ISTP-33

Multidisciplinary optimization design for electronic devices： based on in-situ measurements of multi-physical fields and integration evaluation

Professor Xing Zhang

Key Laboratory for Thermal Science and Power Engineering of Ministry of Education,
Department of Engineering Mechanics, Tsinghua University,
Beijing, China.

Electronic devices are widely used in electronic manufacturing, information and communication technology, and other fields. In order to improve information transmission capacity, electronic devices need high power and operate at a higher frequency, significantly increasing heat generation. Thus, the cooling problem has become a critical bottleneck to be broken through, while the most common solution is to optimize the device design. However, effective optimization is based on a precise evaluation through the design parameters of the device and the physical properties of the semiconductor, which will be greatly influenced by multiple physical fields and must be accurately determined. Unfortunately, the physical field inside the electronic device is complex and its performance is controlled by many factors. The optimization considering only one factor is difficult to effectively improve the device performance. Therefore, it is necessary to develop a system optimization scheme considering multi-disciplines, multi-factors, and multi-performance, and the in-situ integrated characterization of the multiple physical fields in the device has become the premise of system optimization.

However, the in-situ integrated measurement of multiple physical fields is still challenging. Raman spectroscopy method is the most promising approach to in-situ simultaneously obtaining multiple physical fields, while high-precision decoupling of the influence of multiple physical parameters, such as temperature and stress, is still a difficult problem. The spatial and temporal resolution of the traditional Raman method also limits its application in miniaturized, integrated, and high-frequency electronic devices. To solve this problem, we built a temperature stress coupling model, deduced the relationship between the measured effective Raman temperature coefficient (ERTC) with the structural constraints in the devices, and determined the influence proportion of temperature and thermal stress in a single Raman peak shift when the local temperature rises. Based on the above analysis, we developed a method to realize the integrated measurement of multiple physical fields with only one Raman peak position, which significantly decreases the measurement uncertainty compared with the previous multi-peak decoupling method by avoiding using the Raman peaks with low signal-to-noise. On this basis, we developed a Gaussian inversion algorithm to increase the spatial resolution of Raman spectral measurements from the optical diffraction limit to 100 nm. Furthermore, we propose an electro-optical synchronous modulation flash measurement method, to determine the transient fluctuation of the physical quantity in the device with 100 ps temporal resolution. It was found that the transient temperature difference in one period of the amplitude modulation device (3V/10V, 0.1kHz) can reach 20K, and the frequency varies from 0.1 kHz to 1000 kHz can lead to a temperature rise of 160K. The transient measurement results of devices under different heat dissipation conditions show that the relative transient temperature difference in one period can be regarded as a new index to evaluate the heat dissipation performance of devices.
Based on the in-situ integrated characterization method, we proposed the multidisciplinary integration evaluation and system optimization principle of electronic devices. The coupling influences of multiple physical fields were considered in the evaluation and optimization. Through in-situ stress measurement, we evaluated the possible burnout position of a GaN-based HEMT device, which is not a central position but a tensile stress concentration point in the boundary, and this evaluation result is completely consistent with the burnout failure test result. Optimization simulation analysis shows that, by regulating the normal (1/10 to 10 times) and in-plane stresses (-3% to 3%), the interface thermal resistance and thermal conductivity of the semiconductor layer in GaN-based devices will significantly change, and the temperature variation of the device hot spot can decrease 50 K. This result strongly shows the necessity and importance of multidisciplinary integration optimization.

About Professor Xing Zhang

Professor Xing Zhang is the Director of the Institute of Engineering Thermophysics in the School of Aerospace Engineering at Tsinghua University, Beijing, China. He received his Ph.D. degree from Tsinghua University in 1988 and worked as a Lecturer at Southeast University after his graduation.
From 1990 to 2006, he worked as a Research Associate, an Assistant Professor and an Associate Professor at Kyushu University in Japan. He returned to Tsinghua University as a Professor in 2006.
His current research interests include micro/nanoscale heat transfer, thermophysical properties of nanostructured materials, multiscale cooling technology for data centers, multidisciplinary optimization design for electronic devices and the efficient use of wind/solar/hydrogen energy sources etc.
He has published over 400 refereed journals and conference publications, and delivered more than 60 Plenary, Keynote, and Invited Lectures at major technical Conferences and Institutions. He serves as the Presidents of Assembly for International Heat Transfer Conference (AIHTC) and Asian Union of Thermal Science and Engineering (AUTSE).
He received the “Best Paper Award” from the Heat Transfer Society of Japan in 2021 and 2008, the “Thermal Engineering Award for International Activity” from JSME in 2020, the “Hartnett-Irvine Award” from International Center for Heat and Mass Transfer (ICHMT) in 2019, the “Natural Science Award (First Class)” from the Ministry of Education of the People’s Republic of China in 2018, the “Significant Contribution Awards” from the 10th Asian Thermophysical Properties Conference in 2013, and the “National Natural Science Award (Second Class)” from the State Council of the People’s Republic of China in 2011.