Standardized Heat Spreader Design for Passive Cooling of Interconnects in the BEOL of ICs

Helou, Assaad El; Venugopal, Archana; Raad, Peter E.

doi:10.1109/tcpmt.2020.2974793

Cited by 3 publications

(1 citation statement)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Past mitigation strategies to suppress self-heating have been reported using active cooling techniques such as microchannel cooling [12][13][14] and jet impingement cooling [15], or passive cooling techniques such as using higher thermalconductivity substrates [16][17][18], or adding a heat-spreading material closer to the active regions to more effectively dissipate the generated heat [19][20][21][22]. The assessment of any proposed thermal management solution, such as different substrates used in this study, requires both physical modeling and complementary experimental thermal mapping for validation.…”

Section: Introductionmentioning

confidence: 99%

Full thermal characterization of AlGaN/GaN high electron mobility transistors on silicon, silicon carbide, and diamond substrates using a reverse modeling approach

Helou¹,

Cui²,

Tadjer

et al. 2020

Semicond. Sci. Technol.

Self Cite

View full text Add to dashboard Cite

Gallium nitride (GaN) high electron mobility transistors (HEMTs) operate at high power levels and are thus especially thermally-critical devices. Not only do they require innovative thermal management strategies, but can also benefit from advanced experimental thermal characterization, both numerical and experimental, in their design and system integration stages. The thermal numerical analysis of microelectronic devices faces the challenges of complex physics and uncertain thermophysical properties which leads to numerically expensive models that are prone to error. By the use of an innovative reverse modeling approach to mitigate the above challenges, this work presents the full thermal characterization of GaN power devices with different substrates aimed at managing performance-limiting self-heating. The approach develops and optimizes a thermal simulation model to match the numerical results to experimentally-obtained thermal maps of the devices under test. The experimentally-optimized simulation model can then be used to extract full 3D temperature distributions, infer in-situ thermal properties, and provide a numerical platform that can be used to conduct further parametric studies and design iterations. The presented analysis provides a full thermal characterization of different GaN HEMT devices and compares the thermal performance of different substrates on the basis of thermal properties. The extracted properties for HEMTs on Si, SiC, and Diamond substrates are compared and a set of conclusions are presented to guide further developments in GaN HEMT thermal management strategies.

show abstract