1. Introduccion

Perovskite manganites compounds with mixed manganese valence, La_1-x A_x MnO₃ where A for a divalent metal, have attracted a great deal of attention due to anomalously large negative “colossal” magnetoresistance (CMR) for 0.2<x<0.5 ^1-3. Understanding the behavior of these Mn-based materials is a challenge for experimentalist and theorist. It is now accepted that both double exchange (DE) ⁴ and Jahn-Teller (J-T) lattice distortions ⁵ play an important role in the magnetic and transport properties of the manganites. Other effects such as magnetic polarons, electron localization and phase separation of carriers have been suggested as further subjects of research^6,7. These effects show that spin, charge and lattice are strongly coupled in the CMR materials⁸. Although most of the attention has been focused on the effect of applied magnetic field and temperature on the conductivity, there is much to be learned from the properties (such as photoluminescence) determined by the electronic structure in La_2/3Ca_1/3MnO₃ (LCMO)⁸ compound. Photoluminescence (PL) in solid materials is always resulting from radiative decay of electronic transitions. In semiconductors, the PL comes from band to band transition or excitonic states, while in insulators it results from impurities, as transition metal or rare earth ions. Some few cases that are not exactly explained by these two mechanisms have been reported, as porous silicon or nanometric particles⁶. In this paper, we present the first report on the PL features and their temperature dependence of Ca_0.90Sr_0.10RuO₃ (CSRO) compound, which provides another case that cannot be explained by the mechanisms mentioned above. The results indicate than the unconventional features for the electronic structure and their variation with temperature, which arise from the extremely strong coupling among the conduction electrons, local spins and lattice.

2. Materials and Methods

The polycrystalline sample of the CSRO compound was synthesized by solid-state reaction method at ambient pressure. Starting materials were RuO₂ anhydrous (99.9% STREM), SrCO₃ (99.5% CERAC) and CaCO₃ (99.99% BAKER). Prior to weighing, SrCO₃ and CaCO₃ were preheated during 10-20 minutes at 120°C in order to dehydrate them. Stoichiometric mixture of these starting materials was done in air during 15 minutes using an agate mortar, resulting homogenous slurry. The produced milled polycrystals were then annealed in air during two days at a temperature between 700 and 800°C in a Thermolyne 46100 furnace (±4°C) with the intention of decompose the carbonates. The resultant CSRO nano-crystals were compressed into pellets (13 mm diameter, 1.0-1.5 ± 0.05 mm thickness), by applying a pressure of 1/4 ton/cm² for 15 minutes in vacuum. The resulting compacted specimens were then sintered in air at 800°C during four days. All reagents and final products were characterized by X-ray powder diffraction (XRD) with a Bruker-AXS D8-Advance diffractometer using CuK₍ radiation (λ=1.5406 Å) and a graphite monocromator. Diffraction patterns were collected at room temperature over a range 5∘≤2θ≤70∘ with a step size of 0.017° and a time per step of 4 s. The change in morphology grain size in CSRO compound obtained by different heat treatments, was observed by scanning electron microscopy (SEM) on a JOEL JSM- 6610LV. The micrographs at 50.00 K.X, were taken with a voltage of 20 kV, current intensity of 1000 pA and WD = 10 mm. The Energy Dispersive X-Ray (EDX) was performed on the same equipment equipped with an Oxford/Link System electron probe microanalyzer (EPMA). Low temperature DC resistance measurements were performed using the standard four-probe method from room temperature down to 10 K. The magnetization was obtained on a MODEL-P525: PPMS vibrating sample magnetometer, Quantum Design, 16 T. To measure the zero field-cooled (ZFC) and the field-cooled (FC) magnetization, the samples were cooled down to 2 K at zero field and 100 Oe, respectively⁹. Finally, the PL spectra of the (Ca_1-x Sr_x)RuO₃ pellets were recorded by mounting them on a cold finger of a Displex system and keeping the temperature range 6.7-296 K. The spectra were analyzed thorough an automated, HORIBA JOBIN YVON IHR320 monochromator and CCD detector. The samples were excited using Ar+ laser with power of 0.73 mW and wavelength of λ=476.5.

3. Results and Discussion

Figure 1 shows the XRD patterns of the (Ca_1-x Sr_x)RuO₃ system with x = 0.07, 0.10, and 0.15^10-12. From those spectra, compounds show a solubility up to x = 0.15. All samples crystallize in a single orthorhombic phase identified as CaRuO₃ PDF (70-2790) and form a continuous solid solution. With respect to the examined diffractograms, it is worth to mention than the compound with x = 0.10 (Ca_0.90Sr_0.10RuO₃) present reflections of a secondary phase identified as RuO₂¹³ and the lattice parameter a = 5.524 Å, b = 7.843 Åand c = 5.432 Å. The solid line corresponds to an orthorhombic phase and it is identified as CaRuO₃ compound with PDF (70-2790). The phase marked with an asterisk (*) corresponds to the RuO₂ compound, PDF (43-1027). The presence of this very small amount of RuO₂ compound was detected in the (Ca_1-x Sr_x)RuO₃ 0.07≤x≤0.15 system then indicated than occur an overload of the reagent in all the system. This is presumably a consequence of loss of volatile ruthenium oxide¹³. This solubility is a result of the anions, which contributed to the formation of the mechanism of the solid solution in this (Ca_0.90Sr_0.10RuO₃) compound. The reason why we selected this (Ca_0.90Sr_0.10RuO₃) compound is that it lies in the formation region of the solid solution.

Figure 1 XRD patterns of (Ca_1-x Sr_x)RuO₃ system with x = 0.07, 0.10, and 0.15. The solid line corresponds to an orthorhom-bic phase and it is identified as CaRuO₃ compound. The phase marked with an asterisk (*) corresponds to the RuO₂ compound. 

The next step was the characterization of the Ca_0.90Sr_0.10RuO₃ compound achieved by SEM, to observe the morphology and grain sizes of the crystals. The micrograph shown in Fig. 2 was taken on the surface of the Ca_0.90Sr_0.10RuO₃ compound with a magnification of the 50.00 K. X. Also, in some regions we observe semi-fusion than can be attributed to the ruthenium content. We can observe the secondary phase in the other gray color and the grain size varies between 0.592 and 1.598 µm. This coincide with the XRD data, than we obtain diverse components in the diffractograms, this generally happens, because SEM technique is more sensitive that XRD.

Figure 2 SEM Image of Ca_0.90Sr_0.10RuO₃ showing a particle size varies between 0.422 and 1.598 µm. 

The electrical resistance as a function of temperature is present in Fig. 3. The Ca_0.90Sr_0.10RuO₃ compound show a metallic behavior and short-range ferromagnetic interactions appear. This indicates that the ferromagnetism has been suppressed through the process of substitution of Sr²⁺ ions by Ca²⁺ ions^10-14. For the compounds with large Ca²⁺ ions doping (x≥0.7), no clear phase transition is discerned, and only some irreversibility is observed in the magnetization curves of these materials. The disappearance of the long-range magnetic order is commonly related to the distortion of the RuO₆ octahedra associated with the partial or total replacement of Sr²⁺ ions by Ca²⁺ ions, and the corresponding narrowing of the 4d bandwidth¹⁵.

Figure 3 Electrical resistance as a function of temperature for the Ca_0.90Sr_0.10RuO₃ compound. 

Figure 4 shows the PL spectra for the (Ca_1-x Sr_x) RuO₃ system with x = 0.0, 0.07, 0.10, 0.15 and 1.0 measured at room temperature. In the measurements performed at low temperature on the (Ca_1-x Sr_x)RuO₃ system with x = 0.0, 0.07, 0.10, 0.15 and 1.0, it was observed a change in the intensity peaks due to temperature effects, although there exists small changes in the peaks position due to effects of the Sr²⁺ ions incorporation. The most notable observation in than the signal of PL is very intense when we have only CaRuO₃, but the gradual increase of the signal is due to Sr²⁺ ions incorporation in the system, than indicates larger density of states near Fermi level EF and smaller lattice distortion¹⁴. Therefore, the intensity is related to the peaks of the 4d band than should be even narrower in CaRuO₃ compound than in SrRuO₃ compound because a Ca-O bond has a more covalent character than a Sr-O bond. This argument is consistent with the photoemission studies, which reveal that the free-electron component in CaRuO₃ compound is weaker than in SrRuO₃ compound and hence the 4d band is narrower in CaRuO₃ compound^15-16. Such changes in interactions are observed through this change of chemical composition. Due to the other compounds of the (Ca_1-x Sr_x)RuO₃0.07≤x≤0.15 system do not show any change in the intensity of the PL spectrum in the temperature range 297 to 6.7 K. The main interest to study the photoluminescence (Ca_0.90Sr_0.10RuO₃) compound is due to than the PL spectrum presents a gradual increase in intensity in the temperature range 297-6.7 K under the same experimental conditions.

Figure 4 PL spectral on (Ca_1-x Sr_x) RuO₃ system with x = 0.0, 0.07, 0.10, 0.15 and 1.0 measured at room temperature. 

In the Fig. 5 displays the PL spectra (1.3-2.5 eV) of the Ca_0.90Sr_0.10RuO₃ compound in the range 6.7-296 K, for which an excitation laser wavelength of the 476.5 nm was used. The main peak at ∼1.73 eV, along with the shoulders at ∼1.38 eV and ∼2.05 eV, respectively, is observed through the whole temperature range. As to the PL bands with the photon energies of 2.2 and 2.5 eV in our studied CSRO compound, W.L. Zhu et al suggest that they do not originate from the oxygen vacancies or other defect states, which should be sample-dependent¹⁹. T. Ding et al suggest that the PL bands around 2.2 and 2.5 eV arise perhaps from the interband transition between the O 2p and Ru 4d bands²⁰.

Figure 5 PL spectra of Ca_0.90Sr_0.10RuO₃ compound at different temperatures from 6.7 to 296 K. Labels indicate the position of the peaks for the fitting. Excitation: λ = 476:5 nm. 

In the following, we only concentrate on the temperature dependence of PL spectra in the photo-energy range 1.3-2.2 eV, ascribed to the Ru 4d band²¹. We use the multi-peaked Gaussian fitting to the mixed band (PL); the results are perfectly adjusted and determine the position of each peak, which is 1.38 eV for B₁ , 1.73 eV for B₂ , and 2.05 eV for B₃ (Fig. 6), respectively, which represent three different types of electronic transitions.

Figure 6 Experimental (open circles) and fitted (solid line) data of PL spectrum at 40 K. The best fit is given by three Gaussian curves B₁ , B₂ and B₃ (dashed lines) centered at 1.38, 1.73, and 2.05 eV, respectively. 

The temperature dependence of the intensity for each peak is clearly shown in Fig. 7. In the entire temperature range 6.7-296 K, the intensity of B₁ and B₂ increased with decreasing temperature. However, the intensity of B₃ is almost temperature-independent. In order to reveal the mechanisms of PL CSRO, its related electronic structure should be understood.

Figure 7 The temperature dependence of the PL intensities for each peak B₁ , B₂ and B₃ . 

Figure 8 shows the electronic structure of the Ru³⁺ and Ru⁴⁺ ions and the possible optical transitions. The upper panel shows the energy of Ru e_g and t_2g levels around the ferromagnetic transition region for adjacent Ru³⁺ and Ru⁴⁺ ions. The spin up e_g levels in the Ru³⁺ ions are split by E_TJ due to the J-T effect. Transition A is the dipole active photoionization of the J-T small polaron. The lower panel shows the energy levels in the metallic ferromagnetic state at low temperatures. The spin up and spin down e_g bands are separated by E_TJ. The aligned core spins in the t_2g levels lie below the spin up e_g levels by the crystal field energy ECF. Process B promotes a t_2g core electron of Ru³⁺ to the spin up e_g bands Ru⁴⁺ by a dipole allowed charge transfer process. The transition between the spin up e_g bands and spin down e_g bands depicted by C is allowed only transition involves electrons from the spin up e_g level of Ru³⁺ to the spin down t_2g level of Ru⁴⁺ with spin flip process.

Figure 8 Schematic electronic structure of the e_g and t_2g levels of Ru³⁺ and Ru⁴⁺ ions and the optical transitions relevant to the PL peaks. From ^{Ref. 8}. 

The existence of these three transition processes was supported by the temperature dependence. At lower temperatures, the spins are all aligned, so the effect of the intensity is expected to be maximum, whereas at higher temperatures, where the spins aligned randomly, and the effect of the intensity to be spin system is reduced. The growing intensity reflects the enhanced metallic character in Ca_0.90Sr_0.10RuO₃ compound and helps to explain the observed ferromagnetism ¹⁰.

4. Conclusion

In conclusion, the visible PL of Ca_0.90Sr_0.10RuO₃ compound in the range temperature 6.7-296 K was measured, and its temperature dependence was also presented. The intensities of the peak B ₁ and B ₂ varied with temperature. The Ca_0.90Sr_0.10RuO₃ compound show a metallic behavior and short-range ferromagnetic interactions appear. The disappearance of the long-range magnetic order is commonly related to the distortion of the RuO₆ octahedra associated with the partial or total replacement of Sr²⁺ ions by Ca²⁺ ions, and the corresponding narrowing of the 4d bandwidth.