For the last three decades, physicists have been baffled by what happens to uranium ruthenium silicide (URu2Si2) at 17.5 kelvin (minus 256 degrees Celsius). By measuring heat capability and different traits, they’ll tell it undergoes some sort of phase transition; however, that is up to anybody can say with certainty.
A Cornell partnership headed by physicist Brad Ramshaw, the Dick & Dale Reis Johnson Assistant Prof. in the College of Arts and Sciences, employed a combination of ultrasound and machine learning to narrow the possible explanations for what happens to this quantum material when it enters the hidden order.
Their paper titled “One-Component Order Parameter in URu2Si2 Uncovered by Resonant Ultrasound Spectroscopy and Machine Learning” was featured in Science Advances on March 6.
Ramshaw and his doctoral pupil Sayak Ghosh used high-resolution ultrasound spectroscopy to review the symmetry properties of a single crystal of URu2Si2 and how these properties change throughout the hidden order phase transition.
The outcomes from the machine-learning algorithm eradicated nearly half of the more than 20 likely explanations for the hidden order.
It might not yet offer a solution to the URu2Si2 query; however, it has developed a new strategy for tackling data evaluation problems in experimental physics.