Accuracy Errors of Modeling of MIS-Transistor Sensor Elements

  • PDF / 1,364,634 Bytes
  • 18 Pages / 612 x 792 pts (letter) Page_size
  • 67 Downloads / 153 Views

DOWNLOAD

REPORT


NSORS AND SYSTEMS

Accuracy Errors of Modeling of MIS-Transistor Sensor Elements B. I. Podlepetsky National Research Nuclear University “MEPhI,” Moscow, Russia e-mail: [email protected] Received December 5, 2018 Revised January 16, 2019 Accepted May 25, 2020

Abstract—Assesses the accuracy characteristics of circuits’, electrical and electrophysical models of MISFET transistor elements of the sensors, taking into account the errors arising from simplifying assumptions, approximations, extrapolations and experimental dispersions’ characteristics in determining the parameters of the models and measured physical quantity. Keywords: MIS transistor, models, approximations, errors of models DOI: 10.1134/S0005117920100094

1. INTRODUCTION Transistors with the metal-dielectric-semiconductor structure (MISFET) are used in electronic devices and systems as elements of integrated circuits (IC), and as sensitive elements (SE) of temperature, pressure, force, light radiation and radiation power, humidity, concentrations of ions and atoms in liquids, and concentrations of gas molecules [1]. Therefore, the MISFET can be considered as a multi-functional SE of sensors of various physical quantities X. To evaluate the metrological and operational characteristics of sensors, devices and systems, simulation can be used, which is considered a relatively fast and inexpensive method compared to direct tests of electronic components. When modeling and designing MISFET-based ICs and sensors, technological and circuit-based CAD systems (for example, TCAD, SPICE, VERILOG-A) are used, based on physical, electrical, and engineering-physical models of MISFET [2–8]. In most models, MISFETs are considered as elements of electronic circuits. When MISFET is used as a SE, the direct application of known models of MISFETs and CAD for the design of sensors and devices becomes difficult or impossible without taking into account the technological features of the sensors and the operational conditions of their applications. First of all, this applies to the subscript (customized) parameters of models, which can only be determined experimentally. The basic characteristic of sensors and converters is the conversion function–the dependence of the informative parameter of the output signal Y on the informative parameter of the input signal X. When using sensors, an important metrological characteristic is the measurement error of value X, which is determined by the error’s band based on the experimental dependence of the calibration characteristic X(Y ) (the inverse function of the conversion function Y (X)). As a rule, models of sensor’s conversion functions are developed based on experimental researches. Researches includes the following procedures: 1) you select the electronics’ circuits to test the sensor; 2) you fabricate SEs and circuits of sensors; 3) the sensors’ dependencies Yi (Xj ) are repeatedly measured with given parameters influencing factors Z; 4) the average values of Va (Xj ) determined and their variances; 5) approximation and extrapolation of