Experimental setup for three-dimensional thermal imaging using a Nipkow disk confocal microscope and phase-locked CCD detector for lock-in thermoreflectance measurements.
Thermoflectance measurements of an encapsulated gold heater comparing confocal imaging (closed squares) to widefield imaging (open circles). Reflection from the encapsulant causes a 6% decrease in signal for widefield. When the surface reflectivity is increased by sputtering the sample with gold, the widefield signal drops by 80% (open triangles) but the confocal signal decreases by less than 5%. The hot areas at the edge of the heater are an artifact caused by thermal lensing in the encapsulant.
(Left) Reflected intensity measured by the confocal microscope vs focal depth, showing bright reflections from the top and bottom of gold-sputtered thin films of two different film thicknesses: , . (Top right) Schematic of the thin film confocal measurement, in which a piezostage is used to adjust the depth of focus and a CCD records the reflected intensity. (Bottom right) Film thickness measured using the transmission spectra (spectrometer measurement) vs the calibrated thickness measurement using the confocal microscope. The theoretical calibration constant, , is within 10% of the experimental results.
Schematic of the gold heater sample and coverslip obstruction used for contrast measurements. Upward-pointing arrows represent light reflected from the coverslip and sample surfaces.
(Left) Widefield contrast (black circle), confocal contrast (gray triangle), and widefield contrast predicted using confocal measurements (open triangle) as the obstructing slide reflectivity is increased. (Right) Confocal and widefield microscope images of the gold heater, obscured by a gold-sputtered glass slide.
(Top) Average heater temperature signal, , vs injected power for widefield and confocal thermoreflectance both with and without an obstructing coverslip. Even though temperature remains constant, the presence of the coverslip strongly decreases the dynamic range and the value of of the widefield measurements; (top right) widefield and confocal thermoreflectance images; (center) predicted and measured decrease in thermal dynamic range due to reduced contrast from coverslip reflections, and (bottom) the same data replotted as a function of the relative reflectivity of the obstruction.
Article metrics loading...
Full text loading...