SUBJECT: Ph.D. Proposal Presentation
   
BY: Alireza Mahdavifar
   
TIME: Thursday, November 20, 2014, 3:00 p.m.
   
PLACE: Love Building, 210
   
TITLE: Computational and Experimental Development of Ultra-Low Power and Sensitive Micro-Electro-Thermal Gas Sensor
   
COMMITTEE: Dr. Peter Hesketh, Chair (Mechanical Engineering)
Dr. Mostafa Ghiaasiaan (Mechanical Engineering)
Dr Todd Sulchek (Mechanical Engineering)
Dr Shannon Yee (Mechanical Engineering)
Dr. Hamid Garmestani (Materials Science and Engineering)
 

SUMMARY

In the present work we propose development of a miniature, ultra-low power, and sensitive, microbridge-based gas sensor. The batch fabrication of the microbridge is a CMOS compatible process. The structural material of the bridges consists of doped polysilicon with critical dimension of 500 nm. Small size of the sensor along with innovative measurement techniques reduces the power required to operate the sensor to about 3 ÁJ per measurement while increasing the measurement sensitivity. It makes the sensor suitable for growing market of gas sensing applications especially as detector for portable gas chromatography solutions. Operation of the sensor includes complex interactions between thermal and electrical phenomena; the output signal depends on variable resistance of the microbridge and the supplied power on one hand, and the heat capacity of the structure along with thermal conductivity of the gas medium on the other hand. Considering that the scales of phenomena are very small, a model of the microbridge was developed to provide accurate insights on the physics of the problem; the COMSOL model that couples electrical and thermal physics together and includes minimal simplifications is the most comprehensive modeling on thermal conductivity detectors (TCDs) so far. Heat loss from constant electric source was observed to be a function of the thermal conductivity of the gas ambient, resulting in different magnitude of resistance change. The sensor is experimentally tested in nitrogen, carbon dioxide, helium and methane. Modeling results coincide with experimental data in predicting resistance changes. An experimental setup including gas flow system and data acquisition tools was developed for the tests. LabVIEW programming, circuit design and data processing are other aspects of this work. Based on our preliminary results we propose a new method of detection based on time constant of sensors transient response. The new method further reduces power consumption and enhances measurement accuracy. Future study include more experiments in different gas mixtures, study effect of humidity, integration of portable and wireless electronics, and finally finite element analysis of thermal stresses.