Submitted to the Department of Electrical Engineering.
Thesis (Ph.D.)--Stanford University, 2013.
Sensor systems are a ubiquitous part of modern life and make huge impacts on how we deal with disease and injury, how we communicate and travel, and how we understand our environment and ourselves. This thesis presents research in two emerging areas, with contributions in front-end mixed signal interface electronics as well as measurement systems. The first project explores interface electronics for in vivo recording of neural signals in human and non-human subjects. Research tools created by IC designers are used to explore the function of the central and peripheral nervous systems, and make an impact in the way we diagnose, treat, and understand a broad range of neurological ailments such as epilepsy, chronic pain, obsessive compulsive disorder, and chronic neurodegenerative diseases. A new front-end circuit architecture based on switched capacitor filtering and windowed integrator sampling is presented. The circuit was integrated into a 96 channel neural recording ASIC, which acquires extracellular neural action potentials and local field potentials from a Utah Electrode Array (UEA) implanted in cortex. The front-end achieves 2.2uVrms input referred noise in a 10kHz bandwidth, and conditions signals before conversion at 31.25kSa/s by 10-bit SAR ADCs with 60.3dB SNDR and a Walden figure of merit of 42fJ per conversion step. This power and area efficient sensor interface consumes 6.4mW from 1.2V while occupying 5mm x 5mm in 0.13um CMOS, and is the cornerstone of the HermesE wireless neural recording platform, enabling basic neuroscience as well as neural prosthetics research. The second work described in this thesis relates to label-free biothreat detection based on transduction of vibrational mode information that couples to electron transport through electron-phonon interactions. This project proposes an ambitious, novel biosensing paradigm that is alternative to traditional micro-array systems, the current workhorse for resolving molecular biology, with potential advantages in specificity, sensitivity, portability, cost, throughput, and manufacturability. The sensor is based on a gold electrode coated with a self-assembled monolayer film, and is placed in an electrolyte solution prepared from a biological sample. The low magnitude of the vibrationally-assisted tunneling current and the need to reduce electronic interactions in the sensor have motivated a number of extensions to the conventional electrochemical instrumentation space, including noise suppressing potentiostat designs, a three-electrode AC measurement system based on a replica path scheme to reject blockers and errors, and a three-electrode DC current measurement system achieving sub-picoampere readout accuracy. These systems were implemented with a mixture of discrete integrated circuits, laboratory instrumentation, and custom software.