PhD defence by Yauheni Belahurau

Supervisors

• Principal supervisor: Professor Frieder Lucklum, Department of Electrical and Photonics Engineering, DTU
• Co-supervisor: Associate Professor Niels Aage, Department of Civil and Mechanical Engineering, DTU
• Co-supervisor: Associate Professor Rasmus Ellebæk Christiansen, Department of Civil and Mechanical Engineering

Evaluation Board

• Associate Professor Vicente Cutanda Henriquez, Department of Electrical and Photonics Engineering, DTU
• Professor Michael J. Vellekoop, University of Bremen, Germany
• Professor Yan Pennec, Université de Lille, France

Master of the Ceremony

• Associate Professor Finn T. Agerkvist, Department of Electrical and Photonics Engineering, DTU

Abstract

Measurement of volumetric properties of liquids in small volumes is a challenge due to the limitations of classical ultrasonic and resonant sensors, especially low sensitivity and only probing surface layers of an analyte.

A phononic-fluidic sensor consists of a fluidic cavity resonator with phononic crystal (PnC) layers around it. The phononic structures around the cavity significantly improve the boundary conditions of the cavity resonator, drastically increasing quality factor and resolution. Thereby, such combination allows to measure volumetric properties of liquids, for example, density and speed of sound.

This work consists of three main parts: numerical design, manufacturing and experimental characterization of the fabricated samples. In the first part, we established an advanced numerical model of the sensor, which takes into account its finite geometry, measured material parameters of the sensor material, vibration pattern of the transducers and fabrication tolerances. As a result, computed and measured transmission spectra are in a very good agreement. Moreover, we applied the methods of shape and topology optimization to create the design of a sensor with the aim to increase the Q-factor of acoustic resonance peaks. Additionally, we take into account the fabrication tolerances of a sensor element during optimization process. These optimization methods allows us to achieve new designs, which provide two times higher Q-factor of resonance peaks. Furthermore, we tested different formulations of objective functions and fabricated some optimized designs.

In the second part, we established the fabrication process of a sensor element, studied all possible tolerances, in order to use them in numerical design as constraints for optimization.

In the third part of this study, we have built an advanced experimental setup to conduct transmission measurements. Our setup is capable to control the alignment of the transducers and the sensor element with a precision of 10 μm.