Phase decomposition and chemical inhomogeneity inNd2−xCexCuO4±

Abstract

Extensive X-ray and neutron scattering experiments and additional transmission electron microscopy results reveal the partial decomposition of Nd2−xCexCuO 4±δ (NCCO) in a low-oxygenfugacity environment such as that typically realized during the annealing process required to create a superconducting state. Unlike a typical situation in which a disordered secondary phase results in diffuse powder scattering, a serendipitous match between the in-plane lattice constant of NCCO and the lattice constant of one of the decomposition products, (Nd,Ce)2O3, causes the secondary phase to form an oriented, quasi-two-dimensional epitaxial structure. Consequently, diffraction peaks from the secondary phase appear at rational positions (H, K, 0) in the reciprocal space of NCCO. Additionally, because of neodymium paramagnetism, the application of a magnetic field increases the low-temperature intensity observed at these positions via neutron scattering. Such effects may mimic the formation of a structural superlattice or the strengthening of antiferromagnetic order of NCCO, but the intrinsic mechanism may be identified through careful and systematic experimentation. For typical reduction conditions, the (Nd,Ce)2O3 volume fraction is ∼ 1%, and the secondary-phase layers exhibit long-range order parallel to the NCCO CuO2 sheets and are 50 − 100 A thick. The presence of the secondary phase should also be taken into account in the analysis of other experiments on NCCO, such as transport measurements.

Microscopic annealing process and its impact on superconductivity in T′-structure electron-doped copper oxides

“…To test this possibility, single crystals of PLCCO (space group I4/mmm, a = 3.98 Å, c = 12.3 Å) were grown using the traveling-solvent floating-zone technique. These included four samples prepared from the same as-grown batch: as-grown nonsuperconducting (ag-NSC), reduced superconducting (r-SC), oxygenated nonsuperconducting (o-NSC), and re-reduced superconducting (r2-SC) crystals.As shown in previous work [18][19][20][21] , annealing necessary to produce superconductivity also induces epitaxial intergrowths of the cubic R 2 O 3 impurity phase.The R 2 O 3 has a cubic structure with space group Ia3 and lattice parameter a c = 2√2a ~ c/1.1, where the "c" subscript indicates "cubic". The Miller indices of the T' and impurity phases are related as ( , , ) [ To test the reversibility of the impurity phase, we carried out x-ray powder diffraction measurements on portions of the ag-NSC, r-SC, and o-NSC single crystal samples of PLCCO that were crushed into fine powders.…”

“…), electron-doping alone is insufficient, and annealing the as-grown sample in a low oxygen environment to remove a tiny amount of oxygen is necessary to induce superconductivity 2,3 . Previous work [4][5][6][7][8][9][10][11][12][13][14][15][16][17][18][19][20][21][22][23] suggests that oxygen reduction may influence mobile carrier concentrations 7 , decrease disorder/impurity scattering 8,10,11,23 , or suppress the long-range AF order 16,17,22 . However, the microscopic process of oxygen reduction, its effect on the large electron-hole phase diagram asymmetry and mechanism of superconductivity 2,3 are still unknown.…”

“…5%) entirely suppress AF order and superconductivity appears over a wide range of hole concentrations (from 6% to 30%). In the case of T' structured electron-doped superconductors, doping alone by substituting the trivalent ions R 3+ in R 2 CuO 4 with tetravalent Ce 4+ is insufficient to induce superconductivity and the as-grown materials such as Nd 2-x Ce x CuO 4±δ (NCCO), Pr 2-x Ce x CuO 4±δ (PCCO), and Pr 0.88 LaCe 0.12 CuO 4 (PLCCO) are semiconducting with static long-range AF order [4][5][6][7][8][9][10][11][12][13][14][15][16][17][18][19][20][21][22] .In these materials, superconductivity (maximum T c ~25 K) is achieved only when the samples within a relatively narrow dopant range (0.1≤x≤0.18) are annealed in a reducing atmosphere . The reduction process suppresses the AF order most severely at higher Ce-doping levels above x≥0.1 (ref.…”

“…31) have recently determined the Fermi surface topology and the superconducting gap function in electron-doped materials. However, the quality of the sample surfaces studied has not been conclusively established, and one must now consider the possible effect of an exposed NSC R 2 O 3 surface in future ARPES and scanning tunneling microscopy experiments.Furthermore, it should also be possible to grow higher-quality single-crystal T' structured copper oxides by adding excess Cu in the starting materials to eliminate Cu deficiencies and subsequent R 2 O 3 contamination.Because of the availability of single crystals of the T' structured copper oxides, the vast majority of transport, structural, magnetic, and electronic-property measurements of electron-doped superconductors have been obtained from these materials [9][10][11][12][13][14][15][16][17][18][19][20][21][22]25,[27][28][29][30][31] . For other electron-doped high-T c superconductors with different crystal structure types, such as the infinite-layer Sr 1-x La x CuO 2 family 32-34 , the microscopic process of inducing superconductivity is still unclear 33,34 .…”

Neutron Scattering Studies of Antiferromagnetic Correlations in Cuprates

“…An initial study [257] on Nd 2−x Ce x CuO 4 with x = 0.14 and T c ∼ 25 K found that applying a field as large as 10 T had no effect on the intensity of an antiferromagnetic Bragg peak for temperatures down to 15 K. Shortly after that came a report [258] of large field-induced enhancements of antiferromagnetic Bragg intensities, as well as new field-induced peaks of the type ( 1 2 , 0, 0), in a crystal of Nd 2−x Ce x CuO 4 with x = 0.15 and T c = 25 K. It was soon pointed out that the new ( 1 2 , 0, 0) peaks, as well as most of the effects at antiferromagnetic reflections, could be explained by the magnetic response of the (Nd,Ce) 2 O 3 impurity phase [251,252]. There now seems to be a consensus that this is the proper explanation [259,260]; however, a modest field-induced intensity enhancement has been seen at ( 1 2 , 1 2 , 3) that is not explained by the impurity-phase model [259].…”