Dynamic FEM analysis of multiple cMUT cells in immersion

BAYRAM B., Yaralioglu G., Ergun A., Oralkan O., Khuri-Yakub B.

IEEE Ultrasonics Symposium, Montreal, Canada, 23 - 27 August 2004, pp.252-255 identifier identifier

  • Publication Type: Conference Paper / Full Text
  • Volume:
  • City: Montreal
  • Country: Canada
  • Page Numbers: pp.252-255
  • Middle East Technical University Affiliated: No


This paper reports on the accurate modeling of immersion capacitive micromachined ultrasonic transducers (cMUTs) using the time-domain, nonlinear finite element package, LS-DYNA, developed by Livermore Software Technology Corporation (LSTC). A capacitive micromachined ultrasonic transducer consists of many cMUT cells. In this paper, a square membrane was used as the unit cell to cover the transducer area by periodic replication on the surface. The silicon membrane, silicon oxide post and insulation layer were modeled, and the contact region was defined on the membrane and the substrate surfaces. The 3-D finite element model also included a 500 mu m-thick substrate and the acoustic fluid medium, to take into account two main sources of coupling in cMUTs: Scholte wave propagating at the solid-fluid interface and Lamb wave propagating in the substrate. A highly efficient perfectly matched layer (PAIL) absorbing boundary condition was designed for the acoustic medium to truncate the computational domain. The cMUT was biased in-collapse or out-of-collapse with an applied potential difference between the membrane and substrate electrodes: a rectangular pulse excitation was then used for the conventional, collapsed or collapse-snapback operations of the cMUT. Collapsed operation of the cMUT generated six times greater acoustic output pressure (641 kPa) than the conventional operation (107 kPa) at both the same bias voltage (83 V) and the pulse amplitude (+5 V). The vacuum backing and impedance-matched backing were compared to determine the influence of wave reflections from the bottom of the substrate in the collapsed operation. The dynamic FEN,I results were compared to the experimental results for conventional and collapse-snapback operations by applying step voltages on biased cMUT membranes. The acoustic output pressure measurements of the cMUT were performed with a hydrophone. The hydrophone calibration data was used to find the sensed pressure. Taking the attenuation and diffraction losses into account, the pressure on the cMUT surface was extracted. The cMUT generated 348 kPa and 1040 kPa in the conventional and collapse-snapback operations, respectively, and good agreement was observed with the dynamic FEM results.