(98d) Silicon Based Microfabricated Sensor for Detection of Organophosphate Compounds

In the present work, we utilize a microscale gas-liquid interface for use in a selective gas micro-detector for detection of toxic organophosphates (OP) compounds. Previous work in our lab has demonstrated that OP compounds can be selectively detected using such a sensor fabricated in polycarbonate; however, the sensor material is not inert and cannot be integrated with other MEMS-based silicon devices. In this work we focus on the design of a MEMS-based silicon sensor using both experiments and COMSOL simulations. A Teflon nano-porous membrane is used to enhance the stability of gas-liquid interface as well as sensitivity of detection. A toxic gas of interest is injected into the vapor microchannel and reacts with an alkaline oxime solution as it dissolves in the liquid phase. In the reaction cyanide ions are produced and detected using a gold electrode on the nano-porous membrane. Response is measured as change in open circuit potential between the working and reference electrodes integrated on a single chip. Due to the toxicity of OP compounds, an OP simulant (e.g. acetic anhydride) which undergoes a similar reaction mechanism has been used. The detection limit of this sensor design is in the parts per trillion levels or approximately 3×10⁹ molecules.

In order to investigate the influences of important geometric parameters on detector performance, a finite element based commercial software, COMSOL 3.4 (Stockholm, Sweden) has been utilized. A 2D simulation of the system consists of gas and liquid micro-channels with a nanoporous gas-liquid interface. Coupled steady state Navier-Stokes, continuity and unsteady state mass transfer equations with chemical reaction have been numerically solved to establish a realistic model of the system. Simulation results indicate that using a nano-porous gas-liquid interface tremendously reduces diffusion time of cyanide ions, leading to a fast response of the detector compared to micron size membrane. Furthermore, it has been shown that the liquid channel depth and nano-membrane porosity are amongst the main parameters affecting the micro-sensor performance.

Experimental and simulation results demonstrate that the silicon based micro-detector proposed in this work can be a promising way to selectively detect ultra low levels of hazardous materials.