In the first part, the thermal and mechanical challenges in microelectronics that range from device to data center room level related to an IUCRC Center called Energy-Smart Electronic Systems (ES2) will be presented. At the board-level, we are concerned with thermal management of high-power modules through design and experimental testing of a first-ever “dynamic cold plate” concept (indirect liquid cooling). Related to immersion cooling, an experimental study will be presented on a single server fully submerged horizontally in white mineral oil. Air-side economization related to the building of an Aztec ASC-15 cooling unit in collaboration with Mestex and best practices of using airside economization with or without direct and indirect evaporative cooling will be addressed. I will also discuss our study in particulate and gaseous contaminants, which is a key concern in using airside economizers. The second part will address the proposed establishment of a new center: Reliability Assessment for Micro and Power Electronic Systems (RAMPES). More than Moore roadmap plans to increase functionality density through heterogeneous integration (HI). In HI, components with different functionality such as DRAM and processors are packaged together. The proposed center will focus on extensive material characterizations at different length scales, testing and design for reliability and implementation of additive manufacturing for cost-effective cooling of HI systems. The reliability work at the center will focus on accurate material characterizations, reliability tests, and failure analysis. With the expansion of heterogeneous integration to new domains such as mobile products, big data systems, internet of thing, and automotive, there is a need to understand new failure

modes and mechanisms of HI, perform extensive sets of reliability tests, and obtain accurate material properties for finite element analysis. Furthermore, the Center envisions development of cooling technologies that enable a holistic integration of the multi-scale thermal design of reliable and energy efficient  electronic systems. For smart allocation of on-demand cooling in response to workload dynamics of electronic systems, the Center seeks to develop diverse predictive and prognostic models, based on the principles of thermodynamics, transport phenomena and machine learning.

 

Bio