| Description |
The design, working principle, fabrication, and characterization of ultrasensitive ferromagnetic and magnetoelectric magnetometer are discussed in this thesis. Different manufacturing techniques and materials were used for the fabrication of the two versions of the magnetometer. The ferromagnetic microelectromechanical systems (MEMS) magnetometer was fabricated using low-pressure chemical vapor deposition (LPCVD) of silicon nitride, yielding low compressive stress, followed by patterning. The built-in stress was found to be -14 Mpa using Tencor P-10 profilometer. A neodymium magnet (NdFeB) was used as a foot-mass to increase the sensitivity of the device. A coil (Ø=3 cm), placed at a distance from the sensor (2.5-15 cm), was used to produce the magnetic field. The response of the ferromagnetic MEMS magnetometer to the AC magnetic field was measured using Laser-Doppler vibrometer. The ferromagnetic sensor's average temperature sensitivity around room temperature was 11.9 pV/pT/-C, which was negligible. The resolution of the ferromagnetic sensor was found to be 27 pT (1 pT = 10-12 T). To further improve the sensitivity and eliminate the use of the optical detection method, we fabricated a Lead Zirconate titanate (PZT) based magnetoelectric sensor. The sensor structure consisted of a 9 mm long, and 0.17 mm thick PZT beam of varying widths. A neodymium permanent magnet was used as a foot-mass in this case as well. The magnetic field from the coil generated a driving force on the permanent magnet. The driving force displaced the free end of the PZT beam and generated a proportional voltage in the PZT layer. The magnetoelectric coupling, i.e., the coupling between mechanical and magnetic field, yielded a sensor resolution of ~40 fT (1 fT = 10-15 T); an improvement by three orders of magnitude. We used high permeability Mu sheets (0.003"") attached to copper plates (0.125"") to shield stray magnetic fields around the sensor. For both the ferromagnetic MEMS and the magnetoelectric magnetometer, the initial output was improved by using external bias and parametric amplification. By applying an external DC magnetic field bias to the sensor, the effective spring compliance of the sensor was modified. Electronic feedback reduced the active noise limiting the sensor's sensitivity. We used magnetic coupling to enhance the sensors' sensitivity and to reduce the electronic noise. Two identical sensors, with identical foot-mass (permanent magnet), was used to show coupling. The magnet on one of the sensors was mounted in NS polarity, whereas, on the other it was in SN polarity. When excited by the same external AC magnetic field (using coil), one of the sensors was pulled towards the coil and the other was pushed away from it. Adding the individual sensor output, using a preamplifier, an overall increase in sensors' output was observed. The techniques mentioned above helped to improve the output from the sensor. The sensitivity of the sensor can be improved further by using a 3-axis magnetic field cancellation system, by eliminating the AC and DC stray magnetic field, by using coupled-mode resonators and by increasing the surface field intensity of the foot-mass. The magnetometers, thus, developed can be used for mapping the magnetic print of the brain. |
