UC Berkeley

Detecting depth varying thermal properties is desirable for applications such as cryosurgery, landmine detection and non-destructive testing. This can be accomplished by periodic heating, exploiting the resulting frequency dependent penetration depth. Using this, objects with thermal conductivity distinct from their surroundings (such as material flaws or landmines) can be detected. In cryosurgery, for treatment of atrial fibrillation, it is desirable to track the location of moving boundaries between two phases (i.e. frozen and thawed tissue) with sub-millimeter precision. This spatial resolution is not attainable with conventional imaging.

In this dissertation, the 3ω thermal property measurement technique is used to make new types of measurements on systems changing in real time. Analytical understanding of measurements with multiple frequency excitations is first developed, and verified with experiments. Next, the 3ω technique is used to sense thickness of samples (200μm and 500μm ice). The same sensors are used to detect step changes in thermal properties, including phase change, of water and mouse liver samples in real time. Finally an experimental setup is constructed to track an oil-air interface as it moves. Numerical modeling is used to convert measured 3ω voltages into spatial dimensions. For slow front velocities, the location of the oil-air front determined from the 3ω approach is found to be accurate to within an average error of under 18μm (RMS) for front distances between 12 and 360μm.