Study of A-site disorder dependent structural properties and magnetic ordering in polycrystalline perovskite Sm0.5Ca0.5−xSrxMnO3

This paper reports the studies on the effect of A-site substitution by strontium on the structural properties and magnetic ordering in polycrystalline perovskite Sm0.5Ca0.5−xSrxMnO3 (x = 0.1, 0.2 and 0.3). The investigated samples are prepared by conventional solid-state reaction technique. XRD analysis at room temperature has confirmed orthorhombic structure of the sample with space group Pnma. The dependence of structural parameter, Curie temperature and coercivity on Sr doping content has been thoroughly investigated. It is observed that substitution of Sr2+ for Ca2+ increases lattice parameter, tolerance factor and the Curie temperature. However, the coercivity (Hc) decreases with increasing Sr content while the charge ordering process is weakened with increasing Sr content. Field cooled (FC) and zero-field cooled (ZFC) dc magnetizations measurements at low field and low temperature indicate that there is a spin-glass (SG) like state occurred. Temperature dependent ac susceptibility at different frequency indicates a spin-glass-like transition of the sample.

chemical composition were determined by using Scanning Electron Microscopy (SEM) which includes an energy dispersive x-ray diffractometer (EDX). Field cooled (FC) and zero field cooled (ZFC) magnetization, magnetization as a function of manetic field at different temperatures and temperature dependent ac susceptibility M´(T) were investigated by means of a Quantum Design Superconducting Quantum Interference Device (SQUID) magnetometer.

Results and discussion:
Room temperature XRD patterns of Sm 0.5 Ca 0.5−x Sr x MnO 3 (x=0.1, 0.2 and 0.3.) are shown in figure 1(a). Detailed analysis (those related to the phase characterization and identification of diffraction reflections) of the crystal structure based on reference code# 04-018-9809 in X′pert Highscore Plus software with jcpds reveals that all the samples are found to be crystallized in orthorhombic system and all the peaks are related to Pnma space group. But the Miller indices of only the major peaks are shown in the figure 1(a). The shift of the XRD peaks towards smaller angles demonstrates an increase of the lattice parameters in the prepared samples with increase in Sr content. Also, it is clearly seen that all the samples are single phase in nature without any detectable impurities within the measurement range of study. All patterns share the same characteristics peaks and there are no anomalous peaks due to impurities.
The lattice parameters a, b and c as well as the cell volume V are calculated from the XRD data analyzed with Rietveld refinement using an 'X′pert Highscore Plus' software and summarized in table 1. It is observed that cell volume increases due to the substitution of Ca 2+ (r A ∼1.18 Å) by a larger Sr 2+ ion (r A ∼1.32 Å). However, no structural phase transition has been found in the system. The lattice parameters a, b and c 2 / verified the relation c a b 2 < < / which indicates that the samples are characterized by the presence of static Jahn-Teller distortion [7]. In perovskite manganites, the Jahn-Teller (JT) effect plays an important role in ground state (GS) and transport properties. Where, in this system three electrons in Mn ion lie in lower t2g levels, contributing a  local spin of S=3/2 due to a strong Hund's rule coupling, while another electron stays at the e g level. The degeneracy leads to Mn 3+ with 3d 4 configuration being JT ions. According to the JT theorem, the highsymmetric MnO 6 octahedral degenerated configuration is unstable and the degeneracy is lifted through a distortion of the octahedron. As a result we can say Jahn-Teller coupling of rather modest strength and may therefore strongly influence such properties of the ion as its magnetic behaviour [8]. From where it is seen that dependence of structural parameters for manganite on the A-site substitution is understood in terms of distortions and rotation of the MnO 6 octahedra induced by the mismatch size of the A-cation. The XRD profiles of the samples in figure 1(a) indicate a decrease in the width of diffraction peaks with increasing x, resulting in an increase in crystallite size. The crystallite sizes (d) are evaluated from the XRD profiles of the samples using Debye-Scherer equation d=Crystallite size (nm) K = 0.9 (Scherer constant) λ=0.15406 nm (wavelength of the x-ray) β = FWHM (radians) θ = Peak position (radians) X'pert HighScore Plus software was used to evaluate 2θ and FWHM values from the XRD profiles of the samples and Microsoft Excel software was used to calculate the d value using Debye-Scherer equation. A regular enhancement in crystallite sizes d is seen in the samples with increase in x content. The tolerance factor which is a measure of deviation from an ideal cubic perovskite structure, is calculated from the ionic radii of the lattice sites in a perovskite structure and defined by Goldschmidt as Where r , A á ñ r Mn á ñ and r o are the average ionic radius of the A-site, Mn and O ions respectively. The average A-site ionic radius for our sample is calculated using As t decreases from 1, the lattice structure transforms to rhombohedral (0.96 < t < 1) and then to orthorhombic structure (t < 0.96) [9]. For our samples the tolerance factor t varies from 0.856 for x=0.1 to 0.870 for x=0.3 (table 1). This is attributed essentially to the replacement of a smaller Ca 2+ ion by a larger ion of Sr 2+ . Figure 1(b) shows SEM image of a representative sample with x = 0.3. It is seen that the sample contains small and larger size particles homogeneously distributed throughout the sample. The average grain sizes (D) are estimated from SEM images by using 'ImageJ' software with line intercept method. In this method

Average grain size Line length No of grain =
In this way average grain size was measured as many times as possible and their mean value was taken as final average grain size. The D values for all the samples are larger in comparison to their crystallite size, indicating a multi-domain structure [10]. Figure 1(c) shows the EDX spectrum of a representative sample with x = 0.3. The EDX spectrum shows some impurity element like Ba (0.4%) and C (0.3%) as shown in figure 1(c) which has been considered negligible within experimental error. Then the elemental composition of the synthesized samples normalized to 100 wt% is given in table 2 which shows that the calculated value agreed well with the experimental value. Figure 2(a) shows zero-field cooled (ZFC) and field cooled (FC) magnetization measured at 100 Oe applied field from temperature 10 K to 150 K and inset shows the temperature range 150 K to 300 K. Field cooled (FC) and zero-field cooled (ZFC) DC magnetization measurements show a divergence at low temperature, which indicates the coexistence of anti-ferromagnetic and ferromagnetic clusters. Some salient features of these curves are that all the samples exhibit a paramagnetic to ferromagnetic transition and the Curie temperature increased from 22 K to 31 K with increasing Sr-doping. It is also noted from figure 1(a) that the magnetization increases with increasing Sr content (x) in Sm 0.5 Ca 0.5−x Sr x MnO 3 . The Sr 2+ replacement does not change Mn 3+ /Mn 4+ content ratio in the samples, but enhances slightly the values of 〈r A 〉 [11]. Therefore the increase of T c on x content could be attributed to the enhancement of the double exchange interaction due to the increase of 〈r A 〉 [12]. There is a second transition visible between 50 K and 58 K which may be ascribed to some ferromagnetic clusters of larger size particles. A broad hump is seen above 200 K (more clear for x = 0.1), which is shown at the inset of figure 2(a) and indicates the charge ordering temperature (T CO ). The sample with x = 0.1 shows a clear charge ordering at temperature T co = 225 K. For x = 0.2 the broad hump is almost disappeared and for x = 0.3 the charge ordering process is weakened and the observed T co = 211 K.
In ZFC curve for x = 0.1 a broad hump is seen around T=225 K due to charge ordering and another kink is found around 110 K (more visible in dM/dT versus T curves) which correspond to antiferromagnetic ordering. With further lowering the temperature magnetic moments of larger size particle tend to align and magnetization increases around 50 K. The magnetization decreases onward due to the relaxation of magnetic moments of the larger size particles and it continues up to temperature 30 K. Upon further lowering of temperature magnetization increases presumably due to the magnetic moment ordering of smaller size particle until the temperature reaches up to 21 K. Afterwards the magnetization decreases due to relaxation of smaller size particles. From the extrapolation of these curves from their peaks it is seen that relaxation of small size particles occur more rapidly than that of large size particles. The ZFC curves for other samples shows the similar trend but the amount of magnetization increases and the Curie temperatures shifted to higher temperature with increasing Sr content. Curie temperature is determined from the minimum of the dM/dT versus T curves which has been shown in figure 1(b). It is found that T c increases with increasing Sr content from 22 K for x = 0.1 to 31 K for x = 0.3.
In figure 3(b), the isothermal M-H curves at temperature T=100 K show a linear behavior which indicates the antiferromagnetic state. The enlarged figure at very low field is shown in the inset of figure 3(b) which shows some coercivity (H c ) for x = 0.1 and x = 0.3 but coercivity (H c ) for x = 0.2 is found approximately zero. At M-H curves at T=10 K, the situation is changed slightly as shown in the figure 3(a), the behavior of the samples is still dominated by paramagnetic contributions and saturation is not reached in fields up to 50,000 Oe (5 T) applied field. The enlarged figure at very low field is included in the inset of figure 3(a) which shows that coercivity (H c ) decreases from 1238 Oe to 308 Oe with increasing Sr content. These values of coercivity (H c ) reveal the characteristics of short-range FM order in Sm 0.5 Ca 0.5−x Sr x MnO 3 . It is also seen from the lower inset of figure 3(a) that hysteresis loop for the sample with x = 0.2 is shifted toward the negative field which is a signature of exchange bias effect that happens due to the ferro and anti-ferromagnetic coupling.
Temperature dependent inverse magnetic susceptibility (1/χ = H/M) curves (shown in figure 4(a)) deduced from M (T) measurement in a magnetic field of 100 Oe show a clear deviation from the mean-field Curie-Weiss behavior above T C [13]. The degree of magnetic frustration can be measured by the divergence  between T c and θ, where θ is the paramagnetic Curie temperature. In our case the θ value are higher than the T c value. This difference between the experimental and theoretical data confirms the magnetic inhomogeneity of the sample around T c . In order to understand the spin glass (SG) behavior a frequency dependent ac susceptibility M´(T) measurement at low field was done. In figure 4(b) the ac susceptibility M´(T) curves at frequencies 0.17 Hz, 1.7 Hz, 17 Hz, 170 Hz for x = 0.3 shows that the peak position of the M´(T) curves is frequency dependent. It is noticed that the height of these peaks descends with increasing frequency which is a signature of spin glass (SG) like state. Figure 4(c) shows M versus T plots obtained from the field cooled warming  condition within the temperature range 10-300 K at 10,000 Oe (1 T) applied magnetic field. Sample with x = 0.1 shows a clear charge ordering at temperature T co = 225 K and at the low temperature below T co the sample undergoes a second order phase transition from paramagnetic to ferromagnetic state. This transition is very broad suggesting a smearing of the transition due to lattice defects. In the sample with x = 0.3 a weak charge ordering at temperature T co = 211 K is found and with decreasing temperature from T co the sample undergoes a second order phase transition from paramagnetic to ferromagnetic state. This transition is comparatively sharper than that of the sample with x = 0.1. This curve also shows a clear anti-ferromagnetic transition at T=25 K for x = 0.3. Because of smearing of the paramagnetic to the ferromagnetic transition the magnetocaloric effect is expected to be small and was not studied for x = 0.2. However, the magnetic entropy change (−ΔS m ) for the sample with x = 0.1 and x = 0.3 shown in figure 4(d) was calculated from M versus T curves taken at H=10000 Oe (1 T). For isothermal processes, total magnetic entropy change ΔS m of a magnetic system due to application of a magnetic field H is given by the Maxwell's thermodynamic relation [14]: In order to evaluate this quantity we used an approximation of the above integral in the form The value of magnetic entropy change (−ΔS m ) was 0.012 J K −1 g −1 K −1 for x = 0.1 and 0.028 J K −1 g −1 K −1 for x = 0.3 around T=66 K for a magnetic field change ΔH=10,000 Oe (1 T).

Conclusion:
Polycrystalline samples of Sm 0.5 Ca 0.5−x Sr x MnO 3 were prepared by conventional solid-state reaction technique. All the compounds crystallize in single phase orthorhombic structure with space group Pnma. Present investigation shows that the structural and magnetic properties of Sm 0.5 Ca 0.5−x Sr x MnO 3 can be tuned by Sr doping. The substitution of Sr 2+ for Ca 2+ increases the lattice parameter, tolerance factor and the Curie temperature. Magnetic measurements show that all the samples exhibit a paramagnetic to ferromagnetic transition though a clear deviation from the mean-field Curie-Weiss behavior has been observed above T c. The increase of tolerance factor and cell volume and decreasing of coercivity (H c ) due to the Sr doping for Ca plays an imperative role for A-site disorder which in turn shifted the Curie temperature from 22 K for x = 0.1 to 31 K for x = 0.3. It is also noted that the magnetization increases with increasing Sr content. Coercivity (H c ) decreases from 1238 Oe to 308 Oe with increasing Sr content and it is seen that the coercivity is temperature dependent. In soft magnetic materials, the coercivity of perovskite usually depends on temperature. We suggest that the temperature dependence of the magneto-crystalline anisotropy has a similar behavior. The sample with x = 0.1 shows a clear charge ordering at temperature T co = 225 K and the charge ordering process is weakened lowering the T co = 211 K for x = 0.3. The value of magnetic entropy change (−ΔS m ) was 0.012 J K −1 g −1 K −1 for x = 0.1 and 0.028 J K −1 g −1 K −1 for x = 0.3 around T=66 K for a magnetic field change ΔH=10,000 Oe (1 T). It is observed that the sample with x = 0.3 shows a spin-glass-like behavior at low temperature.