A submicrometre silicon-on-insulator resonator for ultrasound detection

The widely available silicon-on-insulator technology is used to develop a miniaturized ultrasound detector, which is 200 times smaller than the wavelengths of sound that it can detect. Ultrasound detectors use high-frequency sound waves to image objects and measure distances, but the resolution of these readings is limited by the physical dimensions of the detecting element. Point-like broadband ultrasound detection can greatly increase the resolution of ultrasonography and optoacoustic (photoacoustic) imaging(1,2), but current ultrasound detectors, such as those used for medical imaging, cannot be miniaturized sufficiently. Piezoelectric transducers lose sensitivity quadratically with size reduction(3), and optical microring resonators(4)and Fabry-Perot etalons(5)cannot adequately confine light to dimensions smaller than about 50 micrometres. Micromachining methods have been used to generate arrays of capacitive(6)and piezoelectric(7)transducers, but with bandwidths of only a few megahertz and dimensions exceeding 70 micrometres. Here we use the widely available silicon-on-insulator technology to develop a miniaturized ultrasound detector, with a sensing area of only 220 nanometres by 500 nanometres. The silicon-on-insulator-based optical resonator design provides per-area sensitivity that is 1,000 times higher than that of microring resonators and 100,000,000 times better than that of piezoelectric detectors. Our design also enables an ultrawide detection bandwidth, reaching 230 megahertz at -6 decibels. In addition to making the detectors suitable for manufacture in very dense arrays, we show that the submicrometre sensing area enables super-resolution detection and imaging performance. We demonstrate imaging of features 50 times smaller than the wavelength of ultrasound detected. Our detector enables ultra-miniaturization of ultrasound readings, enabling ultrasound imaging at a resolution comparable to that achieved with optical microscopy, and potentially enabling the development of very dense ultrasound arrays on a silicon chip.