Open Access
Hatipoglu, Gokhan
Graduate Program:
Electrical Engineering
Doctor of Philosophy
Document Type:
Date of Defense:
June 09, 2015
Committee Members:
  • Srinivas A Tadigadapa, Dissertation Advisor
  • Srinivas A Tadigadapa, Committee Chair
  • Suman Datta, Committee Member
  • Sumeet Kumar Gupta, Committee Member
  • Nitin Samarth, Committee Member
  • Magnetic Sensors
  • Magnetostriction
  • Quartz resonators
Magnetic sensor technologies are gaining further importance as they are used in smart mobile devices, navigation and especially to bio-imaging. Despite the fact that the sensitivity has been the major magnetic sensor specification, the bulkiness, overall price and the power consumption of the sensors also act as important specifications for practical applications. Micro-Electro-Mechanical Systems (MEMS) are the chip-scale actuators and sensors, which can offer very sensitive sensors that can i) operate at room temperature, ii) consume very low power iii) be manufactured in arrays iv) the overall cost is extremely cheap. In this dissertation, two different MEMS magnetic sensors are proposed utilizing quartz micro-resonators. First sensor operates on quantifying the magnetoviscosity of ferrofluids. Magnetovisvosity is defined as the viscosity changes occuring at ferrofluids as a function of applied magnetic field. These viscosity changes as a function of magnetic field are detected utilizing micromachined AT-cut thickness Shear Mode Quartz arrays acting as micro-viscometers. Ferrofluids typically consist of 10  3 nm iron oxide (Fe2O3) nanoparticles suspended in mineral oil. However, the high magnetic susceptibility of ferrofluid suspensions results in the modulation of the magnetoviscosity due to applied magnetic fields. These viscosity changes due to in-plane incident field shifts the at-resonance admittance characteristics of µQCR and is tracked in real time to achieve a novel magnetic sensing mechanism to detect and quantify the low frequency, low strength magnetic fields For improved sensitivity, the in-plane sensed magnetic flux density is concentrated using a high relative permeability (µr = 45000 as bulk) thin film of Metglas® (Fe85B5Si10) deposited on the resonator electrode. Furthermore, by patterning the Metglas® film in a bow-tie shape and aligned at the center of the µQCR electrode 2D vector sensing is achieved. Using these improvements, a minimum detectable field of 1.5 nT/√Hz at 1 Hz has been experimentally demonstrated. Furthermore, the high frequency and small amplitude shear waves are created with the Quartz resonators and the highly magnetoviscous interface is modeled. The ferrofluid is modeled as a viscous loading on the thickness-shear mode resonator via the modified Buttorworth-Van-Dyke Model. This sensor will be referred as the magnetoviscous sensor throughout the dissertation. Second magnetic sensor concept also exploits micro-machined quartz resonators’ admittance shifts, but the mechanism is based on the transverse force-frequency effect. Magnetostrictive Metglas thin film coated AT-cut thickness shear mode (TSM) Quartz thin plate microresonator structures are uniquely released using focused ion beam (FIB) milling. Therefore, the whole structure is able to do flexural bending due to magnetostriction induced as a result of applied magnetic field in the magnetic thin film that is elastically coupled to the quartz microresonator. As a result of transverse loading and bending, the admittance characteristics of the resonator shifts. The transverse force-frequency sensitivity as a function of azimuth angle for magnetostrictive thin film coated AT-cut thickness shear mode (TSM) quartz thin plate are experimentally tested and successfully modeled via coupling Lee’s theory and magnetostrictive unimorph equations. The first device (500 x 500 x 19 µm) has been tested as the proof-of-concept device, where 1 µT at 10 Hz was measured. The latter thinner device (750 x 500 µm x 7.5 µm) is the optimized device, where the sensitivity is improved to 79 nT at 10 Hz input frequency inducing 94.5 nS of conductance shift. This corresponds to a frequency shift of Δf / f0 that is equal to 1.4692 x 10-9. The ultimate expected sensitivity for this system is simulated using the coupled domain analysis. According to the theoretical model, it is predicted that low nanoTesla to high picoTesla magnetic flux densities would be detected after further thickness reduction. To sum up, both sensors are proposed and studied as unique MEMS sensing mechanisms. They i) are relatively low power consuming devices, ii) can be manufactured in arrays, which improves the spatial density, and iii) can both operate at room temperature. All aspects of the devices, sensing mechanisms along with the experimentations and theoretical modeling will be discussed in this dissertation.