PSPICE model for silicon nanowire field‐effect transistor biosensors in impedimetric measurement mode

Abstract

In this work, we demonstrate a behavioral PSPICE model for silicon nanowire (SiNW) field-effect transistor (FET) biosensors, which is suitable to simulate frequency domain electrical measurements. The model is divided into two separated components: an electrochemical part for the liquid/solid interface at the gate input and an FET part simulating the SiNW characteristics. In our study, the parameters of the FET model are obtained from characterization measurements of real devices, which are fabricated in our research group. Measurements were performed with a self-developed readout system. The model can be used to investigate the effects of solution conductivity and the drain and source capacitances of the SiNW sensor. We observed that the impedance spectra consist of a low-pass filter domain at low frequencies and a resonance domain at higher frequencies. The former is mainly influenced by the source capacitance and the oxide/liquid interface of the devices, whereas the latter is mainly dominated by the drain capacitance, in case that the solution conductivity is kept constant. In addition, based on the relationship of the solution conductivity and the experimental data, we suggest a value called representative resistance of the solution. In this study, the model was compared to real spectra and we achieved good agreement between simulation and experiments. Our model provides a clearer view for frequency domain measurements with SiNW sensors, offers the possibility to optimize our SiNW design and forms the basis to include and explain sensing events in future biomedical assays.

DNA detection with top–down fabricated silicon nanowire transistor arrays in linear operation regime

“…For comparison, also a fit with the former level 2 model is shown. It can be seen that the device can be regarded as a long‐channel transistor and that this model is representing the device characteristics more faithfully compared to the previously used level 2 model . The PSPICE circuit for the SiNW‐FETs in our model is including physical parameters of the fabrication processes such as gate oxide thickness, junction depth, and doping concentration.…”

“…Our SPICE model to simulate bio‐sensing experiments with ISFET devices is composed of two components: (i) a membrane model describing the binding event of receptors immobilized on FET devices and biomolecules of interest in the measuring solutions; (ii) a MOSFET model representing the electrical properties of transistors. The MOSFET level 2 was used in some earlier publications to simulate either micro‐sized ISFETs or SiNW‐FETs . However, this model was not efficient in this study to faithfully simulate the DNA experiments with SiNW‐FETs with widths of 100 nm and length of 40 and 20 μm, respectively, in DC measurement mode.…”

“…We already treated our silicon nanowire devices as ideal long‐channel field‐effect transistors. To describe the device characteristics, we used a level 2 model in PSPICE, which did not include advanced CMOS effects . Therefore, this model did not faithfully represent the characteristic lines of our devices.…”

Handheld readout system for field‐effect transistor biosensor arrays for label‐free detection of biomolecules

“…Commonly, there are two read‐out methods used in the field: (1) Comparison of the shift of V TH before and after the binding events or, (2) measuring the real‐time difference of the drain–source current ( I DS ) during the reaction. Usually, such devices are recorded by high accuracy semiconductor parameter analyzers or home‐made desktop systems . These are both big and expensive or lack high accuracy and portability.…”

“…The working mechanism of the FET-based biosensors is that the variation of the surface potential induced by the binding of the charged molecules to the modified surface leads to a change in the threshold voltage (V TH ) of the device. Commonly, there are two read-out methods used in the field: (1) Comparison of the shift of V TH before and after the binding events or, (2) measuring the real-time difference [6,7,11]. These are both big and expensive or lack high accuracy and portability.…”