The Nanoscale Structure of Charge Order in Cuprate Superconductor Bi2201
Citation
Webb, Tatiana A. 2019. The Nanoscale Structure of Charge Order in Cuprate Superconductor Bi2201. Doctoral dissertation, Harvard University, Graduate School of Arts & Sciences.Abstract
In 1986, high-temperature superconductivity was first discovered [J. G. Bednorz and K. A. Muller, Zeitschrift fur Phys. B Condens. Matter 64, 189 (1986)] in layered cuprate materials based on CuO2 planes. The subsequent decades uncovered a complex phenomenology including a variety of additional unconventional electronic behaviors arising from strong electron-electron correlations. Superconductivity occurs near the meeting point of a regime characterized by mysteriously open "Fermi arcs” in place of a conventional closed Fermi surface and a metallic phase at high hole doping. Describing the doping evolution of the ground state is crucial to uncovering the mechanisms underlying the complex phase diagram. Here, we use the structure of broken translational symmetry, namely d-form factor charge modulations in (Bi,Pb)2(Sr,La)2CuO6+d, as a probe of the reorganization marked by the transition from Fermi arcs to a large Fermi surface. From real-space imaging in a scanning tunneling microscope, we disentangle the spatial inhomogeneity of the electronic structure to obtain the doping dependence of both the Fermi surface and the charge modulation structure. We discover a commensurate to incommensurate transition accompanies the Fermi surface transition, reflecting the distinct nature of electronic correlations governing the two sides of this quantum phase transition.Charge order in cuprates is one example of a physical relationship encoded in the spatial structure of inhomogeneous properties. Today’s imaging technology enables collection of large spatially resolved data sets, but disentangling the complex structure of correlations among properties is challenging in the presence of strong disorder. Machine learning has revolutionized image analysis with its abilities to model large multidimensional data sets and to detect features despite strong variability. We explore the use of convolutional neural networks to uncover the relationships among multiple spatially resolved properties by training networks to classify the structure of disordered d-form factor modulations according to an underlying doping-dependence of the wave vector.
Terms of Use
This article is made available under the terms and conditions applicable to Other Posted Material, as set forth at http://nrs.harvard.edu/urn-3:HUL.InstRepos:dash.current.terms-of-use#LAACitable link to this page
http://nrs.harvard.edu/urn-3:HUL.InstRepos:42029523
Collections
- FAS Theses and Dissertations [6136]
Contact administrator regarding this item (to report mistakes or request changes)