Testing the response of high-temperature superconductors to the presence of spins: Nickel, zinc, and gallium substitutions in yttrium barium copper oxide
Abstract
Since the discovery in 1986 of high-temperature superconductors, the physical mechanism of superconductivity in these materials has remained a mystery. One approach to determining the validity of the conventional BCS theory has been to dope these copper oxides with magnetic ions and look for the characteristic drop in transition temperature. Results have been inconclusive thus far. We describe in this thesis a systematic program to test for a BCS-like response to the presence of spins by varying the ratio of two dopants which substitute for copper, one magnetic and the other nonmagnetic. Electrical resistance measurements of $T\sb{c}$ have been made for the doped high-temperature superconductor $YBa\sb2\lbrack Cu\sb{0.95}(\lbrack Ni,Zn\rbrack\sb{1-x}Ga\sb{x})\sb{0.05}\rbrack \sb3O\sb{7-\delta}$, with 0 $\leq\ x \leq\ 1$, to determine the effect of Ni spins on $T\sb{c}$. Using the co-doped scheme of magnetic Ni with nonmagnetic Ga, and also nonmagnetic Zn with Ga, it was possible to isolate the effect of the presence of spins from those effects due to changes in the number of charge carriers. Ni and Zn have a valence of +2 while Ga has a valence of +3, so it was possible to vary the number of available charge carriers by adjusting the ratio of Ni:Ga or Zn:Ga. It has been determined through anomalous dispersive x-ray diffraction on single crystals that Ni resides in the Cu(2)-planes, while it is assumed, based on other dopant site occupancy studies, that Zn also resides in the Cu(2)-planes and that Ga resides in the Cu(1)-chains. The Zn/Ga co-doped samples were used as a control because they do not contain spin but do not have dopant site occupancy and charge carrier effects similar to the Ni/Ga co-doped samples. Results are consistent with the presence of competing effects from the Ni spins and changes in the number of available charge carriers. Analysis indicates that the spins of Ni could indeed by playing a significant part in the reduction of $T\sb{c}$. It was estimated that the change of $T\sb{c}$ due to spins was $-$10K at a Ni concentration of 2%, while non-charge balanced Ni reduces $T\sb{c}$ by only 8K. That is, Ni is twice as effective as poisoning superconductivity when the charge imbalance is corrected. This study is the first direct evidence of a BCS-like response to magnetic dopants in the $Cu - O$ planes of a high-temperature superconductor.
Degree
Ph.D.
Advisors
Durbin, Purdue University.
Subject Area
Condensation
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.