Models and Experimental method

5.2.1 Models and Epitaxial design of the interfaces

Both YBCO and LCMO have ABO₃ perovskite related structures. Two possible atomic stacking sequences can be formed along the (001)-oriented heterostuctures: a) La_0.7Ca_0.3O–MnO₂–BaO–CuO2 (MnO₂-terminated interface) and b) MnO₂–La0.7Ca_0.3O–CuO₂–BaO (La0.7Ca_0.3O–terminated interface). Realizing this interface design requires atomically precise interface control which can be achieved by designing the LCMO layers with well-defined atomic terminations using reflection high-energy electron diffraction (RHEED) assisted pulsed laser deposition. The clear intensity of the oscillations indicated a layer–by–layer growth mode with the unit cell precision during the growth of the LCMO, YBCO and SrRuO3 (SRO) layers. The SRO layer was inserted to switch termination of the LCMO layer [42]. Two distinct interfaces can be fabricated based on the control of the LCMO termination layer. The details of the heterostructure growth can be found in Figure 5.2a) and 5.2b). The schematics of the MnO2–terminated and La0.7Ca0.3O–terminated interfaces (i.e., the interfacial control of a heterostructure built with two perovskites stacked along the (001) direction) are shown. We deposited YBCOd/LCMO on different interfaces with a constant LCMO layer thickness of n=25 unit cells (u.c.) (corresponding to 10nm). The thickness d of YBCO layer, however, varied from 2nm to 100nm. In this chapter of thesis,

“MnO2-terminated” corresponds to the STO/LCMO10nm/YBCOd structure (black in the online data) while “La_0.7Ca_0.3O-terminated” corresponds to the following STO/SRO_1.5u.c./LCMO_10nm/YBCO_d structure (red in the online data).

Figure 5.2 Epitaxial design of heterointerfaces: Schematic of the interfacial control of LCMO/YBCO_d with different interfaces; a) in the MnO₂-terminated interface (La0.7Ca0.3O–MnO2–BaO–CuO2) the charges are very difficult to transfer because the CuO chain is very far from the interface (indicated by a dashed line) while b) switches into the La0.7Ca0.3O-terminated (MnO2–La0.7Ca0.3O–CuO2–BaO) interface by using SRO; electrons transfer easily from LCMO to YBCO because of the CuO chain at the interface (indicated by solid lines).

5.2.2 Experimental details 5.2.2.1 Sample preparation

YBa₂Cu₃O_7-x/La_0.7Ca_0.3MnO₃ (YBCO/LCMO) hetrostructures were prepared on 5x5 mm²(100)-oriented SrTiO₃ (STO) single crystal. We used in-situ reflection high-energy electron diffraction (RHEED) to monitor layer growth. The LCMO and YBCO layers were deposited at respective growth temperatures of 700^oC and 750^oC, and oxygen pressures of 80 mTorr and 150 mTorr. To switch the (La,Ca-O) termination at the interface, a buffer layer of SRO (1.5u.c.) was deposited between the substrate and the LCMO layer. For MnO2-terminated interface, we used uniform single termination of TiO2. The TiO2 terminated STO (100) surfaces were obtained by chemical treatment with an HF-NH4F buffer solution.

The growth processes and the switching of terminations at different interfaces of YBCO/LCMO are demonstrated on Figure 5.3 Inset of Figure 5.3 a) and b) (on top) show the in-situ RHEED patterns and TiO2 terminated surface of STO (100).

105

Figure 5.3 Interface control “Growth model”: Layer–by –layer growth is monitored by RHEED, a) the MnO₂-terminated interface corresponds to the STO/LCMO_10nm/YBCO_d structure while b) the La_0.7Ca_0.3O-terminated interface has the following STO/SRO1nm/LCMO10nm/YBCOd structure.

Following layer deposition, full oxygenation was achieved by annealing the film at 550^oC in an oxygen atmosphere of 700Torr for an hour followed by slow cooling to room temperature.

5.2.2.2 XAS and XMCD

XAS and XMCD are the most appropriate techniques by using an extremely sensitive local probe to study the valence and spin characters as well as the orbital contribution to the magnetic moment. The XAS and XMCD spectra of the Mn-L2,3 edge and the XAS of the O K-edge were recorded using the Dragon and 20A beamlines of National Synchrotron Radiation Research Center (NSRRC) in Taiwan with respective energy resolutions of 0.2 eV, and 0.3 eV. The sharp peaks at 640.1eV of the Mn-L3 edge of single crystalline MnO, and at 934.7 eV and 531 eV of the Cu-L3 edge and the O K-edge of single crystalline Cu2O were measured simultaneously in a separate chamber for energy calibration, which enabled us to achieve accuracy better than 0.05 eV for relative energy alignment. Both the XMCD spectra

of the Mn-L_2,3 edge were measured under a magnetic field of 1T at the temperature 30K with approximately 80% circularly polarized light. The magnetic field direction makes an angle of 30° with respect to the Poynting vector of the soft x-rays. The spectra were recorded using the total electron yield (TEY) method (by measuring the sample drain current) under an ultrahigh chamber (UHV) with a base pressure of 1x10⁻⁹ mbar.

The polarized O K-edge XAS spectra were carried out by the synchrotron linear polarized light with E//ab in normal incidence on the sample and the signals were detected in the total x-ray fluorescent yield (FY) mode and sample current mode. The probing depth of FY detection is about in the order of 200 nm and several nano meters for FY and sample current modes, respectively. The base pressure in the UHV chamber was about 10^-9Torr. The resolution of the spectra was controlled by the spherical grating monochromator and was estimated to be better than 0.22eV. As a routine procedure, following pre-edge background subtraction, the spectra were normalized using the incident beam intensity, keeping the energy range between 580-620 eV for the O K-edge spectra.

Mn K-edge XAS spectra were recorded in a fluorescence mode with a Lyttle detector at BL17c1 beamline of NSRRC. A double Si(111)-crystal monochromator was used for energy selection with a resolution ΔE/E better than 2 x 10^-4. Higher harmonics were eliminated by detuning the double crystal Si(111) monochromator. X-ray energy was calibrated by the known Mn K-edge absorption of Mn foil.

5.3 Results and Discussions