The Heat is on

The Heat is on

One of those persistent problems that has never fully been overcome in electronics is thermal management. At the moment the best we can do is try design more efficient products that give out less heat in combination with trying dissipate any heat generated effectively. These methods can range from a simple vent in a casing that allows the free flow of air, to heatsinks that draw the heat away from the components and let the airflow do its job. Often for more demanding roles a fan will be used to force air over a heatsink and out through the vents. If that combination isn’t enough, the final resort is often a more complex, expensive and bulky solution, such as using liquids to cool the component. Those methods don’t solve the problem, but they take away the worst of the potential effects. These measures have to be taken, as overheating will shorten the life of any components and stop them from acting as their specifications suggest. It often takes many man/hours of simulation and design to get circuits working as intended, especially if the design has to be as small as possible – putting components closer together compounds the heat and makes the situation worse. Every solution either adds cost, complexity or size to the final solution, and on occasion all three.


One team of scientists is now looking at the fundamental physics of components heating and trying to develop a method of lessening its effects by managing the flow of the heat. The physicists at CU Boulder started on the problem after observing a phenomenon that some ultra-small heat sources cool down faster if you pack them closer together. That phenomenon was first observed in 2015 by a team of scientists led by Margaret Murnane and Henry Kapteyn at JILA, a joint research institute between CU Boulder and the National Institute of Standards and Technology (NIST). The physicists were experimenting with bars of metal that were many times thinner than the width of a human hair on a silicon base. When the bars were heated with a laser, they behaved very counter-intuitively. Nano-scale heat sources do not normally dissipate heat efficiently. However, when they were packed together closely, they cooled down quicker than would be expected.


The research team investigated the phenomenon and now know why it occurs. They have now published the results of their research in the journal Proceedings of the National Academy of Sciences (PNAS). In the future, those results could help us design electronic devices that do not overheat as much. The research used computer-based simulations to track the passage of heat from the nano-sized bars. When heat sources were positioned close together, the vibrations from the energy they produced started to bounce off each other, which scattered the heat away and cooled the bars down.


Murnane and Kapteyn and their team partnered with a group of theorists led by Mahmoud Hussein, professor in the Ann and H.J. Smead Department of Aerospace Engineering Sciences. The latter group specializes in simulating, or modeling, the motion of phonons.


The researchers recreated their original experiment, but this time, entirely on a computer. The simulations were so detailed that the team could follow the behaviour of each and every atom in the model from start to finish. The researchers found that when they spaced their silicon bars far enough apart, heat tended to escape away from those materials in a predictable way. The energy leaked from the bars and into the material below them, dissipating in every direction.


When the bars got closer together, as the heat from those sources scattered, it effectively forced that energy to flow more intensely away from the sources. This discovery could allow the heat generated in electronics to be channeled energy along a desired path.


CU Boulder co-authors on the new research include Hossein Honarvar, a postdoctoral researcher in aerospace engineering sciences and JILA, Brendan McBennett, a graduate student at JILA and Joshua Knobloch, postdoctoral research associate at JILA. Former JILA researchers Travis Frazer, Begoña Abad and Jorge Hernandez-Charpak also contributed to the study.