ORIGINS OF MICROSCALE COOLING
The development of advanced cooling solutions for microelectronics has long been an active topic of interest for many companies and research institutions. Thousands of scientific articles can be found in this field ever since microelectronics were invented, as thermal management has always been a limitation. The origins of micro-scale cooling approaches date to the 1980's when researchers realized how heat removal from silicon chips was significantly enhanced by circulating liquid through etched micro-channels. However, associated challenges included high flow restrictions, complexity to manufacture, and integration at competitive costs. Although many significant and exciting advancements have been made since then, the majority of these have stayed as fundamental demonstrations published in research journals. Bridging technology breakthroughs from a research laboratory to actual industry applications or consumer products is a tremendous challenge that requires a significant amount of financial investment and human resources. In many cases, no matter how promising the breakthrough may look like, the technology will be doomed if cost-effective methods for transition to market are not found. 
RESEARCH ON MICROFLUIDICS AT GEORGIA TECH
Among the many institutions dealing with advanced cooling technologies, Georgia Institute of Technology  (Georgia Tech) stands as one of the most active ones across the world. The G.W. Woodruff School of Mechanical Engineering at Georgia Tech is one the largest in the United States, with over 100 academic faculty members. From its many laboratories, EMCOOL's embedded microfluidic technology was first developed and demonstrated at the Microelectronics & Emerging Technologies Thermal Laboratory (METTL), which has over 30 years of trajectory at the forefront of heat transfer technologies, and is led by Dr. Yogendra Joshi. Microfluidic cooling has been a main research activity at the METTL in the past two decades, where pioneering studies have taken place. In these studies, the use of embedded microstructures on the back side of the silicon wafer was proposed in order to eliminate contact and spreading heat resistances; thereby, allowing the removal of larger amounts of heat at lower device temperatures. Although this concept might sound straightforward, its research and experimental demonstration were quite a challenge that took several years and resources. Using the cleanroom facilities at the Georgia Tech Institute for Electronics and Nanotechnology (IEN) through a sophisticated process called Deep Reactive-Ion Etching (DRIE), the microfabrication of heat transfer structures with high aspect ratio was achieved by Georgia Tech researchers in silicon wafers. 
Figure 1. Scanning Electron Microscopy (SEM) picture of heat transfer microstructures on a silicon wafer. (Photo courtesy by METTL).
Figure 1 shows a Scanning Electron Microscopy (SEM) photograph of microfabricated cylindrical pin fins on a silicon wafer; these microstructures significantly increase the surface area available for heat transfer and therefore allow the removal of higher power densities. The bigger pin fins in the picture have a diameter of 150 micrometers, while the center pin fins have a diameter of 75 micrometers. For the sake of comparison, a human hair has a very similar diameter ranging between 40 to 180 micrometers. These pin fins have a height of 200 micrometers, and liquid flows through the gaps between the structures. For a typical silicon die size of 10 mm on each side, we are talking about nearly four thousands of such micro pin fins for cooling the billions of transistors generating the heat on the other side of the silicon die.
The results from a series of investigations by the METTL and collaborating Georgia Tech teams, indicated that by using this approach of embedded microfluidic cooling, heat dissipation could be as high as 500 to 1,000 Watts from the small area of 1 square centimeter (similar to the area of a thumbnail).  To put these numbers in perspective, an Intel Core™ i9-9900K microprocessor has a maximum power consumption (when heavily overclocked and under synthetic stress loads) of around 200 W (twice as more as the manufacturer thermal design power of 95 W when running at base frequency); this processor has an area of around 2 square centimeters, therefore being just 1/5 to 1/10 of the demonstrated cooling capacity by METTL.
Video 1 shows the flow of a dielectric coolant (HFE7200) though the small gaps formed between the microstructures, where very interesting flow patterns followed by the fluid (from left to right) can be observed. Such videos were taken by a specialized microscopic lens, coupled with a high-speed (PHANTOM™) camera, allowing the investigation and better understanding of the flow physics at the microscale. Results from these investigations allowed the researchers at METTL to optimize the hydraulic features and thermal performance of these devices.
Video 1. Microscopic, high-speed video of dielectric coolant (HFE7200) flowing through micro pin fins on a silicon wafer. (Video courtesy by METTL, taken at 2000 FPS and played at 5 FPS).
With these promising results, researchers at METTL refined the technology for years at their laboratory with the ultimate objective of finding ways to integrate such promising technology for industrial and/or consumer applications. For those interested in how this technology was exhaustively studied in the last decade by the METTL through multiple numerical and experimental investigations, a record of publications of these scientific articles is available in this link.
TRANSITIONING TO COMMERCIALIZATION- VENTURELAB - NSF I-CORPS
Although the amazing results obtained by the METTL laid the foundations of practical microfluidic cooling into silicon devices, going from a laboratory demonstration to a reliable, and cost-effective solution for actual applications is a difficult task. Bearing such challenges in mind, the researchers from METTL approached VentureLab, an organization that helps identifying technologies, based on Georgia Tech research, with a potential for commercialization. After an initial assessment of the technology and potential applications, VentureLab guided the researchers from METTL through the several steps for transitioning the technology to commercialization. With the mentoring of Jonathan Goldman, a seasoned entrepreneur and VentureLab's Principal, a team was created and accepted to participate in the National Science Foundation Innovation Corps (I-Corps). After an intensive 6-week customer discovery bootcamp, and more than 100 interviews with stakeholders in potential fields for implementing the developed technology, the team identified their main customer segments, value propositions and business structure.
​
​
GEORGIA RESEARCH ALLIANCE
The Georgia Research Alliance (GRA) is a non-profit organization that works with the University System of Georgia and the Georgia Department of Economic Development, providing commercialization grant programs for startup companies based on research from universities in the state of Georgia. GRA provided funding to the researchers leading the technology transition at METTL, helping them to reduce market and technical risks as they worked towards commercialization. Through the clever use of GRA funding, the METTL team was able to develop new prototypes for microfluidic cooling based on its original research, but this time with a new focus on cost-effectiveness for business purposes. After the successful demonstration of the cooling capabilities of these new (market-oriented) prototypes, EMCOOL Inc. was formed in May 2019 to license the technology from Georgia Tech Research Corporation  (GTRC), being the first company in the world to bring a reliable and cost-effective embedded microfluidic cooling™ solution to the market. 
MOVE TO NEW LOCATION
REMARKS
Although we tried our best to effectively summarize the history of how this technology was conceived and evolved, there are many details we did not cover in this brief. An important remark is the timeline, the fundamental research on this technology took more than 5 years through multiple iterations, and the transition to commercialization took nearly 3 years of continuous efforts to finally bring a product that end users can purchase. A special acknowledgement is given to the numerous staff at Georgia Institute of Technology, Georgia Research Alliance, and National Science Foundation that helped in one way or another to make this possible.
​
SUPPORT
FAQs
Shipping
Payment methods