Home > Research > Micro-Cooling

Micro-Cooling

In the ever-shrinking land of MEMS, heat can be a problem. As these devices shrink in size, power per unit volume increases, and components get hot. Blowing cool air over them is not very effective because the heat capacity of air does not provide enough cooling for these tiny areas. A possible solution is to cycle fluid cooled via a MEMS-scale thermoacoustic refrigerator around a device, according to Professor Cila Herman, an expert in thermoacoustic refrigeration.

A thermoacoustic refrigerator uses sound energy to transport thermal energy. Using a sound source such as a loudspeaker, a standing wave is set up in a tube filled with noble gas. As the wave travels back and forth in the chamber, the gas compresses (heating up) and expands (cooling off). The gas also oscillates in the chamber. To exploit this energy, a “thermoacoustic core,” consisting of a densely packed stack of plates, is placed in the chamber. As the gas oscillates, it compresses and heats up, transferring heat to the plates. Then, as the gas expands and cools down, it absorbs heat from the plates. This sets up a temperature gradient within the plates, effectively pumping heat from the cool side to the hot side of the core. Attach heat exchangers to the thermoacoustic core, and this device becomes a useful refrigerator. Fluid cooled with the thermoacoustic refrigerator can cycle over microelectronic components, absorbing their heat, and then return to the heat exchanger to cool down and repeat the cycle. Prof. Herman and former graduate student Martin Wetzel (currently with BMW Research in Munich, Germany) are well-known for their groundbreaking experimental and theoretical work in thermoacoustic refrigeration. The Heat Transfer Lab in the basement of Latrobe Hall houses a working thermoacoustic device, and Prof. Herman and her students study the heat transfer using a variety of techniques, including holographic interferometry and digital image processing.

Building an efficient device that can tackle a specific cooling load involves applying theory carefully to the scale of the problem at hand. The cooling that a MEMS-scale thermoacoustic refrigerator can do is a problem involving many parameters, such as the thermal properties of the fluid being used for the cooling, the material of the plates in the stack, the length of the stack, and the length of the tube. Optimization of the system involves combining groups of these parameters with a specific outcome in mind. Interestingly, Prof. Herman found that sets of parameters leading to two seemingly similar outcomes—maximum efficiency and maximum cooling—were not the same. In addition, she, former graduate student Martin Wetzel, and post-doctoral researcher Zdenek Travnicek found a novel way to collapse the number of parameters needed to perform the optimization, streamlining the approach significantly. Instead of 20 parameters, she now works with a manageable six. Finding the optimal set of design parameters for a MEMS-scale thermoacoustic refrigerator is therefore much more straighforward, and Prof. Herman predicts that commercially viable models won’t be far behind.