PULLMAN – WSU Professor Matt McCluskey has discovered a new type of atomic oscillation that could impact solid-state phenomena ranging from diffusion to electronic device performance. The result was published in the April issue of the journal Physical Review Letters, http://link.aps.org/abstract/PRL/v102/e135502.
 
The practical implications of the phenomenon McCluskey has discovered include better understanding of matters like how heat is dissipated from computer chips. 
 
“This is an example of the long-term work needed to address fundamental questions in the physical sciences. The work – like a great deal of science research – is a tale that builds on results made clear by previous discoveries and theories,” said McCluskey.
 
When McCluskey was a graduate student at UC Berkeley and the Lawrence Berkeley National Laboratory (LBNL) 10 years ago, he studied the interaction of hydrogen with impurities in semiconductors.
 
“Hydrogen, the most abundant element in the universe, is a common contaminant in semiconductor growth and processing. It is therefore highly important to understand what it does when it finds its way inside a semiconductor,” he said.
 
In this case, the semiconductor was selenium-doped aluminum antimonide (AlSb:Se). When McCluskey cooked AlSb:Se in hydrogen gas, the hydrogen atoms diffused into the crystal and formed bonds with the aluminum atoms. The aluminum-hydrogen pairs exhibit vibrational motion, like two balls connected by a spring. The stretch mode is a vibration where the spring is stretched and compressed during the vibrations. The wag mode is where a mass moves side to side.
 
Using infrared spectroscopy, McCluskey measured the frequencies of the vibrational modes. He performed the experiment first with hydrogen, then deuterium, which has a lower vibrational frequency than hydrogen.
 
McCluskey noticed that while the aluminum-deuterium pairs had one stretch-mode frequency, the aluminum-hydrogen pairs showed two frequencies. Nothing like this had ever been seen before. No one could explain why the aluminum-hydrogen stretch mode should split in two.
 
At that time, McCluskey and his collaborators speculated that the aluminum-hydrogen stretch mode frequency just happened to equal some combination of lower-frequency modes.
 
At WSU, McCluskey finally had the tools to test this theory. Using the National Science Foundation’s Teragrid system, he performed atomic simulations on machines at the National Center for Supercomputer Applications. After months of numerically intensive calculations, the results came in. Along with the stretch mode and wag mode, the aluminum-hydrogen pair also had a transverse mode, where the aluminum and hydrogen atoms oscillate together, as a single unit.
 
The solution to the puzzle popped out. The combination mode was a transverse mode plus two wag modes. The frequencies added together perfectly.
 
Then came the real surprise. Normally, a stretch mode is highly localized. Only the hydrogen atom oscillates. The neighboring atoms barely move. When the stretch mode interacts with the combination mode, though, its spatial extent increases dramatically. Instead of only one atom moving, hundreds do. The accidental resonance completely changes the vibration’s character.
 
Solid-state physicists refer to collective oscillations of atoms as phonons, meaning literally ‘particles of sound.’ McCluskey and coworkers discovered a new quasi-particle that exists somewhere between a phonon and a localized mode. In the future, it is possible that these strange quasi-particles will be found in a range of condensed-matter systems. They could potentially help dissipate energy in devices, like computer chips, that generate heat.
 
“It took an amazing coincidence to produce this quasi-particle excitation,” McCluskey said. “Now the challenge will be to produce them in other solids and put them to good use.”