CMUT Electrode Resistance Design: Modelling and Experimental Verification by a Row-Column Array

This paper addresses the importance of having control over the resistivity of the electrodes for capacitive micromachined ultrasonic transducers (CMUTs) devices. The electrode resistivity can vary depending on the fabrication technology used, and resistivity control becomes especially important in the cases where metal electrodes cannot be used. This raises the question: When is the resistivity of an electrode sufficiently low? To answer this question, we have developed a simple design criterion. The criterion describes the attenuation of ac signals along a CMUT element. It is shown that the nondimensional product between the angular excitation frequency, resistance, and capacitance ${\omega \textit {RC}}$ of an element has to be smaller than 0.35 to ensure an ac potential drop along the element of less than 1%. The optimal magnitude and directionality of the transmit pressure will be achieved if CMUT elements are designed according to the developed criteria. Hence, the model can be used to estimate device parameters that will ensure the CMUT is suitable for generating ultrasound images. An example is given where the model is used to predict the required electrode thickness for structured electrodes made of gold, aluminum, and indium–tin–oxide, respectively. To verify the model, two row-column addressed (RCA) CMUT transducers were used to illustrate the effect of high and low electrode resistivity. One transducer had a sufficient electrode resistivity, and the other had an insufficient electrode resistivity. The RCA CMUT transducers were fabricated using fusion bonding, where the top electrode is made of aluminum and the bottom electrode is made of doped silicon. The resistivity of the aluminum top electrode is ${2 \times 10^{-6}} \,\, {\Omega }$ cm for both transducers, whereas the resistivity for the bottom electrode is ${0.1} ~ {\Omega }$ cm for the first transducer and ${0.005} ~ {\Omega }$ cm for the second transducer. The transducer with low resistivity emits pressure uniformly along both the rows and columns, whereas the transmit pressure field from the other transducer has a uniformly distributed pressure field along the rows, but a decreasing pressure field along the columns due to the high resistivity in the bottom electrode. The pressure drop, along the columns, is frequency dependent and has been observed to be 63%, 74%, and 82% for the excitation frequencies 2, 4.5, and 7 MHz, respectively.