In a study published in Nature Physics, researchers identified the existence of Bragg glass behavior in a charge-density-wave material using machine learning analysis of X-ray data.
(a) Crystal structure of PdₓErTe₃. (b) Machine learning clusters showing two CDW transitions at 135K & 260K (inset: CDW Q-vectors). (c) Phase diagram showing the extent of Bragg Glass correlations.
Scientific Achievement
A machine-learning analysis of charge-density-wave (CDW) peaks in PdxErTe3 has identified Bragg glass behavior due to pinning disorder.
Significance and Impact
By analyzing the linewidths of hundreds of peaks, Bragg glass correlations have been confirmed for the first time in a CDW system.
Research Details
- Bragg glasses, whose structural correlations diverge algebraically, are predicted in strongly pinned CDWs but have not been conclusively observed.
- ErTe3 is an almost tetragonal system with two separate incommensurate CDW transitions at ~135K and 260K, which are suppressed by palladium-doping.
- X-ray data spanning 20,000 Brillouin zones was collected, and machine learning (ML) was used to identify Q-vectors associated with the two CDW phases across the entire phase diagram of PdxErTe3, 0 ≤ x ≤ 2.9%.
- ML clusters associated with each CDW phase were used to determine the Q-dependence of hundreds of peak linewidths to distinguish Bragg glass behavior from vestigial nematic order.