Authors: M.J. Deen and M.W. Shinwari
Affilation: McMaster University, Canada
Pages: 599 - 602
Keywords: FETs, BioFETs, Modeling, Biosensors
Uniquely examining and identifying biological pathogens is of great importance in health care, as it facilitates early detection, isolation and possibly prevention of many diseases. Availability of highly efficient and easily produced sensors could have saved millions of dollars in prevention of the spread of many diseases, including the Foot and Mouth disease, Avian Bird Flu and Classical Swine Fever. Current nucleic acid detection techniques are based on optical detection of the hybridization of fluorescent-labeled DNA targets on surfaces containing probes of the complementary strand. This detection is usually done using laser scanners, which are very expensive. Recently, research has been directed towards direct detection of DNA molecules by means of their intrinsic negative charge. The aim of this research is to arrive at cost effective, highly sensitive, small-sized microarray biosensors that can be realized and integrated in mainstream CMOS technology. In particular, MOSFET devices can be used to sense DNA charges if they were placed close to the gate dielectric. Thus, by removing the gate of a conventional MOSFET, and exposing the dielectric, any DNA charges close to that region would induce a charge in the channel of the MOSFET. That charge could enhance the inversion of the channel and cause extra current conduction. Quantitative modeling that relate the amount of DNA charges to the inversion charge and, consequently, the sensed current change, has been carried out through complete electrostatic modeling of the gate-modified FET (BioFET). This modeling takes into effect the charge screening of the ionic medium and any possible chemical adsorption of ions onto the surface of the dielectric. Using this model, small signal parameters and the expected noise performance of the device has been investigated. The next step in this research would be to identify the mode of operation of the BioFET that would provide the maximum sensitivity to the DNA charges. Circuit amplification and noise reduction techniques can then be used to enhance the overall sensitivity and signal to noise ratio of this sensor. Eventually, several BioFETs can be placed in an array, creating an electronic microarray sensor that provides the DNA gene expression as electrical signal rather than optical. This signal can be sampled with a multiplexer and processed as a regular image. All the image processing techniques carried out for the optical microarray would be possible with this sensor. Optimization of the BioFET can be carried out in many different ways. Firstly, the relationship between the sensitivity and many different parameters has to be investigated. For example, the effect of changing the electrolytic strength, composition, pH concentration, dielectric type, DNA probe density, DNA strand length, or other parameters can be investigated. Two dimensional modeling of the potential distribution can give insight into the required DNA probe density that would result in predictable performance of the sensor. In this presentation, we will describe in detail our work on modeling the electrical characteristics of these BioFETs, including the effects of type of materials, design and operating conditions on the performance. Particular attention will be given to describing the signal and noise characteristics to demonstrate how highly sensitive detection systems can be designed and modeled to improve the detection limits of BioFETs-based sensing systems.
Nanotech Conference Proceedings are now published in the TechConnect Briefs