A new type of sensor, namely, Biologically sensitive Field Effect Transistors (BioFET) is potentially beneficial over established ones. BioFETs exhibit a biofunctionalized surface by replacing the gate structure of a conventional MOSFET. This surface allows the device to sense certain molecules in an aqueous solution. The analyte molecules specifically bind to the biofunctionalized surface layer and thus change the charge distribution and surface potential of the device, which alters the conductivity of the transducer. This little change in conductivity enables label-free detection within minutes. Due to the versatile surface modifications, BioFETs are gaining importance in the biomedical field.
Very promising sensors built from conventional silicon structures and silicon nanowires have been developed. BioFETs able to detect single-stranded DNA and tumor markers have been demonstrated. Despite these successes, the theoretical framework required to understand the experiments in a quantitative manner is still absent. The reason for this is the complexity of the system, which demands a consistent analysis of the semiconductor, the surface chemistry, and the transport mechanisms in the aqueous solution. As mentioned before, the device senses due to the charge distribution and potential on the biofunctionalized surface layer. Therefore it is important to understand the underlying physics and to take care of an adequate multi-physics model. For the semiconductor it is convenient to employ drift-diffusion-equation, while the liquid remains an object of discussion.
There are two commonly employed descriptions for the analyte. Firstly the Poisson-Boltzmann model, which treats the salt concentration as a continuous quantity and requires the salt ions to be in thermal equilibrium with their environment and secondly the Debye-Hückel model, which can be deduced by linearizing the Poisson-Boltzmann model. However, the often utilized Poisson-Boltzmann model fails at high potential values, especially for low salt concentrations. This is due to the low number of counter ions at small buffer concentrations and in conjunction with the overestimation of screening of the proteins/DNA by the Poisson-Boltzmann model. Therefore a self-consistent simulation of the whole device and a thorough treatment of the employed models is necessary to quantitatively understand and engineer these novel devices. However, the physical behavior of the system is far more challenging and requires further investigation.