Ultrasonic transducers and sensors

Piezoelectric transducers

Contact: franck.levassort@univ-tours.fr / marc.lethiecq@univ-tours.fr / tran@univ-tours.fr

Ultrasound devices operating in the high-frequency range (over 20 MHz) are required for several medical diagnostic or NDT applications, including skin, eye, intravascular and small animals imaging. In its simplest configuration, a piezoelectric transducer consists of a piezoelectric material, one or several matching layers, a lens to focus the acoustic beam, and a backing that can be used as a mechanical support for the piezoelectric element. For high-frequency operation, the piezoelectric layer should be a few tens of micrometers thick, and hence, can be fabricated by thick-film technologies.

In this context, several collaborations have been established with laboratories and companies specialized in piezoelectric material developments and/or specific fabrication processes. PZT thick films were developed using electrophoresis process in collaboration with Jozef Stefan Institute. This process allows delivering integrated structures where piezoelectric layer is directly deposited on a backing (porous PZT, figure 1). Single element transducer was fabricated to deliver skin images (figure 2) (PhD thesis André-Pierre Abellard).


FIG 1: Photograph of curved gold-coated porous PZT substrate (collaboration Jozef Stefan Institute, Slovenia)


FIG 2: High resolution in vivo image of human forearm skin (collaboration with U930, Tours, France)

Today, lead-based compositions such as PZT are the most used piezoelectric materials but they are associated with health and environmental problems because of the lead they contain. As a result, since the mid-2000s, the EU and other countries around the world have included PZTs in their legislation as hazardous substances and to be replaced by other materials safe for health and the environment . More recently, since 2012, these compositions have been included in the European REACH directive. Research for substitute materials therefore becomes an emergency.

In collaboration with the CEA Le Ripault (near Tours), piezoelectric thick films with BHT composition have been developed by sol-gel composite and deposited by dip/coating on silicon wafer, always for high frequency medical imaging applications (PhD thesis Thomas Richadot). Finally, in the framework of an ANR project (HYPERCAMPUS) dedicated to the development of lead-free piezoelectric materials, in collaboration with VERMON and the U930 INSERM (Tours), we characterized a new lead-free piezoelectric crystal (figure 3) (Ph.D Thesis Rémi Rouffaud) for the fabrication of a 1-3 connectivity composite and its integration into a high frequency (30 MHz) linear array with 128 elements (figure 4). For the first time in the international scientific community, real-time images were obtained with this probe on the skin of a human forearm (figure 5). All this work confirms the interest of these new lead-free compositions for the manufacture of future generations of piezoelectric micro-devices.

FIG 3: Orientational dependence of the longitudinal velocity (m/s) vl and the two shear velocities vs1 and vs2 in the XZ-plane of a KN piezoelectric single crystal.


FIG 4: Photograph of a 30 MHz high-frequency probe (linear array, 128 elements, with VERMON)


FIG 5: Ultrasound image of a human forearm skin performed with 30 MHz lead-free probe (collaboration with U930, Tours).

cMUTs Capacitive Micromachined Ultrasonic Transducers

Contact: dominique.certon@univ-tours.fr / daniel.alquier@univ-tours.fr

Two original applications of cMUT applied to medical applications were recently investigated and funded by the French Research National Agency. The first, ANR COSTUM, was the development (in collaboration with LIB – UMR CNRS 1146 / UMR 7371) of a linear array (figure 6) dedicated to bone quality imaging. The probe is made of ultrasonic emitters and several dozens of ultrasonic receivers. The probe is placed in contact with the skin of the forearm and when emitters fire, they generate in the cortical layer of the bone a set of guided modes, so-called Lamb waves. The experimental position of these modes in the k-wave domain is then used for adjusting, by inverse problem, the elastic properties of the bone. This experimental technique is called the axial transmission measurement.

Successful results were obtained with the first prototypes. Measurements of guided modes in a bone mimicking plate obtained with the cMUT probe were compared with a piezoelectric probe (figure 7). Advantages of cMUT technology were clearly their ability to cover a large frequency range, i.e. from 500 kHz up to 3 MHz (figure 8). 

FIG 6: Axial transmission measurement : principle (a), probe topology (b) and medical apparatus developed by the LIB Laboratory (c)

FIG 7: Dispersion curves of the bone mimicking plate measured with the cMUT probe.

FIG 8: Results for a 2.2mm thick bone mimicking plate: theoretical (continuous and dashed lines) and experimental Lamb modes obtained with cMUT (circles) and PZT (dots) probes. For the cMUT probe, the circles correspond to the maxima observed on the Figure 7.

The second (THERANOS project) aimed to fabricate a dual-ultrasonic probe dedicated to imaging and therapy. Targeted applications are the fabrication of probes to help in the development of new drug delivery protocols. Here, thermal sensitive liposomes are used to carry drugs and then, to release it, under thermal elevation effect due to the application of moderate intensity acoustic field.

The fabricated prototype (figure 9.a) was designed for small animal imaging. It is composed of two kinds of arrays: four low frequency probes (@ 1 MHz) driven by high power CW emitters in order to produce the heating ultrasonic beam and one linear imaging arrays (@ 16 MHz). The four low frequency arrays are placed on a cylindrical shape frame in order to focus ultrasound at 20 mm from the probe. Moreover, each LF array contains 8 emitters that can be triggered individually to create electronic focusing. The mechanical and electronic focusing allowed obtaining peak-to-peak pressure amplitude, at the focusing point, of 3 MPa. Furthermore, elevation temperature tests were performed in glycerol and +7 °C was measured. The measurements were repeated several times in the same day, with 10 minutes rest between each experiment, without impact on the probe behavior neither on its ageing.

A second prototype is under investigation, where the LF arrays are replaced by four annular arrays in order to gain in focusing and so, to keep a flat shape dual-transducer (figure 9.b).

FIG 9: Photograph of two prototypes of dual-probe : curved transducer (a) and flat transducer (b)

Surface acoustic wave sensors

Contact: laurianne.blanc@univ-tours.fr/ julien.bustillo@univ-tours.fr / gael.gautier@univ-tours.fr                  Project linked with CARACUS theme

Surface acoustic wave sensors are based on the generation of a wave through interdigital transducers. Conventionally, transducers are deposited on a piezoelectric material (Quartz, LiNbO3, AlN...) and the wave propagation takes place in the piezoelectric material. This type of device can be used to characterize a thin film deposited between the IDTs (Figure 10). Changes (in magnitude and frequency) of the sensor resonance peak are related to the film properties (density, stiffness tensor, porosity, thickness).

The objective of this work is the characterization of silicon when it is made porous by electrochemical etching (porosity, thickness). For this, the propagation of the acoustic wave must take place in the silicon. Transduction is then made through IDT deposited on piezoelectric AlN (Figure 11).

Two approaches are studied:

  • The surface wave propagates only in the crystalline silicon. The porous zone is the sensitive layer of the device which will vary the acoustic charge of the surface wave during species detection.
  • The surface wave propagates integrally in the porous layer. In this case, the sensor sensitivity can be improved, but the frequency may be limited due to the high attenuation in the two-phase structure.

FIG 10: Schematic representation of a classical SAW sensor

               FIG 11: Schematic representation of the SAW sensor dedicated to silicon characterization