An integrated biosensor is a miniaturized device which performs biochemical analysis normally handled by a laboratory equipment. By incorporating signal processing electronics, it automates the massive processing of biological samples, avoids noise and interferences from wiring and interfacing with other systems and enables real-time detection of biochemical reactions. With these numerous benefits, integrated biosensors are poised to change the face of chemical and biological analysis and eventually revolutionize the pharmaceutical industry.
Among different types of biosensing techniques, impedance spectroscopy is peculiarly attractive for the purpose of integration. It is fully electronic in nature, since the sensing frontend can be simply an electrode pair or an inductor coil, which are native to the standard CMOS processes and can be readily adapted into various IC building blocks. Recent years have seen a tremendous advancement in integrated impedance biosensing, a handful of which have already pervaded people's daily lives. Nevertheless, the majority of the impedance biosensors focus on dc-MHz frequency range even though the transistor cut off frequency is approaching half terahertz.
To fulfill the frequency gap, this thesis aims to develop mmWave integrated impedance spectrometers for biomolecular sensing. Operating at mmWave frequencies offers additional advantages. For example, biomolecules interact with EM fields very differently at mmWave frequencies, which allows the discovery of new biochemical fingerprints. In the first part of the thesis, an oscillator-based dielectric sensor is developed which allows the characterization of biomolecules around 40 GHz with sub-ppm sensitivity. It is integrated with single-photon-avalanche-diode based optical sensors to allow highly selective sensing by creating a high-dimensional sample dataset. The second part of the thesis presents an integrated electron paramagnetic resonance spectrometer for in vivo applications. Several design challenges related to in vivo mmWave frequency sensing and detection are discussed and addressed.
