An optimized piezoresistive microcantilever arsenic(III) sensor with CMOS-compatible active readout: Towards in-situ subsurface characterization using microsystems technology

Abstract

Arsenic (As) currently ranks as the number one substance in the most recent (ATDSR, 2007a) Comprehensive, Environmental, Response, Compensation and Liability Act (CERCLA) Priority List of Hazardous Substances published by the Agency for Toxic Substances and Disease Registry (ATSDR). The chemistry of this metalloid, emerging from various natural and anthropogenic sources, poses serious toxicological concern. Existing procedures for subsurface characterization rely primarily on direct sampling & off-site analysis techniques viz. liquid scintillation counting and ICP-MS. Not only do they impose severe limitations on readout-integration & packaging, but also lack precision. In this context, low-cost real-time in-situ diagnostic measurements for arsenic contamination are vital. This paper proposes a novel MEMS based low-power piezoresistive microcantilever sensor system, that can be effectively implanted in an on-site monitoring cavity to detect contamination. The design is based on a single crystal silicon substrate exploiting the benefits of anisotropic chemical etching. The transduction is based on the binding of the contaminant to the Al <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> O <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</sub> self-assembled-monolayer (SAM), and the adsorbtion-induced stress change is monitored via the bending of the beam and thus piezoresistivity. The device uses a readout interface consisting of a common-gate MOS transistor configuration unlike the conventionally used Wheatstone bridge, which allows significant reduction in power consumption as compared to its counterparts. The results strongly indicate the realizability of low-cost in-situ arsenic detection and monitoring. The working of the sensor is simulated using the Finite Element Analysis (FEA) module in COMSOL Multiphysics.