Honors Thesis Examination Committee:

ii In the past, design of transmit/receive coils for magnetic resonance imaging was performed using simple lumped circuit models; however, as developments in technology have allowed for increasingly high magnetic fields, the frequency at which the imaging must be performed has increased to the hundreds of megahertz, causing technical difficulties to arise. This project seeks to overcome the problem of inhomogeneous magnetic excitation through a redesign of the resonator coil based in electromagnetic theory. Several designs were attempted, with the primary modifications being in the dielectric filling and length of the resonator. Since the lumped circuit model is no longer valid at today’s high fields, it must be replaced by a full-wave electromagnetic solution. Therefore, all simulations for this project were performed with a FDTD program previously developed at Ohio State for this purpose. The simulations have shown that a significant increase in homogeneity can be achieved by loading the resonator with a high permittivity dielectric. This in turn lowers the resonant frequency, so the cavity length must be shortened to raise the frequency back to normal values. The best case design found here is with a 10 cm resonator filled with relative dielectric of 60. A better design can likely be found with a resonator of approximately 8 cm in length, but this was unable to be tested due to technical difficulties and time constraints. The project also confirms that dielectric loading of a certain region will cause the field peak to shift towards the load. This could be useful for locating the field peak at a certain point in the head in order to obtain an MRI image in just that area