UC San Diego

Sensors link the physical and electronic worlds, making them useful in environmental, automotive, industrial, communication, and medical applications, among many more. In the first chapter of this dissertation, current sensors and current-sensing front-ends are reviewed, aiming to provide readers with all-around design guidance from both sensor and circuit perspectives. Starting from the transduction method, capacitive, resistive, diode/FET-based, and MEMS sensors are individually reviewed with a focus on applications, circuit models, and nonidealities that must be considered for front-end design. This is followed by a discussion of current-sensing front-ends, specifically transimpedance amplifiers (TIAs), current conveyors (CC), and current-mode delta-sigma (I-ΔΣ) modulators. Each front-end is analyzed in terms of gain, bandwidth, stability, noise, and general design considerations are presented. In this chapter, state-of-the-art works for each front-end are also summarized, and tradeoffs between different architectures are discussed. The following chapters describe two application-specific current front-ends.In Chapter 2, a novel label- and immobilization-free biosensing technique, transient induced molecular electronic spectroscopy (TIMES), was introduced. An 8-channel array of low-noise (30.3 fA/√Hz) current sensing front-ends with on-chip microelectrode electrochemical sensors was proposed to observe real-time protein-ligand interactions. The analog front-end (AFE) consists of a 1st-order continuous-time delta-sigma (CT-ΔΣ) modulator that achieves 123 fA sensitivity over a 10 Hz bandwidth and 139 dB cross-scale dynamic range with a 2-bit programmable current reference. A digital predictor and tri-level pulse width modulated (PWM) current-steering DAC realize the equivalent performance of a multi-bit ΔΣ in an area- and power-efficient manner. The AFE consumes 50.3 µW and 0.11 mm2 per readout channel. In Chapter 3, an AFE for fast-scan cyclic voltammetry (FSCV) with analog background subtraction using a pseudo-differential sensing scheme to cancel the large non-faradaic current before seeing the front-end. As a result, the AFE can be compact and low-power compared to conventional FSCV AFEs with dedicated digital back-ends to digitize and subtract the background from subsequent recordings. The proposed AFE, fabricated in a 0.18-µm CMOS process, consists of a class-AB common-mode rejection circuit, a low-input-impedance current conveyor, and a 1st-order I-∆Σ modulator with an infinite impulse response quantizer. This AFE achieves an effective dynamic range of 83 dB with a state-of-the-art 39.2 pArms input-referred noise loaded with a 1 nF input capacitance (26.5 pArms open-circuit) across a 5 kHz bandwidth while consuming only 3.7 μW. This design was tested with carbon-fiber microelectrodes scanned at 300 V/s using flow-injection of dopamine, a key neurotransmitter.