Have you ever played with a slinky? If you have never seen one, please click here: http://www.jimrigby.org/wp-content/uploads/2012/03/slinky.jpg . To see a video of how slinky stretches, click here : http://www.youtube.com/watch?v=vwjoUmJ4J1w . Why is the slinky behaving that way? Slinky is essentially a spring with a very low spring constant; which means that when you stretch a slinky, the restorative force generated by the slinky is'nt much. Hence slinky is essentially a "soft spring".

Solids can be thought of as atoms packed together, which are attached to one another using "springs". The reason why atoms behave as if they are attached via "springs" is because they are bonded chemically to one another. Under some circumstances, solids can behave like slinkies (at least in some specific directions). As you might imagine, such "soft modes" can result in large scale displacement of atoms, which in some cases may result in an entirely new arrangement of atoms.

Why am I saying this? Power diodes are electronic devices that carry a few amperes per square cm. This is a large current density. Wide band gap material called silicon carbide (SiC) is a popular choice for making power diodes. However these diodes were seen to fail faster than was initially predicted, especially considering their thermal and chemical stability. What people observed was that the electronic properties of these diodes were getting modified over time. Think of it as some "aging" occuring in the device. It was soon learnt that it was because of stacking faults (a mislignment in the arrangement of the atoms) being created at the junction of the device. This observed phenomenon did not have an explanation, which is why it piqued our interest. Along with Prof. Umesh Waghmare (JNCASR), I demonstrated that there are "slinky like" modes (formally called "soft modes"), which are responsible for stacking fault expansion in these systems. The soft mode is also responsible for making the faulted structure thermodynamically more stable than the perfect SiC crystal. This is why SiC power diodes failed faster than was originally predicted. For a more technical account on this work, please do consider reading our paper.

Ref: Physical Review B 77, 121203(R) (2008)