Exploring the Limits of Cooling for Extreme Heat Flux Applications: Data Centers and Power Electronics

Default ARPA-E Project Image

OPEN 2018
Stanford, California
Project Term:
09/13/2019 - 03/12/2023

Critical Need:

Energy-efficient computing and heterogeneous integration (assembling many separately manufactured components onto a single chip to improve functionality and performance ) significantly reduce energy demand for computing needs. The problem is when power density is increased so does the heat generation, reducing the effectiveness of conventional cooling technology solutions. It is critical to develop a high heat flux cooler with the lowest possible thermal resistance and pressure drop for high power electronics and microprocessors, which are of great interest to the automotive, avionics defense, information technology, and semiconductor industries, among others.

Project Innovation + Advantages:

Stanford will develop an innovative cooling technology, the Extreme Heat Flux Micro- (EHFμ-) Cooler, to improve reliability and performance in power electronics by offering improved chip thermal management. The cooler employs a novel liquid wicking, thin-film evaporator, with microchannels to route liquid and the resulting vapor, with the net effect of improved heat removal rates at manageable pressure drops. This significantly increases heat flux thereby reducing the device (chip) temperature. The design could increase heat transfer rates by an order of magnitude compared with today’s corresponding state-of-the-art cooling technologies. Improved cooling devices could greatly increase efficiency, reliability, and performance for microprocessors and power electronics.

Potential Impact:

The EHFμ-Cooler significantly reduces device temperature, resulting in 25-50% improvement in reliability and device performance.


The proposed EHFμ-Cooler technology helps the United States to maintain its technological lead in high-performance computing and high-power electronics.


By improving energy efficiency, the technology wouuld reduce energy-related emissions such as greenhouse gases.


Better cooling, leading to more efficient, higher performance and more reliable power electronics, would save the nation a significant amount of primary energy. Just for automotive applications, a 10% increase in efficiency for silicon carbide-based power electronics equates to >20 million gallons of fuel savings annually.


ARPA-E Program Director:
Dr. Peter de Bock
Project Contact:
Prof. Kenneth Goodson
Press and General Inquiries Email:
Project Contact Email:


National Renewable Energy Laboratory

Related Projects

Release Date: