hence, the reaction kinetics can be exploited to enhance the sensor selectivity. In this paper, some measurements obtained with an array of four sensors with different polymeric sensitive layers are presented. An exponential fitting of the transient responses is used for feature extraction. Finally, to reduce data dimensionality, principal component analysis is used.

-Based Thick Film Sensors: Part I—Model Derivation

This paper and its companion (Part I) are devoted to the development and to the experimental verification of three simple gray-box models able to predict the behavior of some commercial thick film SnO2-based sensors in presence of oxygen and a reducing gas (CO). In this paper the models developed in Part I are applied to different commercial tin oxide sensors, and experimental results are discussed in order to gain a deeper insight into the sensor behavior

A measurement system for odor classification based on the dynamic response of QCM sensors

Metal oxide gas sensors (MOX) are widely used in olfactory systems for their high sensitivity and low-cost. These sensors vary their conductivity in presence of oxidizing and reducing gases and their performance is strictly dependant on the measurement technique adopted. In this paper the authors present results concerning the selectivity enhancement of an electronic nose based on MOX sensors, proposing also a gray-box model to describe the sensor response in presence of oxygen and CO during sensor temperature modulation.

Surface State Model for Conductance Responses During Thermal-Modulation of SnO

A new multisensor headspace and electronics for gas detection are presented. The system is PC-based and fully software controllable and reconfigurable. It was designed to allow for a very versatile use with different metal oxide sensor elements. In particular, the system supports different measuring techniques such as sensor impedance measurement in a frequency range spanning from DC to 15 MHz, and sensor gas sensitivity measurement when the temperature follows an arbitrary-defined trend. Some preliminary results concerning wine classification are also presented.

_{2}

In this paper a comparison among different parameter extraction techniques, applied to the dynamical responses of QCM (quartz crystal microbalance) sensors, is presented. Some results concerning the application of the studied technique to the wine aroma recognition are presented.

-Based Thick Film Sensors: Part II—Experimental Verification

In this paper an innovative measurement system for odor classification, based on Quartz Crystal Microbalances (QCHs), is presented. In particular, the problem of detecting typical wine aroma components in mixtures where ethanol is present is taken into account. In QCM sensors the sensitive layer is a polymeric layer deposited on a quartz surface. Chemical mixtures are adsorbed by the polymeric layer, inducing a change in the quantz mass and therefore in its resonance frequency. The frequency shift is measured by a dedicated fully digital front-end hardware implementing a technique proposed by Cantoni (2000). This approach allows reducing the measurement time while maintaining a high frequency resolution. The developed system allows measuring variations of the QCM resonance frequency shifts during chemical transients obtained with abrupt changes in odor concentration. Hence the reaction kinetics can be exploited to differentiate among different compounds. In this study some measurements obtained with an array of 4 sensors with different sensitive layers are presented. An exponential fitting of the transient responses is used for feature extraction. Principal Component Analysis (PCA) to reduce data dimensionality is used.

Electronic noses based on metal oxide gas sensors: the problem of selectivity enhancement

In this paper, the performance improvement of a gas-sensing system by digital correction techniques is discussed. The considered system operates as a vectorial impedance meter and performs impedance measurements of eight sensors arranged in an array in the frequency range 10 Hz-15 MHz. The measurements of the chemical sensors impedance is an innovative technique that allows highlighting different adsorption mechanisms taking place when the sensors are exposed to gases. Of course, impedance analyzers are commercially available, but they usually make measurements on only one device at time and they are very expensive. The proposed PC-based impedance analyzer is a versatile one and shows good performances for gas-sensing applications. A digital correction technique is used in this work to improve the impedance measurement accuracy of each channel of the gas-sensing system (eight sensors /spl rarr/ eight channels), in order to compensate for the conditioning electronics response. The latter is evaluated in a characterization procedure. A linear black box two-port model is used to take into account crosstalk, amplitude, and phase distortions. Two different techniques to evaluate the response of the measurement system are discussed in this paper, and experimental results are presented on both the measure of reference impedances and on the measure of chemical sensors.

Versatile headspace and electronics for measurements with gas sensor arrays

In this paper an ultrasonic multi sensor acquisition and processing system is presented, that can handle up to 32 air-ultrasound transducers in the frequency range of 40-200 kHz. The system was developed for an obstacle avoidance applications in the robotics field. Both the processing and the sampling capability of the apparatus (it is based on the Digital Signal Processor TMS320C25 and can sample up to five parallel channels) allow to perform tasks of different difficulty degree, laying between the target-ranging and the pattern recognition.