Rethinking Phonons

The current understanding of phonons treats them as plane waves/quasi-particles of atomic vibration that propagate and scatter. This concept of treating phonon transport like that of a gas is termed the phonon gas model (PGM), however, it only seems to rigorously hold true in the limit of a pure, infinite, perfect crystal. The problem is then that whenever any level of disorder is introduced or symmetry is broken, the character of vibrational modes changes, yet nearly all theoretical treatments continue to assume phonons are still waves. For example, the phonon contributions to alloy thermal conductivity rely on the PGM and are most often computed from the virtual crystal approximation. Good agreement is obtained in some cases, but there are many instances where it fails —both quantitatively and qualitatively. Similarly, for amorphous materials the PGM fails at describing the contributions to thermal conductivity, and new questions surrounding its validity at interfaces are also emerging – even when it is an interface between two crystalline materials. This talk will review a growing body of recent work that suggests that our understanding of phonons requires revision, because the critical assumption that all phonons/normal modes resemble plane waves with well-defined velocities is no longer valid when some form of disorder/symmetry breaking occurs. An emphasis will be placed on the new methodologies that have been developed to treat phonon transport more rigorously and generally. Additionally, some of the rather non-intuitive results/predictions that have emerged from applying these new approaches will be highlighted in an effort to identify some of the interesting questions that must be addressed in future work.