Structural, magnetic and magnetocaloric properties of La0.67Ba0.22Sr0.11Mn1 − xFexO3 nanopowders

Solid State Sciences 37 (2014) 121e130

Contents lists available at ScienceDirect

Solid State Sciences journal homepage: www.elsevier.com/locate/ssscie

Structural, magnetic and magnetocaloric properties of La0.67Ba0.22Sr0.11Mn1
xFexO3 nanopowders F. Ben Jemaa a, S. Mahmood b, M. Ellouze a, *, E.K. Hlil c, F. Halouani d, I. Bsoul e, M. Awawdeh f a

Sfax University, Faculty of Sciences of Sfax, B. P. 1171-3000, Tunisia The University of Jordan, Amman, Jordan Institut N
eel, CNRS et Universit
e Joseph Fourier, B. P. 166, 38042, Grenoble, France d Sfax University, National School of Engineers, B.P.W 3038, Tunisia e Al al-Bayt University, Mafraq, Jordan f Yarmouk University, Irbid, Jordan b c

a r t i c l e i n f o

a b s t r a c t

Article history: Received 6 February 2014 Received in revised form 30 August 2014 Accepted 12 September 2014 Available online 16 September 2014

The influence of Fe-doping on the structural, magnetic and magnetocaloric properties of La0.67Ba0.22Sr0.11Mn1
xFexO3 perovskite was investigated. These samples were synthesized by SoleGel method. X-ray diffraction (XRD) measurements for all samples indicates that all samples, except that with x ¼ 0.2 and 1.0, crystallize in the orthorhombic single-phase with Pnma space group. The Rietveld refinement of the X-ray nanopowder data confirms this result. Room-temperature transmission € ssbauer spectra for all samples except that with x ¼ 1.0 consisted of doublets with similar isomer shift Mo (d) values consistent with Fe3þ valence state. The spectrum for the sample with x ¼ 1.0 showed the presence of an additional magnetically ordered phase and showed evidence of the presence of Fe4þ valence state. Magnetization measurements versus temperature under magnetic applied field of 0.05 T showed that all of our investigated samples exhibit a paramagneticeferromagnetic transition at low temperature except that with x ¼ 1.0. The Curie temperature decreased with increasing Fe content from 345 K for x ¼ 0e80 K for x ¼ 0.3. The magnetic entropy change jDSMj for the parent compound was evaluated using Maxwell's relation. The maximum value of the magnetic entropy change obtained from




1
1 K for applied magnetic field of 5 T. Our result on magthe M(H) plot data is
DSMax M
¼ 2.26 J kg netocaloric properties suggests that the parent sample La0.67Ba0.22Sr0.11MnO3 could be a candidate refrigerant for low temperature magnetic refrigeration. © 2014 Published by Elsevier Masson SAS.

Keywords: Perovskite manganites X-ray diffraction Isothermal magnetization Magnetocaloric effect

1. Introduction Perovskite-type oxides have the general formula ABO3 where A is an alkali or alkaline-earth ion and B is a 3d transition metal ion. A special class of these oxides is lanthanum manganites (La, A) Mn1
xTxO3 (where A is a divalent ion such as Ba2þ, Sr2þ, Ca2þ etc, and T ¼ Mn, Fe, Co, Ti, etc.) which provides the opportunity for tuning the Mn3þ/Mn4þ ration and consequently modifying their properties. These perovskites attracted considerable interest due to their potential applications in solid oxide fuel cell (SOFC) cathodes, oxygen separating membranes, read heads in modern hard disk devices, and chemical sensors for gases [1e4]. In addition, these

* Corresponding author. E-mail address: [email protected] (M. Ellouze). http://dx.doi.org/10.1016/j.solidstatesciences.2014.09.004 1293-2558/© 2014 Published by Elsevier Masson SAS.

oxides exhibit several important characteristics such as magnetocaloric effect (MCE), colossal magnetoresistance (CMR), charge ordering, orbital ordering, isotope effect, etc. [5e7]. These effects could be due to Mn3þ/Mn4þ mixed-valence state of the oxides, and as a consequence, a large number of studies were devoted to Lamanganites with La ions substituted by rare earth or alkaline earth ions and Mn ions substituted by 3d-transition metal ions [8,9]. Due to the similarity of the ionic radii of Mn3þ and Fe3þ ions (0.645 Å), Fe3þ ions substitute Mn3þ ions in six-fold octahedrally coordinated sites [10]. Accordingly, the substitution of Fe3þ for Mn3þ is not expected to modify the tolerance factor and, therefore, the Jahn-Teller distortion effect can be neglected. However, the deformation of the local structure in manganites which exhibit € ssbauer spectroscopy, such distortions can be investigated by Mo which is an effective technique to examine deviations from octahedral symmetry at the Fe sites [11]. On the other hand, MCE in

122

F. Ben Jemaa et al. / Solid State Sciences 37 (2014) 121e130

these oxides is defined by the adiabatic temperature change (DTad) or the isothermal magnetic entropy change (DSM), which is a function of both magnetic field and temperature. Generally speaking, there are three principal methods to evaluate MCE [12]. First, by direct measurement of the adiabatic temperature change (DTad), which is carried out by exposing thermally insulated sample to the magnetic field. The second method is based on Maxwell's relation, and DSM is calculated from magnetization measurements, and the third method is based on heat capacity measurements in zero-field cooled (ZFC) and field cooled (FC) processes [13e15]. In this article, we report on the experimental results concerning the effects of Fe-doping on the magnetic properties of LaMnO3 prepared by Sol-Gel method. The manganites were characterized

by X-ray diffraction (XRD), Scanning Electron Microscopy (SEM), € ssbauer Spectroscopy (MS), and magnetization measurements Mo at different temperatures and in different applied magnetic fields. The magnetization data near the transition temperature were examined with the assistance of the modified Kouvel-Fisher Arrott plot method and the magnetocaloric effect was evaluated. 2. Experimental details Nanopowders of La0.67Ba0.22Sr0.11Mn1
xFexO3 (x ¼ 0, 0.1, 0.2, 0.3 and 1.0) oxides were prepared by SoleGel method [16e17], and single perovskite phase was obtained by appropriate heat treatment in accordance with the following reaction:

Fig. 1. XRD patterns of La0.67Ba0.22Sr0.11Mn1
xFexO3 ((a): x ¼ 0, (b): x ¼ 0.1, (c): x ¼ 0.2, (d): x ¼ 0.3 and (e): x ¼ 1.0) with Rietveld refinement (experimental data in red, calculated patterns in black, difference between them in blue and Bragg positions in green). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

F. Ben Jemaa et al. / Solid State Sciences 37 (2014) 121e130 Table 1 Cell parameters, La0.67Ba0.22Sr0.11Mn1




unit cell volume and Curie xFexO3 with (x ¼ 0, 0.1, 0.2, 0.3 and 1).

temperature

Samples

a (Å)

b (Å)

c (Å)

v (Å3)

TC (K)

La0.67Ba0.22Sr0.11MnO3 La0.67Ba0.22Sr0.11Mn0.9Fe0.1O3 La0.67Ba0.22Sr0.11Mn0.8Fe0.2O3 La0.67Ba0.22Sr0.11Mn0.7Fe0.3O3 La0.67Ba0.22Sr0.11FeO3

5.49299 5.49824 5.48432 5.49878 5.55271

7.79671 7.79275 7.81047 7.80175 7.85943

5.52697 5.53661 5.53495 5.53988 5.53789

236.70 237.22 237.09 237.66 241.68

345 190 130 80 e

of

Table 2 Refined structural parameter for La0.67Ba0.22Sr0.11Mn1
xFexO3 (x ¼ 0.0, 0.1, 0.2, 0.3 and 1.0), and the numbers in subscript represent the error bars. Samples La/Ba x y z Mn/Fe x y z O1 x y z O2 x y z c2 tG Dhkl (nn) DSEM (nm)

x¼0

x ¼ 0.1

x ¼ 0.2

x ¼ 0.3

x¼1

0.00501 0.25 0.00521


0.00038 0.25 0.00295

0.00815 0.25 0.00447

0.00594 0.25
0.00565

0.01771 0.25
0.00484

0 0 0.5

0 0 0.5

0 0 0.5

0 0 0.5

0 0 0.5

0.48283 0.25
0.00521

0.47779 0.25 0.02663

0.48975 0.25
0.02111

0.48456 0.25
0.00252

0.48989 0.25 0.02259

0.25557
0.00245 0.77323

0.26318 0.01423 0.76292

0.26795
0.02998 0.76104

0.25934
0.02098 0.77377

0.27292 0.02935 0.74165

1.43 0.9253 28.11 675

1.52 0.9255 29.02 441

1.57 0.9257 69.15 472

1.38 0.9259 31.45 567

1.33 0.9268 35.12 151

123

0.335 La2O3 þ 0.22 BaO þ 0.11 SrO þ x/2 Fe2O3 þ (1
x) MnO2 / La0.67Ba0.22Sr0.11Mn1
xFexO3 þ d CO2 The appropriate amounts of high purity (99.9%) La2O3, MnO2, Fe2O3, BaO and SrO powder precursors were dissolved in nitric acid and distilled water. The solution was then heated at 90  C on a hot plate under constant stirring to eliminate the excess water and to obtain a homogeneous solution. Subsequently, citric acid and ethylene glycol were added under thermal agitation to obtain a viscous gel. The gel was then dried at 300  C and calcinated at 600  C for 5 h resulting in fine powder. Finally, the powders were ground and mixed in a mortar for 10 min and then placed in platinum crucibles and compressed into pellets 12 mm in diameter and 2 mm thick. The pellets were finally sintered at 1100  C for 3 h. The structure, phase purity and homogeneity of as-elaborated samples were checked by powder X-ray diffraction (XRD) using CuKa radiation (l ¼ 1.5406 Å) at room temperature. Structural analysis was carried out using the standard Rietveld technique [18]. The XRD patterns indicated that all prepared samples consisted of a major phase with orthorhombic structure. The particle morphology and average physical particle size, as well as the chemical stoichiometry of the samples were investigated by Scanning Electron Microscopy (SEM). The hyperfine structure of the samples was €ssbauer spectroscopy using a conventional examined by 57Fe Mo € ssbauer spectrometer. Magnetization constant acceleration Mo measurements versus temperature in the range 5e500 K and isothermal magnetization curves in an applied magnetic field of up to 6 T were obtained using a Vibrating Sample Magnetometer (VSM). Isothermal curves around the ferromagnetic ordering transition temperature (TC) of the samples were obtained insteps of 5 K (and 2.5 K for the sample with x ¼ 0.1). The room-temperature magnetic hysteresis loops of the samples were measured in applied

Fig. 2. Crystal structure of La0.67Ba0.22Sr0.11MnO3 and the coordination polyhedron on Mn.

Table 3 Interatomic distance and angles for La0.67Ba0.22Sr0.11Mn1 0.3 and 1). Distances MneO1 (Å) MneO21 (Å) MneO22 (Å) Angles MneO1eMn ( ) MneO2eMn ( )

x¼0 1.95 1.84 2.06 174.2 175.8


xFexO3

(x ¼ 0.0, 0.1, 0.2,

x ¼ 0.1

x ¼ 0.2

x ¼ 0.3

1.96 1.85 2.05

1.96 1.85 2.07

1.95 1.83 2.09

168.8 173.5

172.4 166.1

174.9 169.82

x¼1 1.97 1.92 2.03 172.1 164.8

magnetic fields ranging from 0 to 1 T. The MCE was calculated from the temperature dependence of the magnetization measurements versus applied magnetic field up to 5 T. 3. Results and discussion 3.1. Structural X-ray diffraction (XRD) measurements for La0.67Ba0.22Sr0.11Mn1
xFexO3 (0, 0.1, 0.2, 0.3 and 1.0) were carried

124

F. Ben Jemaa et al. / Solid State Sciences 37 (2014) 121e130

Fig. 3. Typical scanning electron micrographs of La0.67Ba0.22Sr0.11Mn1


xFexO3

out at room temperature. Fig. 1, indicates that all samples, except that with x ¼ 0.2 and 1.0 consist of an orthorhombic single-phase with Pnma space group. The weak diffraction peak at 2q ¼ 31.53 for the sample with x ¼ 0.2 is due to the precipitation of a very small amount of Mn3O4 (see Fig. 1c) which is consistent with previously reported results [19,20]. However, the sample with x ¼ 1.0 contained small amounts of additional phases. The weak diffraction peaks at 2q ¼ 28.0 and 34.3 correspond to the precipitation of a very small amount of (Ba,Sr)Fe2O4 spinel phase, while the weak diffraction peaks at 2q ¼ 25.4 , 33.0 and 53.6 are consistent with the precipitation of a very small amount of a-Fe2O3 (see Fig. 1e). Rietveld refinement of the X-ray powder data confirmed the presence of these minor phases, and the good agreement between the observed and calculated inter-planer spacing for the perovskite major phase is coherent with the value of the Goldschmidt tolerance factor tG:

rA þ rO ffi tG ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 2ðrB þ rO Þ Here rA, rB and rO are respectively the average ionic radii of the A and B perovskite sites and oxygen anion [21]. Rietveld refinement of the X-ray powder data was carried out to determine the lattice parameters, unit cell volume, and atomic positions using the FULLPROF code [22]. In Table 1 we list the refined unit cell parameters for all samples where unconstrained refinement of the data gave good values of Rf and c2 parameters. The results of our refinement are summarized in Table 2. The structural analysis indicated that La/Ba/Sr ions were located at 4c (x, 0.25, z) position, Mn/Fe ions at 4b (0, 0, 0.5), and oxygen ions occupied two different sites, namely O1 at 4c (x, 0.25, z) and O2 at 8d (x, y, z) positions. We also noticed that the lattice parameters and the unit cell volume increased significantly only upon complete substituting manganese by iron. However, the volume and lattice parameters should be essentially independent of the Fe content due to the similarity of the radii of Mn3þ and Fe3þ ions in the octahedral sites as indicated by the constancy of the average tolerance factor tG (see Table 2). We can therefore associate this increase of the lattice parameters and cell volume with the distortion of the original crystal lattice structure and the

samples; (a) x ¼ 0, (b) x ¼ 0.1, (c) x ¼ 0.2, (d) x ¼ 0.3 and (e) x ¼ 1.0.

adjustment of the MneO bond length and MneOeMn bond angles. Our results are nearly comparable to those obtained by Jonker et al. [10] and Kallel et al. [23]. The average crystallite size was estimated from the XRD data using the Scherer relation [24]:

D¼

Kl b cos q

Here K, l, b, and q are the grain shape factor, the X-ray wavelength, the full width at half maximum (FWHM) of the diffraction peak, and the Bragg diffraction angle, respectively. The values of the effective crystallite sizes are summarized in Table 2. It is evident that the crystallite sizes of the samples were similar, with the exception of that with x ¼ 0.2 which exhibited a better degree of crystallinity. “Diamond” program was used to draw the unit cell (Fig. 2) and determine the distances between Mn ions and the nearest neighbor oxygen ions. The Pnma structure of La0.67Ba0.22Sr0.11Mn1
xFexO3 at room temperature is evident, and although the MnO6 octahedral are almost regular, they are slightly distorted as observed in a previous work [25]. The interatomic distances and MneOeMn angles for the La0.67Ba0.22Sr0.11Mn1
xFexO3 samples, in the orthorhombic structure, are listed in Table 3. Scanning Electron Microscopy (SEM) was employed to examine the morphology of the La0.67Ba0.22Sr0.11Mn1
xFexO3 samples (x ¼ 0, 0.1, 0.2, 0.3 and 1.0). Fig. 3 shows that agglomerated particles are generally parallelepipedic with a homogeneous microstructure, fine granulation and porous structure. We estimated in (Table 2) the average grain size by SEM based on the average grain size of ten compared with those calculated by the Deby-Scherrer formula using data of X-ray diffraction. The size of the grains observed by SEM are greater than those calculated by the Deby-Scherrer formula, this can be explained by the fact that each particle observed by SEM is formed by several grains crystallized [26]. The chemical composition of the samples was estimated from the elemental analysis using energy dispersive X-ray spectra (EDX). Fig. 4 is a representative spectrum for the un-doped sample, where X-ray counts are presented vs. energy (in keV). The observed energy peaks correspond to the various elements in the sample. The

F. Ben Jemaa et al. / Solid State Sciences 37 (2014) 121e130

Fig. 4. DAX analysis spectral at room temperature of La0.67Ba0.22Sr0.11Mn1


xFexO3

125

samples; (a) x ¼ 0, (b) x ¼ 0.1, (c) x ¼ 0.2, (d) x ¼ 0.3 and (e) x ¼ 1.0.

obtained chemical composition of La0.67Ba0.22Sr0.11Mn1
xFexO (x ¼ 0, 0.1, 0.2, 0.3 and 1.0) is close to the nominal ones. €ssbauer measurements 3.2. Mo €ssbauer The transmission Mo spectra of La0.67Ba0.22Sr0.11Mn1
xFexO3 (x ¼ 0.1, 0.2, 0.3 and 1.0) compounds recorded at room-temperature are shown in Fig. 5. The spectra for the samples, with the exception of that with x ¼ 1.0, each consists of a €ssbauer spectroscopy is quadrupole doublet. Due to the fact that Mo a local effect, the presence of the doublet is characteristic of fast paramagnetic relaxation of the Fe moments. The sample with x ¼ 1.0 shows a magnetically split component in addition to the central paramagnetic component. Accordingly, the spectrum for this sample was fitted with a magnetic sextet and a doublet; the remaining spectra were fitted each with a doublet only. The resulting hyperfine parameters (the isomer shift (d) quadrupole splitting (D) and line width) are given in Table 4. The isomer shifts of about 0.35 mm/s and quadrupole splitting of 0.40e0.45 mm/s for the samples with x ¼ 0.2 and 0.3 are characteristic of F3þ ions at distorted octahedral sites [27e30]. This splitting could be partially due to distortion arising from the presence of small polarons at room temperature [27,29]. The line width in the spectra of these samples is significantly larger than the intrinsic line width of about 0.2 mm/s, which could be attributed to the random chemical environment around the Fe sites. The spectrum for the sample with x ¼ 0.1 is characterized by a significantly lower isomer shift (0.08 mm/s) and quadrupole splitting (0.17 mm/s). These hyperfine

€ssbauer Fig. 5. Mo spectra recorded at room-temperature of La0.67Ba0.22Sr0.11Mn1
xFexO3 ((a): x ¼ 0.1, (b): x ¼ 0.2, (c): x ¼ 0.3 and (d): x ¼ 1). Recorded at room temperature.

F. Ben Jemaa et al. / Solid State Sciences 37 (2014) 121e130

Fe

contends

for

d

D

W

Chi2

La0.67Ba0.22Sr0.11Mn0.9Fe0.1O3 La0.67Ba0.22Sr0.11Mn0.8Fe0.2O3 La0.67Ba0.22Sr0.11Mn0.7Fe0.3O3 La0.67Ba0.22Sr0.11FeO3

0.08 0.34 0.36 0.36 0.15

0.17 0.40 0.45
0.01 0.27

0.21 0.40 0.39 0.59

0.54 0.42 0.58 1.05

80 X=0.2

40

X=0.0 X=0.1 X=0.2 X=0.3 X=1.0

60

2

La0.67Ba0.22Sr0.11Mn1-xFexO3

35

22

42

M(Am /kg)

Samples

La0.67Ba0.22Sr0.11Mn1-xFexO3

100

40

38

X=0.3

20 18

36

M (emu/g)

Table 4 €ssbauer Mo parameter as a function of the La0.67Ba0.22Sr0.11Mn1
xFexO3 (x ¼ 0.0, 0.1, 0.2, 0.3 and 1).

M (emu/g)

126

34 32

16 14 12 10

30

8

28

6

26 0

1

2

3

4

5

0

6

1

2

3

4

5

6

H(T)

H(T)

20 30

0

25

0

1

2

3

4

5

6

M(Am2/kg)

H(T) 20

Fig. 8. Magnetization versus applied magnetic field at 5K for La0.67Ba0.22Sr0.11Mn1
xFexO3 with (x ¼ 0, 0.1, 0.2, 0.3 and 1). The insets show the fitting of isothermals curves corresponding respectively for each samples with x ¼ 0.2 and x ¼ 0.3.

15

X=0 X=0.1 X=0.2 X=0.3 X=1

10

5

broadened. These parameters are associated with the presence of a high Fe4þ/Fe3þ ratio of 0.33 in this sample, and the electron hopping between the two valence states in neighboring octahedral sites. This electron relaxation process is responsible for the observation of one paramagnetic component with hyperfine parameters resembling some sort of an average of those for the two valence states.

0 0

100

200

T(K)

300

400

500

600

Fig. 6. (a) Temperature dependence of magnetization in a magnetic applied field of 0.05 T for all samples La0.67Ba0.22Sr0.11Mn1
xFexO3 (with x ¼ 0.0, 0.1, 0.2, 0.3 and 1). (b) The minimum values of dM/dT curves are defined as TC of the samples.

3.3. Magnetic measurements Fig. 6 shows the magnetization of La0.67Ba0.22Sr0.11Mn1
xFexO3 (x ¼ 0.1, 0.2, 0.3 and 1.0) as a function of temperature for an applied field of 0.05 T. The samples with x ¼ 0, 0.1 and 0.2 exhibited ferromagnetic (FM) to a paramagnetic (PM) phase transition as the temperature increased. Curie temperature (TC) for each sample was determined from the minimum of dM(T)/dT at the inflection point of the thermomagnetic curve, and the results are listed in Table 1. It is clear that TC could be determined for x ¼ 0, 0.1, 0.2 and 0.3, which indicates that the level of substitution of manganese by iron in these samples does not destroy the ferromagnetic behavior below a temperature which is sensitive to the value of x [31]. Further, TC for the un-substituted sample was found to be above room temperature, while even a low level of the substitution lowers TC to below room temperature. The observed decrease in magnetization and TC

parameters are too low in comparison with those characteristic of Fe3þ valence state, and could be associated with Fe4þ valence state. The presence of this valence state was previously reported in La0.67Sr0.33FexMn1
xO3 system [29]. The line width in the spectrum for this sample is not broadened with respect to the intrinsic line width, which could be due to the fact that at such a low Fe concentration, the local chemical environments are not perturbed significantly. The spectrum for the sample with x ¼ 1.0 consists of a magnetically split component with hyperfine parameters consistent with those for a-Fe2O3, and a doublet with intermediate values of the isomer shift (0.15 mm/s) and quadrupole splitting (0.27 mm/ s). Further, the spectral line width for this sample is significantly

4

60

a

b

3

40

La0.67Ba0.22Sr0.11Mn1-xFexO3

2

T=300K

La0.67Ba0.22Sr0.11MnO3

M(Am2/kg)

M(Am2/kg)

20

0

1 0

-1

x=0.1 x=0.2 x=0.3 x=1

-20 -2 -40

-60

-3

-1

-0.5

H(T)

0.5

-4 1

-1

-0.5

H(T)

0.5

1

Fig. 7. (a) Magnetization versus applied magnetic field at room temperature for La0.67Ba0.22Sr0.11MnO3. (b) Magnetization versus applied magnetic field at room temperature for La0.67Ba0.22Sr0.11Mn1
xFexO3 with (x ¼ 0.1, 0.2, 0.3 and 1).

F. Ben Jemaa et al. / Solid State Sciences 37 (2014) 121e130

La0.67Ba0.22Sr0.11Mn0.8Fe0.2O3

La0,67Ba0,22Sr0,11MnO3

40

70

35

60

30 M(Am /kg)

M (Am /kg)

50 40 30

25 20 15 10

20

5

10 0

127

0 0

1

2

3 μ0H(T)

4

0

5

1

3

4

5

μ H(T) 0

La0.67Ba0.22Sr0.11Mn0.9Fe0.1O3

60

2

La0.67Ba0.22Sr0.11FeO3

1,0

50

0,8 M (Am /kg)

M (Am /kg)

40 30 20

0,4 0,2

10 0

0,6

0

1

2

3 μ H(T) 0

4

5

0,0

0

1

2

3

4

5

μ H(T) 0

Fig. 9. Isothermal magnetization curves taken at different temperatures.

with increasing x could be associated with the fact that Fe3þ ions do not participate in the ferromagnetic double exchange (DE) interaction with Mn3þ ions, but encourage the antiferromagnetic Fe3þeOeMn3þ, Mn4þeOeFe3þ and Fe3þeOeFe3þ super-exchange (SE) interactions [32]. With regard to the sample with x ¼ 0.3, the magnetization exhibited a cusp at about 50 K. Such a behavior is typical of spin-glasses due to the competition between ferromagnetic (DE) and antiferromagnetic (super-exchange) interactions [33,34]. Finally the sample with x ¼ 1.0 exhibited paramagnetic behavior down to the lowest temperature in this study, as expected for such iron oxides. In order to better understand the magnetic properties of the samples, room temperature hysteresis loops were recorded in an applied magnetic field up to 1 T. Fig. 7a shows that the magnetization for the un-substituted sample is almost saturated (52 A m2/ kg) at 1 T applied field. This result is consistent with the thermomagnetic results which indicated that this sample is ferromagnetic at room temperature. However, the rest of the samples (x ¼ 0.1, 0.2, 0.3 and 1) magnetization curves at room temperature did not saturate in magnetic fields up to 1 T (Fig. 7b). Further, it was observed that the increased level of iron substitution resulted in a significant drop in the spontaneous magnetization, which is associated with the weakening of the Mn3þeMn4þ DE interactions. The magnetization curves for all samples demonstrated paramagnetic behavior. This is also consistent with the thermomagnetic results that TC for all substituted samples, if any, is below room temperature. The variation in the magnetic behavior with increasing Fe substitution level is due to the progressive weakening of the net (DE) interaction due to the decrease in number of Mn3þeO2
eMn4þ bonds [31e34]. The small hysteresis loop for the samples with x ¼ 0.2 and 0.3 could be due to the presence of about 1% of magnetically ordered phases, which reflects the presence of a small degree of in homogeneity in the magnetic structure in these samples. The magnetization curves M(H) at T ¼ 5 K are shown in Fig. 8. The samples with x ¼ 0.0 and 0.1 were obviously ferromagnetic at

this temperature and almost saturated at an applied field of 6 T. The saturation magnetizations for these samples are about 90 and 70 A m2/kg, respectively. However, the curves of the samples with higher x values did not show saturation up to 6 T, which could be due to the increased SE antiferromagnetic coupling and the decreased DE ferromagnetic coupling with increasing Fe concentration, the competition between which could lead to frustration and the development of the spin-glass state [33e35]. The saturation magnetizations for these samples were determined from the low-of-approach to saturation in the high-field range, and found to be 54 and 29 A m2/kg for the samples with x ¼ 0.2 and 0.3, respectively (Fig. 8einsets).

3.4. Magnetocaloric study Magnetization measurements versus applied magnetic field up to 5 T were obtained at several temperatures below and above TC for the samples with x ¼ 0.0, 0.1, and 0.2, and in the temperature range between 100 K and 300 K for the sample with x ¼ 1.0. The isothermal curves (Fig. 9) indicated that the magnetization for the un-substituted sample in the ferromagnetic regime increased sharply and saturated at an applied field below 0.5 T. This behavior is characteristic for a soft magnetic material, and is consistent with the room temperature hysteresis behavior. However, the curves for the substituted samples continued to rise and did not saturate even at 5 T. Fig. 10 shows M2 vs. (m0H/M) Arrott plots derived from the magnetization curves. The plots exhibited a positive slope around TC which confirms that our samples (with the exception of that with x ¼ 1.0) exhibit second order ferromagnetic to paramagnetic phase transition [36,37]. The plots for the sample with x ¼ 1.0, demonstrated the absence of such transition. In order to study the magnetocaloric properties of our samples, the temperature dependence of magnetic entropy change was evaluated (Fig. 11). The magnetic entropy change,
DSM, induced by the magnetic field change was calculated according to the classical

128

F. Ben Jemaa et al. / Solid State Sciences 37 (2014) 121e130

thermodynamic theory based on Maxwell's relations using the following equation [38]:

! DSM T; H

¼

X Miþ1 ðTiþ1 ; HÞ
Mi ðTi ; HÞ DH Tiþ1
Ti i

Here Mi(Ti,H) and Mi þ 1(Ti þ 1, H) are the experimental values of magnetization measured at temperatures Ti and Tiþ1, respectively, under applied magnetic field Hi. Fig. 11a shows the temperature dependence of the magnetic entropy change (eDSM) vs.

La0.67Ba0.22Sr0.11MnO3

5000 4500 4000

2500

2

2000

2

M (A m /kg

)

2

3000

4

3500

1500 1000 500 0 0,00 0,05 0,10 0,15 0,20 0,25 0,30 0,35 0,40 μ0H/M (T.g/emu)

La0.67Ba0.22Sr0.11Mn0.9Fe0.1O3 3000

2000 1500

2

2

4

M (A m /kg

2

)

2500

1000 500 0 0,00

0,05

0,10

0,15

0,20

0,25

0,30

2

μ0H/M (Tkg/(Am ))

La0.67Ba0.22Sr0.11FeO3

1,0

0,6 0,4

2

2

4

M (A M /kg

2

)

0,8

0,2 0,0

0

1

2

3 4 5 2 μ0H/M (Tkg/(Am ))

6

7

Fig. 10. Arrott plots for the samples La0.67Ba0.22Sr0.11Mn1
xFexO3 ((a) x ¼ 0, (b) x ¼ 0.1, (c) x ¼ 0.2, and (d) x ¼ 1).

Fig. 11. Temperature dependence of magnetic entropy change for samples: (a) La0.67Ba0.22Sr0.11MnO3, (b) La0.67Ba0.22Sr0.11Mn0.9Fe0.1O3, (c) La0.67Ba0.22Sr0.11Mn0.8Fe0.2O3.

F. Ben Jemaa et al. / Solid State Sciences 37 (2014) 121e130

temperature for the un-substituted sample at different external applied magnetic fields of 1e5 T. As expected, the magnetic entropy change enhanced with increasing applied magnetic field, and exhibited a maximum value around TC. The magnitude of the maximum entropy change increased from 0.675 J kg
1 K
1 at an applied field of 1 Te2.258 J kg
1 K
1 at an applied field of 5T. Fig. 11(b) and (c) shows eDSM of La0.67Ba0.22Sr0.11Mn1
xFexO3 (x ¼ 0.1 and x ¼ 0.2) samples as a function of temperature at different external magnetic fields. The magnitude of the maximum change in magnetic entropy for the sample with x ¼ 0.1 improved only slightly at high fields with respect to the un-substituted sample, and the peak became sharper around TC. However, the maximum entropy change for the sample with x ¼ 0.2 dropped down to about half of the value for the previous two samples (1.03 J K
1 kg
1at 139 K). This is due to the reduction in the rate of change in magnetization with increasing temperature as a result of the increase in the antiferromagnetically aligned moments of Mn and Fe ions. Our magnetic results are consistent with those obtained by P. Nisha et al. [39]. The small value of magnetic entropy change can be associated with the fact that the magnetic phase transition in these samples is of second order. Generally, materials with first order transitions exhibit much larger magnetocaloric effect than materials with second order transitions. However, the importance of the studied system stems from the ability to tune their properties and control the structural and morphological factors, in addition to the ease of synthesizing such materials with good chemical stability. .In Table 5, we list the MCE values of the present compound in comparison with those reported in the literature having the same chemical elements. The relative cooling power (RCP) was calculated using the formula:

  T; H  dTFWHM RCP ¼
DSMAX M Where dTFWHM is the full-width at half-maximum of jDSMj versus temperature [40,41]. Relatively high RCP values of 193.27 J/kg and 153.2 J/kg at DH ¼ 5 T were obtained for the samples
x ¼ 0 and


x ¼ 0.1. Despite the fact that iron doping reduces TC and
DSMAX M
at relatively high iron concentration, the observed properties of the investigated system are promising and open the way for investigations of materials useful for magnetic refrigeration. 4. Conclusion In this work we investigated structural, magnetic and magnetocaloric properties of nanopowder samples Table 5



Curie temperature TC, magnetic field change, maximum entropy change
DSMAX M
and RCP values for the present sample and for materials with comparable grain size from the literature.


MAX
Samples RCP DH Ref.
DSM

1
1 (J kg K ) (J kg
1) (T) Gd Gd5(Si2Ge2) La0.8Ba0.2MnO3 La0.8Ba0.2Mn0.95Fe0.05O3 La0.8Ba0.2Mn0.9Fe0.1O3 La0.67Ba0.22Sr0.11MnO3

9.5 18.5 4.15 3 2.62 2.258

410 535 230 238 211 193

5 5 5 5 5 5

La0.67Ba0.22Sr0.11Mn0.9Fe0.1O3

2.261

153

5

La0.67Ba0.22Sr0.11Mn0.8Fe0.2O3

1.03

91

5

La0.67Ba0.33MnO3 La0.67Ba0.33Mn0.98Ti0.02O3 La0.7Sr0.3Mn0.093Fe0.07O3

1.48 3.19 4

161 307 25

5 5 2

[35] [35] [35] [35] [35] This work This work This work [35] [35] [35]

129

La0.67Ba0.22Sr0.11Mn1
xFexO3 (x ¼ 0, 0.1, 0.2, 0.3, and 1.0) prepared by solegel method. By virtue of Rietveld refinement we generally found that all samples were almost single-phase and crystallize in the orthorhombic structure with Pnma space group. However, a weak diffraction peak corresponding to a very small amount of Mn3O4 precipitates was identified in the pattern of the sample with €ssbauer spectra were x ¼ 0.2. Room temperature transmission Mo consistent with single paramagnetic phase for all samples except that with x ¼ 1.0 where a magnetic component, in addition, was detected. The hyperfine parameters for the paramagnetic subspectrum of the sample with x ¼ 1.0 were consistent with the presence of both Fe3þand Fe4þ valence states with electron hopping between these states in adjacent octahedral sites. The substitution of Mn by Fe was found to influence the magnetic characteristics significantly as a result of the development of SE interactions at the expense of the ferromagnetic interactions. Relatively large values of



MCE were obtained at low Fe substitution levels, and
DSMAX M
reached the highest value for the sample with x ¼ 0.1 at an applied magnetic field of 5 T.

References [1] E. Bucher, W. Sitte, Solid State Ionics 173 (2004) 23. [2] S.P. Simner, J.F. Bonnett, N.L. Canfield, K.D. Meinhardt, J.P. Shelton, V.L. Sprenkle, J.W. Stevenson, J. Power Sources 113 (2003) 1. [3] S. Tanasescu, N.D. Totir, D.I. Marchidan, A. Turcanu, Mater. Res. Bull. 32 (1997) 915. [4] K. Yan, X. Fu, Y. Cui, J. Inorg. Organomet. Polym. (2011), http://dx.doi.org/ 10.1007/s10904-011-9567-6. [5] M. Paraskevopoulos, F. Mayr, J. Hemberger, A. Loidl, R. Heichele, D. Maurer, V. MÜller, A. Mukhin, A. Balbashov, J. Phys. Condens. Matter 12 (2000) 3993. [6] E. Dixon, J. Hadermann, M.A. Hayward, Chem. Mater. 24 (2012) 1486. [7] A. Urushibara, Y. Moritomo, T. Arima, A. Asamitsu, G. Kido, Y. Tokura, Phys. Rev. B 51 (1995) 14103. [8] N. Chau, H.N. Nhat, N.H. Luong, D.L. Minh, N.D. Tho, N.N. Chau, Phys. B: Condens. Matter 327 (2003) 270. [9] N.H. Luong, D.T. Hanh, N. Chau, N.D. Tho, T.D. Hiep, J. Magn. Magn. Mater. 290e291 (2005) 690. [10] G.H. Jonker, Phys. Amst. 20 (1954) 1118. [11] R.D. Shannon, Acta Cryst. A32 (1976) 751. [12] J. Mira, J. Rivas, L.E. Hueso, F. Rivadulla, M.A. Lopez Quintela, J. Appl. Phys. 91 (2002) 8903. [13] J.Z. Sun, W.J. Gallagher, P.R. Duncombe, L. Krusin-Elbaum, R.A. Altman, A. Gupta, Y. Lu, G.Q. Gong, G. Xiao, Appl. Phys. Lett. 69 (1996) 3266. ~ a, M. Oumezzine, Mater. Lett. 64 (2010) 1045. [14] S. Kallel, N. Kallel, O. Pen [15] N.A. Viglin, S.V. Naumov, Y.M. Mukovskii, Phys. Solid State 43 (2001) 10. [16] M.I. Mendelson, J. Am. Ceram. Soc. 52 (1969) 443. [17] C.J. Brinker, G.R. Scherer, SoleGel Science, the Physics and Chemistry of SoleGel Processing, Academic Press, New York, 1990. [18] H.M. Rietvel, J. Appl. Cryst. 2 (1965) 65. [19] P.A. Joy, C.R. Sankar, S.K. Date, J. Phys. Condens. Matter 14 (2002) L663. [20] A.G. Mostafa, E.K. Abdel-Khalek, W.M. Daoush, S.F. Moustfa, J. Magn. Magn. Mater. 320 (2008) 3356. [21] P. Schiffer, A.P. Ramírez, W. Bao, S.-W. Cheong, Phys. Rev. Lett. 75 (1995) 3336. [22] T. Roisnel, J. Rodriguez-Carvajal, Computer Program FULLPROF, LLB-LCSIM, May 2003. [23] N. Kallel, M. Oumezzine, H. Vincent, J. Magn. Magn. Mater. 320 (2008) 1810.
orie et Technique de la Radio[24] A. Guinier, in: X. Dunod (Ed.), The cristallographie, third ed., 1964, p. 482. [25] M. Ellouze, W. Boujelben, A. Cheikhrouhou, H. Fuess, R. Madar, Solid State Commun. 124 (2002) 125.
rrez, A. Pen ~ a, J.M. Barandiara
n, J.L. Pizarro, T. Herna
ndez, L. Lezama, [26] J. Gutie M. Insausti, T. Rojo, Phys. Rev. B 61 (2000) 9028. [27] S.H. Mahmood, J. Dawood, A.-F. Lehlooh, A. Cheikhrouhou, A. Ammar, Hyperfine Interact. 196 (2010) 385. [28] M. Pissas, G. Kallias, E. Devlin, A. Simopoulos, D. Niarchos, J. Appl. Phys. 81 (1997) 5770.
rif, M. Ellouze, A. Lehlooh, S.H. Mahmood, F. Elhalouani, Hyperfine [29] W. Che Interact. 211 (2012) 153. [30] A. Simopoulos, M. Pissas, G. Kallias, E. Devlin, N. Moutis, I. Panagiotopoulos, D. Niarchos, C. Christides, R. Sonag, Phys. Rev. B 59 (1999) 1263. [31] K.H. Ahn, X.W. Wu, K. Liu, C.L. Chien, Phys. Rev. B 54 (1996) 15299. [32] Y.L. Chang, Q. Huang, C.K. Ong, J. Appl. Phys. 91 (2002) 789. [33] M. Baazaoui, S. Zemni, M. Boudard, H. Rahmouni, A. Gasmi, A. Selmi, M. Oumezzine, Mater. Lett. 63 (2009) 2167.
na, M. Oumezzine, J. Alloys Compd. [34] N. Kallel, Ben S. Abdelkhalek, S. Kallel, O. Pe 501 (1, 2) (2010) 30.

130

F. Ben Jemaa et al. / Solid State Sciences 37 (2014) 121e130


rif, M. Ellouze, A.-F. Lehlooh, F. Elhalouani, J. Alloys Compd. 543 (2012) [35] W. Che 152. [36] S.K. Banerjee, Phys. Lett. 12 (1964) 16. [37] N.K. Singh, K.G. Suresh, A.K. Nigam, Solid State Commun. 127 (2003) 373. [38] R.D. McMichael, J.J. Ritter, R.D. Shull, J. Appl. Phys. 73 (1993) 6946.

[39] P. Nisha, S. Savitha Pillai, A. Darbandi, M. Varma, K.G. Suresh, H. Hahn, J. Mater. Chem. Phys. 136 (2012) 66e74. [40] S. Tapas, I. Das, S. Banerjee, Appl. Phys. Lett. 91 (2007) 082511. [41] M.H. P., S.C. Yu, J. Magn. Magn. Mater. (2007) 308.