High-Performance Graphene FET Integrated Front-End Amplifier Using Pseudo-resistor Technique for Neuro-prosthetic Diagnosis

• Published in: Biochip journal 2022, 16(3), , pp.270-279
• Main Authors: Naik, Jatoth Deepak, Gorre, Pradeep, Akuri, Naga Ganesh, Kumar, Sandeep, Al-Shidaifat, Ala’aDdin, Song, Hanjung
• Format: Article
• Language: English
• Subjects: Action potential; Amplifier design; Amplifiers; Art works; Bandwidths; Biocompatibility; Biomedical Engineering and Bioengineering; Biotechnology; Brain research; Chemistry; Chemistry and Materials Science; CMOS; Diagnosis; Electrodes; Field effect transistors; Firing pattern; Gallium arsenide; Graphene; Input impedance; Leakage current; Low voltage; Monitoring systems; Neurons; Neurosciences; Optimization techniques; Original Article; Power consumption; Power management; Prostheses; Resistors; Semiconductor devices; Semiconductors; Silicon wafers; Transconductance; Transistors; Voltage; 생물공학
• ISSN: 1976-0280; 2092-7843
• DOI: 10.1007/s13206-022-00060-5

A complex analysis of spike monitoring in neuro-prosthetic diagnosis demands a high-speed sub-nanoscale transistors with an advanced device technologies. This work reports the high performance of Graphene field-effect transistor (GFET) based front-end amplifier (FEA) design for the neuro-prosthetic application. The 9 nm Graphene FET device is optimized by characterization of transconductance and drain current towards high sensitivity and small factor. The proposed GFET-based FEA with pseudo-resistor technique demonstrates very high-input impedance in Tera-ohms that nullify the input leakage current. Here, gain-bandwidth product and noise optimization of GFET FEA enhances the overall gain with negligible noise. The proposed design operates at low voltage, further reduces the power consumption, and achieves less chip area in sub-nano size so it could be more suitable for implantable devices. The GFET-based FEA architecture achieves an action potential spike of 1.4 µV while the local field potentials spike of 1.8 mV. The proposed architecture is implemented in Advanced Design System using the design kit of the GFET process. Power consumption of 3.14 µW is observed with a supply voltage of 0.9 V. The simulated and experimental results of the proposed design achieve an input impedance of 2 TΩ with excellent noise performance over a wideband of 13.85 MHz. The proposed work demonstrates better neural activity sensing when compared to the state-of-the-artwork, which could be highly beneficial for future neuroscientists.