R. Lec

Macroscopic theory of surface acoustic wave gas microsensors

Recently it has been experimentally demonstrated (see J.F. Vetelino et al., 1987) that a surface acoustic wave gas microsensor can detect concentrations of certain gases in the parts per million to parts per billion range. Here, a theoretical model is presented which is capable of predicting the change in sensor response as a function of gas concentration. The geometry of interest consists of a metal oxide film on a piezoelectric substrate. It is assumed that on exposure to a gas, the most significant change in the film is the creation of free charge carriers which cause the film to change from a dielectric to a semiconductor. The free carriers then interact with the electric field associated with the acoustic wave traveling at the film-piezoelectric interface, causing a corresponding change in the sensor response. Results are presented for a tungsten trioxide film on a lithium niobate substrate which has been exposed to hydrogen sulfide gas. The theoretical results are compared to experimental data.<<ETX>>

A remote acoustic engine oil quality sensor

A prototype model of an Acoustic Engine Oil Quality (AEOQ) sensor is presented. The AEOQ sensor utilizes changes in oil viscosity as an indicator of the oil quality. The supporting electronic circuitry employs a novel VCO based measurement set-up. The output signal from the VCO set-up is a DC voltage, which simplifies the accompanying circuitry, and is compatible with existing automotive electronics. Experimental results showed that the AEOQ sensor is capable of differentiating between engine oils of different grades and monitoring the degradation processes of oils due to their use in automotive engines and due to artificial contamination by dilution with water, ethylene glycol, and gasoline.

An acoustic plate mode immunosensor

The application of an acoustic plate mode (APM) device as an immunosensor is presented. The APM sensor utilizes a Z-cut X-propagating lithium niobate (LiNbO/sub 3/) piezoelectric plate in which various types of acoustic waves are excited and received by means of a photolithographically deposited aluminium interdigital transducer (IDT). A judicious choice of the sensor operating frequency results in the selective excitation of an acoustic mode that is highly sensitive to fluid loading of the LiNbO/sub 3/ surface. This mode is then utilized as the sensing element in the sensor. Experiments have been performed using polyclonal antibodies for which the sensor has shown a strong response to the associated antigen. The biokinetics of the antigen-antibody reaction have also been studied. Experimental data compare favorably with theoretical results predicted by an affinity-purified human immunoglobulin antibody-antigen model. This immunosensor also has potential application in detecting various types of viruses such as AIDS (HIV), herpes simplex, and hepatitis.<<ETX>>

Ultrasonic sensor for the characterization of colloidal slurries

A bulk wave acoustic sensor capable of measuring PSD (particle size distribution) for both low- and high-concentration slurries has been designed, fabricated, and tested. The principle of operation of the sensor is based on the measurement of the variation of the attenuation of the acoustic wave in the slurry as a function of frequency. The attenuation data are then incorporated into a theoretical model to predict the PSD of TiO/sub 2/ particles in the slurry. The sensor is capable of detecting the PSD of the TiO/sub 2/ particles with sizes ranging from 0.1 to 10 mu m in high-concentration slurries, up to 75 weight percent. A potential application of the sensor is in process control in the manufacture of pigments or slurries for use in paints and paper coatings.<<ETX>>

Optimization of a SAW metal oxide semiconductor gas sensor

A simple theoretical model is applied to the problem of a SAW (surface acoustic wave) gas sensor which operates on the principle of conductivity changes within the sensing film. The model allows the prediction of SAW sensor response for a film of known conductivity variations, and the theoretical value of SAW velocity change agrees very well with that determined by experiment. The electroacoustic attenuation was also determined to be a significant factor in designing a SAW sensor. SAW sensor response can be optimized by designing the device such that the sheet conductivity of the film is within some region dependent upon the substrate material. A WO/sub 3/ film could be designed to have a sheet conductivity within a specific range by tailoring parameters such as film thickness, doping, and heat treatment. A sensor consisting of an ST-cut SiO/sub 2/ substrate and a WO/sub 3/ film selectively sorbent to H/sub 2/S was investigated. The SAW gas sensor is configured so that the film sheet conductivity and sensor response are measured simultaneously upon exposure to H/sub 2/S. Specific operating conditions are obtained that optimized the SAW gas sensor response.<<ETX>>

Influence of an External Electric Field on the Electroacoustic Properties of PZT-4

The influence of a static external electric field on the electroacoustic properties of the piezoelectric ceramic, PZT-4, has been investigated. The velocity of the acoustic waves were found to depend strongly on the applid. electric field. Both linear and nonlinear velocity variations as a function of the applied electric field were observed. In particular, under the influence of an electric field along the 0011 axis 1001 axis, the changes of the veloci are linear. direction the changes. in the acoustic wave velocity in the [loo] direction are nonlinear. PZT-4 displays a very strong electroacoustic effect, signifgicantly greater than that known for LiNbO , one of the strongest known nonlinear piezoelecTric materials. and a longitudinal wave propagating a I ong the t owever, for the electric field along t 3: e [OlO]

Scattering- and mixed-matrix analysis of BAW devices

Bulk acoustic wave (BAW) transducers have been employed for fluid and solid property measurements as well as for signal processing applications for many years. BAW measurement and signal processing techniques are robust and miniature, making them ideal for industrial and military applications. Several structures have been considered, including resonators, delay lines and hydrophones. All of these BAW transducer geometries require an accurate and efficient analytical model of the piezoelectric plate and associated bonding, backing, electrode and sample media. Previously, lumped element models, impedance matrix and hybrid matrix analysis have been utilized. The current work details a scattering matrix analysis of multilayered BAW transducers which is rigorously derived from the wave equation and the coupled mode differential equations. It agrees well with the previous methods while allowing additional insight into the problem

Application of a radial basis function neural network to sensor design

One of the important tasks in sensor design is the development of a model for a sensing phenomena. Artificial neural networks are ideal for such a task because of their capability for representation of the mapping functions describing the processes and phenomena which are mathematically difficult or even intractable. We examined a radial basis function (RBF) neural network for modeling of acoustical properties of colloidal TiO2 slurry. The colloidal slurry is a very complex multiphase medium. The RBF network with a set of local Gaussian functions was trained using the data from the earlier developed physical model of TiO2 slurry. Next the TiO2 neural model was used for a prediction of the TiO2 particle size distribution. The resulting prediction accuracies of the RBF network were 99.8% for the data used in the training process and 88% for the data not used in the training. Compared to other available techniques neural networks can offer an effective and time efficient approach for the modeling of complex materials.

Neural network approach to sensor design

A radial basis function (RBF) neural network is examined for modeling of acoustical properties of colloidal TiO/sub 2/ slurry. The colloidal slurry is a very complex multiphase medium. The RBF network with a set of local Gaussian functions is trained using the data from a previously developed physical model of TiO/sub 2/ slurry. The TiO/sub 2/ neural model is used for a prediction of the TiO/sub 2/ particle size distribution. The resulting prediction accuracies of the RBF network is 99.8% of the data used in the training process and 88% for the data not used in the training. Compared to other available techniques neural networks can offer an effective and time efficient approach for the modeling of complex materials.<<ETX>>