This paper presents a novel non-contact method for evaluating the resonant frequency of a microstructure, Firstly, the microstructure under test is excited by ultrasonic waves. This excitation method does not impose any undefined load on the specimen like the electrostatic excitation and also this is the first actual use of ultrasonic wave for exciting a microstructure in the literature. Secondly, the amplitudes of the microstructure are determined by image edge detection using a Mexican hat wavelet transform on the vibrating images of the microstructure. The vibrating images are captured by a CCD camera when the microstructure is vibrated by ultrasonic waves at a series of discrete high frequencies （〉30 kHz）. Upon processing the vibrating images, the amplitudes at various excitation frequencies are obtained and an amplitude-frequency spectrum is obtained from which the resonant frequency is subsequently evaluated. A micro silicon structure consisting of a perforated plate （192 × 192 μm） and two cantilever beams （76 × 43 μm） which is about 4 μm thickness is tested. Since laser interferometry is not required, thermal effects on a test object can be avoided. Hence, the setup is relatively simple. Results show that the proposed method is a simple and effective approach for evaluating the dynamic characteristics of microstructures.
This paper studies the static deformation behavior of a piezoelectric micromachined ultrasonic transducer （PMUT） actuated by a strong external electric field. The transducer membrane consists of a piezoelectric layer, a passive layer and two electrode layers. The nonlinearities of the piezoelectric layer caused by electrostriction under a strong electric field are analyzed. Because the thickness of the transducer membrane is on the microscale, the size dependence of the deformation behavior is evaluated using the couple stress theory. The results show that the optimal ratio of the top electrode diameter and the membrane diameter is around 0.674. It is also found that this optimal value does not depend on any other parameters if the thicknesses of the two electrodes are negligible compared with those of the piezo- electric and passive layers. In addition, the nonlinearities of the piezoelectric layer will become stronger along with the increase of the electric field, which means that softening of the membrane stiffness occurs when a strong external electric field is applied. Meanwhile, the optimal thickness ratio for the passive layer and the piezoelectric layer is not equal to 1.0 which is usually adopted by previous researchers. Because there exists size dependence of membrane deforma-tion, the optimal value of this thickness ratio needs to be greater than 1.0 on the microscale.
