In a real lab on earth, IBM scientists Bernd Gotsmann and Mark Lantz have actually done experiments on heat transport as part of IBM’s overall interest in developing energy efficient 3D chip stacks. Roughness plays an important role in understanding how heat is transported at an atomic scale across interfaces, like between stacked computer chips. And their work on heat transport across nano-scale interfaces was recently published in "Nature Materials."
|Dr. Bernd Gotsmann|
Q. What is a good example to explain the science behind tribology?
BG: While Spiderman isn’t real, I think everyone can appreciate the following analogy. When Spiderman climbs on the side of a building his finger tips come into contact with the bricks. Adhesion forces acting on such contact areas of say a square centimeter are strong enough, in principle, to hold Spiderman. But in reality things do not stick so easily due to irregularities and roughness of the surfaces. In fact, you may find that by looking at his fingertips only a square millimeter will actually be in contact with the surface.
Now imagine looking at this same square millimeter with a microscope. You will find that the roughness will prevent most of the smaller touch points to make contact. Now zoom in even further and again, there are even fewer contact points.
So now the question is whether the notion that roughness governs the real contact between surfaces can reasonably be scaled down to the atomic level.
|Dr. Mark Lantz|
ML: One of the critical challenges for scientists in developing 3D chip stacks is heat dissipation. IBM is currently developing chips that use water cooling on the back of the chip to keep them at operating temperatures. But chips are made of silicon, which at an atomic scale is rough. And when several chips are stacked together the dissipation of heat across the entire surface of the chip is strongly influenced by the roughness of the interfaces between the chips.
Q. What findings did you report in your paper?
BG: Through a series of experiments we measure how heat flows across contacts that are extremely smooth, having roughness only on the nanometer scale and below.
We previously could not explain this data using conventional understanding. We then asked ourselves, what would happen if atomic scale roughness forces the contact to be governed by individual atoms? In this case, thermal transport is easy to describe because each instantaneous atom-atom contact carries a certain amount of heat, called quantized conductance. We found that this explains the data very well.
|Atomic roughness of two surfaces|
BG: Well, despite the concurrence between the data and the explanation, we unfortunately cannot look directly into such contacts and see which atoms make contact and which don't. We therefore hope that other scientists also apply the same notion and gather further support for our hypothesis.
While the questions raised fit into a current debate in the nanotribology community, the impact on the understanding of thermal transport could be just as large. Our work implies that thermal contacts are much easier to describe than previously thought. I am eager to learn whether our understanding holds for only few a few special laboratory cases or if it is relevant for more general and technologically relevant systems.
If you are familiar with IBM Fellow Benoit Mandelbrot’s work in fractal geometry then you already have a good basis for tribology. Fractals are used in modeling uneven or rough structures in Nature, like an eroded coastline, in which similar patterns recur at progressively smaller scales. Fractals use math to explore these rough irregularities.
ML: A very exciting extension of this work would be to perform similar heat transport measurement on sliding surfaces. Our research, to date, has only looked into static contacts formed by pressing two surfaces together, in order to investigate how the atomic scale roughness of the surfaces effects heat transport between them.
In contrast, tribology encompasses the study of interacting surfaces in relative motion. In this dynamic case, the roughness of the two surfaces is thought to play a critical role in the resulting friction and wear. However, the details are not well understood, especially at the atomic scale. At this level, it is difficult to look into the contact to understand what parts of the two surfaces interact and how this changes with time.
Our work on static contacts has shown that heat transport can be used as a kind of fingerprint to study the nature of the contact on the atomic scale. Therefore, by measuring friction and heat transport simultaneously, we have a means to see into the contact. We could then study the nature of the atomic scale interaction between the two surfaces as they move, and then relate this to the measured friction to gain insight into the basic physical mechanisms that cause friction and wear.
"Spiderman" fans should take note and follow this research.