G. Brocero, D. Kendig, A. Shakouri, Y. Guhel, P. Eudeline, J. Sipma, B. Boudart
{"title":"Innovative submicron thermal characterization method for AlGaN/GaN power HEMTs with hyperspectral thermoreflectance imaging","authors":"G. Brocero, D. Kendig, A. Shakouri, Y. Guhel, P. Eudeline, J. Sipma, B. Boudart","doi":"10.1109/CSICS.2017.8240438","DOIUrl":null,"url":null,"abstract":"Significant advances in AlGaN/GaN heterostructure based technologies in the last decade, with AlGaN/GaN high electron mobility transistors (HEMTs) has led to high power performance at Gigahertz frequencies for communication, space, radar, and defense applications. The conjunction of the remarkable properties of the AlGaN/GaN heterojunction and the high thermal conductivity of the silicon carbide substrate enables GaN on SiC-based high electron-mobility transistors (HEMTs) to be very efficient for RF and microwave applications. However, this very high power density leads to self-heating under operating conditions and has important consequences for both performance and reliability. Since these devices, over time, are going to be increasingly smaller and more powerful a method for measuring the self-heating is of great interest for thermal management, temperature control, and optimizing simulation software. This paper presents early results of a new high spatial resolution thermal characterization technique using a thermoreflectance imaging system on a commercial sample. This technique, hyperspectral thermoreflectance imaging, enables us to obtain a clean thermal image in CW mode with 45 nm spatial resolution. We will show the thermal imaging results for a AlGaN/GaN HEMT on a SiC substrate. Although the zone of interest has narrow geometry and some grainy surfaces, the results show a very good linearity of reflection response with changing temperature with a significantly smaller error.","PeriodicalId":129729,"journal":{"name":"2017 IEEE Compound Semiconductor Integrated Circuit Symposium (CSICS)","volume":"53 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"3","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2017 IEEE Compound Semiconductor Integrated Circuit Symposium (CSICS)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/CSICS.2017.8240438","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 3
Abstract
Significant advances in AlGaN/GaN heterostructure based technologies in the last decade, with AlGaN/GaN high electron mobility transistors (HEMTs) has led to high power performance at Gigahertz frequencies for communication, space, radar, and defense applications. The conjunction of the remarkable properties of the AlGaN/GaN heterojunction and the high thermal conductivity of the silicon carbide substrate enables GaN on SiC-based high electron-mobility transistors (HEMTs) to be very efficient for RF and microwave applications. However, this very high power density leads to self-heating under operating conditions and has important consequences for both performance and reliability. Since these devices, over time, are going to be increasingly smaller and more powerful a method for measuring the self-heating is of great interest for thermal management, temperature control, and optimizing simulation software. This paper presents early results of a new high spatial resolution thermal characterization technique using a thermoreflectance imaging system on a commercial sample. This technique, hyperspectral thermoreflectance imaging, enables us to obtain a clean thermal image in CW mode with 45 nm spatial resolution. We will show the thermal imaging results for a AlGaN/GaN HEMT on a SiC substrate. Although the zone of interest has narrow geometry and some grainy surfaces, the results show a very good linearity of reflection response with changing temperature with a significantly smaller error.