BESOI MOSFET transistors optimization for biosensor platform application.

Author(s)

Leonardo Shimizu Yojo

• Document type: Doctoral Thesis
• Press: São Paulo.
• Institution: Universidade de São Paulo (USP). Escola Politécnica (EP/BC)
• Defense date: 2022-03-08

Examining board members

João Antonio Martino; Salvador Pinillos Gimenez; Sebastião Gomes dos Santos Filho; Ana Neilde Rodrigues da Silva

• Advisor: João Antonio Martino

Abstract

Biosensors are defined as the devices that are capable of transforming chemical reactions from biological interactions into electrical signals. They have many applications, such as the medical area and food industry, and they provide a better life quality. Among the numerous types of biosensors, the ones based on the field effect transistors are particularly interesting due to the benefits brought by micro and nanoelectronics evolution over the last decades. The BESOI MOSFET (Back-Enhanced Silicon-On-Insulator Metal-Oxide-Semiconductor Field-Effect-Transistor) was developed and fabricated in Laboratorio de Sistemas Integraveis (LSI) at Escola Politecnica da Universidade de São Paulo (EPUSP). This planar transistor built on a silicon on insulator technology stands for the simple fabrication process and for the operation flexibility, i.e., a single device can act as a n-channel transistor or a p-channel transistor, depending on the bias at the back gate (programming gate). This work proposes the use of the BESOI MOSFET as a biosensor, taking advantage of its structure, in particular the channel underlap regions (between the front gate electrode and the drain/source contacts), to create a device sensitive to different biological elements. In the first part, numerical simulations based on the fabricated transistors measurements will help with the understanding of its electrical behavior when a biological sample is inserted. The permittivity constant (k) and the charges (Qbio) of the biological material were analyzed. The simulations showed that the drain current varied as a function of the analyzed parameters. The evaluation of the physical parameters also showed that the gate underlap length is the main parameter that affects the sensitivity of the devices. The optimization of the dimensions resulted in a sensitivity (according to de definitions adopted in this work) increase in the permittivity-based sensors from 1.43 to 18.43 units for the n-type bias, and from 1.62 to 13.82 units for the p-type bias., in the best case. The charge-based sensors optimization provided a two times increase in the sensitivity for the n-type bias, from 0.79 to 1.78 units; and from 12.69 to 1812 units for the p-type bias, in the best case. From the proposed methods and adopted parameters, in the second part of the work, a BESOI MOSFET sensor was fabricated, followed by its electrical characterization and the analysis of the results. The proof of concept was performed by means of a glucose charge-based biosensor. Among the glucose-oxidase enzyme immobilization methods that were used, the crosslinking process resulted in a higher sensitivity (1.8 in the best case) in comparison with the enzyme covalent bonding with the device surface method (0.52 in the best case). However, the second method presented a more uniform deposition. Both methods resulted in a drain current increase as a function of the gate bias, that was proportional to the glucose concentration of the solutions on the sensors. (AU)

• FAPESP's process: 18/01568-4 - BE SOI MOSFET transistors optimization for biosensor platform application
• Grantee: Leonardo Shimizu Yojo
• Support Opportunities: Scholarships in Brazil - Doctorate