Search

This study presents a straightforward, efficient dual-mode thermoreflectance microscopy (TRM) system for analyzing the stationary and dynamic thermal properties of microelectronic devices. The proposed TRM system employs a standard visible microscope and implements two distinct imaging process schemes to obtain thermal images at different time scales. For TRM imaging of an ohmic microdevice, the optimal probing wavelength is predetermined from the thermoreflectance spectrum, acquired using a white light-emitting diode source and tunable bandpass filters to enhance thermal sensitivity. Thermoreflectance images are obtained in both stationary and transient modes using a four-bucket method and a pulsed boxcar averaging method, respectively. This thermography approach demonstrates highly sensitive, time-integrated stationary thermal imaging and time-gated transient thermal imaging with a temporal resolution of 200 ns. The system shows significant potential as an analytical tool with a sub-micron spatiotemporal resolution for evaluating heat generation and self-heating behavior in microelectronic devices. [ABSTRACT FROM AUTHOR]

In this paper, we propose a sensitivity-enhanced thermoreflectance microscopy (TRM) system employing an electronic auto-balancing photoreceiver for effective thermal fault localization through a Si substrate. The proposed system in the auto-balanced detection not only improves the system signal-to-noise ratio (SNR) by approximately five times compared to that achievable via normal detection but also mitigates the inherent system noise, such as pseudothermoreflectance signals, by using the optical zooming. Moreover, temperature variations as small as ~47 mK were detectable under the measurement conditions employed in this paper. Finally, we experimentally demonstrated clear localization of a thermal fault, which was dimly visible in the normal backside detection results, in an optical zoomed thermal image. [ABSTRACT FROM AUTHOR]