Accepted Manuscript Influence of crystallite size reduction on the magnetic and magnetocaloric properties of La0.6Sr0.35Ca0.05CoO3 nanoparticles R. Tlili, M. Bejar, E. Dhahri, A. Zaoui, E.K. Hlil, L. Bessais PII: DOI: Reference:
S0277-5387(16)30474-0 http://dx.doi.org/10.1016/j.poly.2016.09.044 POLY 12232
To appear in:
Polyhedron
Received Date: Revised Date: Accepted Date:
22 June 2016 15 September 2016 16 September 2016
Please cite this article as: R. Tlili, M. Bejar, E. Dhahri, A. Zaoui, E.K. Hlil, L. Bessais, Influence of crystallite size reduction on the magnetic and magnetocaloric properties of La0.6Sr0.35Ca0.05CoO3 nanoparticles, Polyhedron (2016), doi: http://dx.doi.org/10.1016/j.poly.2016.09.044
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Influence of crystallite size reduction on the magnetic and magnetocaloric properties of La0.6Sr0.35Ca0.05CoO3 nanoparticles R. Tlilia,*, M. Bejara, E. Dhahria, A. Zaouib, E.K. Hlilc and L. Bessaisd a
Laboratoire de Physique Appliquée, Faculté des Sciences de Sfax, B.P. 802, Université de Sfax, Sfax 3018,
Tunisie. b
Laboratoire de Physique Computationnelle des Matériaux – LPCM, Université Djillali Liabes de Sidi Bel-
Abbes, 22000 Sidi Bel-Abbes, Algerie c
Institut Néel, CNRS et Université Joseph Fourier, BP 166, F-38042 Grenoble cedex 9, France.
d
CMTR, ICMPE, UMR 7182, CNRS-UPEC 2 rue Henri Dunant, F-94320 Thiais, France
ABSTRACT In this paper, three samples of La0.6Sr0.35Ca0.05CoO3 with different crystallite sizes were prepared by a combination of the sol-gel and mechanical milling methods, with milling times tm of 0 (M0), 1 h (M1) and 2 h (M2). From the XRD and SEM analyses, it is confirmed that the grain size decreases with increased milling time. The value of the grain size observed from SEM is much larger than that observed from XRD analysis, which indicates that each particle observed by SEM consists of several crystallized grains. The reduction of the crystallite size (D) decreased the Curie temperature (TC) and magnetization values, and broadened the phase-transition region of the materials. The magnetocaloric effect was calculated in terms of the isothermal magnetic entropy change. The maximum value of the magnetic entropy change of the M0 sample, obtained from the M(µ0H) plot data, is | ∆ S Mmax |= 1.35 J/kg K for an applied magnetic field of 5 T. In addition, the µ0 H/M vs. M2 curves at temperatures around TC prove that the samples exhibit a second-order magnetic phase transition. The universal behavior obtained from the ∆SM variation curves confirmed the second-order transition for all the samples. Keywords: Nanoparticles; Cobaltites; Magnetic properties; Magnetocaloric effect; Universal curve Corresponding author: Riadh Tlili
[email protected] Tel: +21696589919; Fax: +21674676609
1. Introduction In recent years, magnetic nanoparticle systems have been important for potential applications in magnetic memory devices, refrigeration, sensors, biology, medicine and catalysts. These
1
systems have also been the subject of intensive research from both fundamental and application points of view [1-4]. Remarkable new phenomena observed in nano-materials have arisen from finite-size effects and inter-particle interactions. Currently, there is an ongoing debate and increased research intensity on the magnetism of the LaCoO3 system, due to the fascinating properties of compounds in this family and their possible multifunctional applications [5,6]. In the LaCoO3 compound, the ground state of the Co3+ ions is nonmagnetic with a low-spin configuration (LS, S = 0, t ), and these ions can be thermally
excited to an intermediate-spin state (IS, S = 1, t e ) at about 100 K.
The La1−xSrxCoO3 system has been studied by many groups [7-9]. The results indicate that replacing different ions in the A-site can cause structural distortion to different extents, leading to various degrees of deterioration of the ferromagnetic exchange interaction between the cobalt ions, and consequently affect their electronic state, magnetic and transport properties. The microscopic inhomogeneity affects the ferromagnetic interaction between the Co 3+ and Co 4+ ions, and the ferromagnetic clusters grow in size and number with increasing Sr content, which leads to an enhanced paramagnetic to ferromagnetic phase transition in the mixed-valence La1−xSrxCoO3 (0 < x ≤0.5) series. The Co 4+ ions have a 3d5 configuration with LS (S = 1/2, t ), IS (S = 3/2, t e ) and HS (S = 5/2, t e ). The theoretical effective magnetic moment (µeff) of Co3+ and Co4+ ions in the LS, IS, and HS states are 0, 2.83 and 4.90 µB and 1.73, 3.87, and 5.92 µB, respectively. Over the last decade, mixed valence manganites have been the subject matter of a large number of studies thanks to their exotic physical phenomena [10-13]. For doped cobaltites, there are relatively few studies. In this article, we prepared cobaltite samples with the nominal composition La0.6Sr0.35Ca0.05CoO3 using a combination of sol-gel and mechanical milling methods and studied the effect of the crystallite size substitution on their structural, magnetic and magnetocaloric properties.
2. Experimental The nano-sized crystalline La0.6Sr0.35Ca0.05CoO3 (M0, M1 and M2) compounds were prepared by the sol-gel method [14,15] using stoichiometric amounts of the La(NO3)3.6H2O, Sr(NO3)2, Ca(NO3)2 and Co(NO3)2.6H2O precursors. This method, based on citric acid as a complexing agent, is very effective for the synthesis of our samples and has the advantage of producing very fine powders of high homogeneity. The solution prepared by dissolving the stoichiometric amounts of the powder precursors in citric acid, ethylene glycol and distilled water was heated to 90 °C under constant stirring to eliminate excess water and obtain a
2
homogeneous solution. Afterwards, heating and stirring was continued until a gel was obtained, which took about 5 h. The gel was dried at 250 °C and then calcinated at 600 °C for 12 h. Thereafter, the powders were ground again, and using a uniaxial pressure system, pellets 10 mm in diameter and 1.5 mm in thickness were made and submitted to heat treatment at different temperatures (800, 900 and 1000 °C), for 12 h. These pellets were then employed for mechanical ball milling (using a Spex 8000M system) with a mass ratio of ball/powder = 5/1(1080 rpm). The milling time (tm) was varied from 0 h (M0), 1 h (M1) to 2 h (M2). The phase purity and homogeneity were determined by X-ray diffraction (Cu Kα radiation) at room temperature. The magnetic measurements were carried out with a BS2 magnetometer developed in the Louis Neel Laboratory of Grenoble.
3. Results and discussions The room-temperature X-ray diffraction (XRD) patterns of the samples under study, shown in Figs. 1.a and b, indicate that each sample has a single phase and can be indexed in the rhombohedral structure (R3 c space group). The broadening of the XRD lines corresponds to a decrease of the particle size. Additionally, the XRD patterns reveal that the peaks are identical and superimposable, except for the fact that the intensity of these peaks is different in these diagrams, which is due to the increase of the grinding time of the samples from 0 to 2 h. The data were analyzed by the Rietveld method using the Fullprof program [16]. The detailed results of the structural refinements are grouped in Table 1. We also have estimated the average crystallite size D from the XRD patterns using the Scherrer formula [17,18]: D=
.
(1)
where λ is the X-ray wavelength, is the diffraction angle for the most intense peak and β is defined as: = −
(2)
Here, m is the experimental full width at half maximum (FWHM) and s is the FWHM of a standard silicon sample. It was found that the crystallite size, summarized in Table 1, decreases with an increase in milling time, which is in good agreement with the previously obtained results [19]. Furthermore, the microstructures of the La0.6Sr0.35Ca0.05CoO3 (M0, M1 and M2) compounds were studied using scanning electron microscopy. Typical SEM micrographs of the three samples are displayed in Fig. 2. The closely packed grains occur in different polygonal forms. These micrographs show that the samples are composed of homogeneous particles which connect with each other. Moreover, the SEM photographs exhibit a decrease in the particle
3
size as the milling time increases (see Fig. 2 and Table 1). In comparison with the average crystallites sizes D, deduced from XRD data, the particle sizes provided by SEM micrographs are larger, which indicates that each particle observed by SEM consists of several crystallized grains. The magnetization and its derivative vs. temperature curves were measured under an applied magnetic field of 0.05 T for M0, M1 and M2 samples (Fig. 3 a and b). All the samples exhibit a clear transition from a paramagnetic to a ferromagnetic state with decreasing temperature. The value of TC was determined as the temperature where the first derivative of the magnetization with respect to temperature, (dM/dT)µ0H, reaches a minimum. The obtained values of TC are 250, 225 and 210 K, respectively, for the M0, M1 and M2 samples. To explain the magnetic properties described above and from the explanations of recent works [20,21], we believe that both the magnetization and TC reduction can be explained in terms of the core-shell model. This behavior is a consequence of a so-called magnetically dead layer arising due to a non-collinear spin arrangement at the surface of the crystallites. The evolution of the magnetization vs. the applied magnetic field (from 0 to 5 T) obtained at various temperatures for the M2 sample is illustrated in Fig. 4. As we can see, the curves M(µ0H) start to be linear only at high temperatures, well above the T C value. Moreover, the magnetization rises rapidly under the low magnetic field signature of ferromagnetic behavior but does not saturate, even in a field of 5 T. The same behavior was observed for the M0 and M1 samples (not shown here). We derived the Arrott plots (µ0H/M vs. M2) from M(µ0H) and the plotted results are shown in Fig. 5. An inspection of the sign of the slope of the isotherms of µ0H/M vs. M2 will give the nature of the phase transition: positive for second order and negative for first order [22-24]. These plots indicate a positive slope in the complete M2 range for all samples, in the vicinity of TC, which confirms that these samples exhibit a second-order ferromagnetic-paramagnetic phase transition. From the Maxwell relation, the entropy change induced by changing the magnetic field from 0 to µ0H is given by [25,26]:
!
µ"#
=$
%
(µ " #)
(
(3)
The magnetic entropy change ∆SM, which results from the spin ordering and induced by the variation of the magnetic applied field from 0 to µ0Hmax, is given by: 2 #345 ! #
ΔS (T, ΔH) = S (T, μH) − S (T, 0) = 0 "
μdH
(4)
4
Fig. 6 reveals the temperature dependence of the magnetic entropy change for all the samples up to 5 T, computed from Eq. 4 using the measured magnetization data. The magnetic entropy change is negative in the entire temperature range and is pronounced over a wide range of temperatures. Moreover, ∆SM increases and reaches a maximum value (∆Smax) when the temperature approaches the Curie temperature. This maximum changes with the applied field, because a high field can shift the transition temperature to a higher value due to the demagnetization field [27]. The different curves of the magnetic entropy change for all our synthesized polycrystalline samples as a function of temperature below 5 T show that the maximum magnetic-entropy value decreases from 1.35 J/kg K for M0 to 0.39 J/kg K for M2. These values are comparable with those of some cobaltites (such as Pr0.7(Ca0.8,Sr0.2)0.3CoO3 [28] and La0.6Sr0.35CoO3 [29]). As discussed above, the observed behavior is a consequence of a non-collinear spin arrangement at the surface layer of crystallites. Franco and Bonilla et al. have put forward a sensitive method to determine the order of the magnetic transition [30,31]. This method is used to compare the properties of several materials and to make extrapolations to temperatures and/or fields outside the accessible experimental range. Its phenomenological construction is based on the assumption that if such a universal curve exists, the equivalent points belonging to several ∆SM(T) curves measured up to different maximum applied fields should collapse onto the same point of the universal curve. The universal curve can be constructed by normalizing all the ∆SM curves using their respective maximum value ∆S Mmax and rescaling the temperature axis as:
{
θ = (TC - T) / (Tr 1 -TC ),T ≤ TC θ = (T - TC ) / (Tr 2 -TC ),T > TC
(5)
where Tr1 and Tr2 are the temperatures of the two reference points that, for the present study, have been selected as those corresponding to ∆SM(Tr1,2)=
∆ S Mmax . 2
The transformed curves of different compositions at several magnetic fields are plotted in Fig. 7. From these results, we can notice that all the data points collapse onto a single master curve, which confirms that the ferromagnetic-paramagnetic phase transition observed for our samples is of a second order. Based on the Landau theory of phase transitions, an attempt at theoretical modelling the magnetocaloric effect was done by Amaral et al. [32]. The Gibbs free energy is expressed as [33,34]:
5
1 1 G (T ,M) = G 0 + AM 2 + BM 4 − µ 0HM 2 4
(6)
The Landau coefficients A and B are the thermodynamic parameters, which depend on temperature. From the energy minimization, the magnetic equation of state is derived from this theory:
µ 0H = A (T ) + B (T ) M 2 M
(7)
It has been argued that a positive value of B is an indication of a second order transition [35]. The values of the Landau coefficients and their dependence on temperature can be obtained from experimental isothermal magnetization measurements of a polynomial fitting of µ0H/M vs. M2 Arrott plots. The variations of the A(T) and B(T) values are shown in Fig. 8. The parameter A(T) varies from negative to positive values with increasing temperature and the temperature corresponding to the value where it crosses zero is consistent with the TC value. The order of the magnetic transition is governed by the sign of B(T) at the transition: a first order transition takes place if B(T) < 0 while a second order occurs when B(T) ≥ 0. As B(TC) > 0, this leads to the conclusion that the transition is second order. On the other hand, for the nanocrystalline sample, both A(T) and B(T) have slight changes of slope near its two transitions.
4. Conclusion La0.6Sr0.35Ca0.05CoO3 (M0, M1 and M2) compounds were prepared by the sol-gel and mechanical ball milling methods with milling times, tm, of 0-2 h. The ratio of the ball and powder weight was 5:1 and the rotating speed was set to 1080 rpm. Rietveld analysis of the X-ray powder diffraction shows that the samples crystallize in the rhombohedral structure with the R3 c space group. The milling time leads to dramatic changes in the magnetic and magnetocaloric properties. The transition temperature from the ferromagnetic to paramagnetic state has been evaluated according to M(T) measurements, and it is found to be 250, 225 and 210 K for M0, M1 and M2, respectively. The magnetocaloric effect was estimated, in terms of the isothermal magnetic entropy change (−∆SM), using the M(T,µ0H) data and employing the thermodynamic Maxwell equation. We have found quite large values for the magnetic entropy change around TC, for the M0 sample, around 1.35 J/kg K for µ0∆H = 5 T. Finally, the fact that the magnetic entropy change curves at different applied fields collapse onto a universal curve proves that the studied cobaltite undergoes a second order magnetic phase transition.
6
ACKNOWLEDGEMENTS This work is supported by the Tunisian Ministry of Higher Education and Scientific Research and the Moroccan, Algerian and French Ministries of Higher Education and Research of the PHC Maghreb 15MAG07 collaboration, within the framework of the Franco-Maghrebin collaboration.
REFERENCES [1] B. Roy, S. Das, Appl. Phys. Lett. 92 (2008) 233101. [2] R. von Helmolt, J. Wecker, B. Holzapfel, L. Schultz, K. Sanwer, Phys. Rev. Lett. 71 (1993) 2331. [3] M. Smari, I. Walha, E. Dhahri, E.K. Hlil, J. Alloy. Compd. 579 (2013) 564. [4] Q.A. Pankhurst, J. Connolly, S.K. Jones, J. Dobson, J. Phys. D: Appl. Phys. 36 (2003) R167. [5] S. Yamaguchi, Y. Okimoto, H. Taniguchi, Y. Tokura, Phys. Rev. B 53 (1996) R2926. [6] T. Saitoh, T. Mizokawa, A. Fujimori, Phys. Rev. B 55 (1997) 4257 [7] T. Saitoh, T. Mizokawa, A. Fujimori M. Abbate, Y. Takeda, M. Takano, Phys. Rev. B 56 (1997) 1290. [8] J. Wu, C. Leighton, Phys. Rev. B 67 (2003) 174408. [9] K. Yoshii, H. Abe, Phys. Rev. B 67 (2003) 094408. [10] R. Tlili, R. Hammouda, M. Bejar, E. Dhahri, J. Magn. Magn. Mater. 386 (2015) 81. [11] Y. Xu, M. Meier, P. Das, M.R. Koblischka, U. Hartmann, Crystal Engineering 5 (2002) 383. [12] A. Rostamnejadi, M. Venkatesan, P. Kameli, H. Salamati, J.M.D. Coey, J. Magn. Magn. Mater. 323 (2011) 2214. [13] R. Tlili, A. Omri, M. Bejar, E. Dhahri, E.K. Hlil, Ceram. Int. 41 (2015) 10654. [14] S.K. Jaiswal, J. Kumar, J. Alloy. Compd. 509 (2011) 3859. [15] H. Gharsallah, M. Bejar, E. Dhahri, E.K. Hlil, M. Sajieddine, J. Magn. Magn. Mater. 401 (2016) 56. [16] H.M. Rietveld, J. Appl. Crystallogr. 2 (1969) 65. [17] F. Elleuch, M. Triki, M. Bekri, E. Dhahri, E.K. Hlil, J. Alloy. Compd. 620 (2015) 249. [18] R. Dudric, F. Goga, M. Neumann, S. Mican, R. Tetean, J. Mater Sci 47 (2012) 3125. [19] T.A. Ho, T.D. Thanh, T.L. Phan, S.K. Oh, S.C. Yu, J. Supercond Nov Magn 28 (2015) 891.
7
[20] H. Baaziz, A. Tozri, E. Dhahri, E.K. Hlil, Chem. Phys. Lett. 625 (2015) 168. [21] L.E. Hueso, P. Sande, D.R. Miguens, J. Rivas, F. Rivadulla, M.A. Lopez-Quintela, J. Appl. Phys. 91 (2002) 9943. [22] S. Chandra, A. Biswas, M.H. Phan, H. Srikanth, J. Magn. Magn. Mater. 384 (2015) 138. [23] R. Skini, A.Omri, M.Khlifi, E.Dhahri, E.K. Hlil, J. Magn. Magn. Mater. 364(2014)5 [24] N. Khan, P. Mandal, K. Mydeen, D. Prabhakaran, Phys. Rev. B 85 (2012) 214419 [25] R. Tlili, A. Omri, M. Bekri, M. Bejar, E. Dhahri, E.K.Hlil, J. Magn. Magn. Mater. 399 (2016)143. [26] T.L. Phan, N.T. Dang, T.A. Ho, T.V. Manh, T.D. Thanh, C.U. Jung, B.W. Lee, A.T. Le, A.D. Phan, S.C. Yu, J. Alloy. Compd. 657 (2016) 818. [27] L. Si, Y.L. Chang, J. Ding, C.K. Ong, B. Yao, Appl. Phys. A 77 (2003) 641. [28] I.G. Deac, A. Vladescu, I. Balasz, A. Tunyagi, R. Tetean, Acta Phys. Pol. A 120 (2011) 306 [29] R. Li, P. Kuma, R. Mahendiran, J. Alloy. Compd. 659 (2016) 203. [30] V. Franco, A. Conde, J.M. Romero-Enrique, J.S. Blázquez, J. Phys. Condens. Matter 20 (2008) 285207. [31] C.M. Bonilla, J.H. Albillos, F. Bartolomé, L.M. García, M.P. Borderías, V. Franco, Phys. Rev. B 81 (2010) 224424. [32] V.S. Amaral, J.S. Amaral, J. Magn. Magn. Mater. 272 (2004) 2104. [33] Ah. Dhahri, M. Jemmali, K. Taibi, E. Dhahri, E.K. Hlil, J. Alloy. Compd. 618 (2015) 488 [34] S. Liu, Physica B 456 (2015) 227 [35] T. Izgi, V.S. Kolat, N. Bayri, H. Gencer, S. Atalay, J. Magn. Magn. Mater. 372 (2014) 112.
8
Figure 1: 1. (a) X-ray diffraction of the La0.6Sr0.35Ca0.05CoO3 compound with different milling times, 0 h (M0), 1 h (M1) and 2 h (M2). (b) Rietveld plot of XRD data for the polycrystalline M2 sample. The points are the observed profile; the solid line is calculated. Tick marks indicate the position of allowed Bragg reflections.
9
Figure 2: Scanning electron micrographs (SEM) of La0.6Sr0.35Ca0.05CoO3 (M0, M1 and M2) samples.
10
Figure 3: (a) Temperature dependence of the magnetization for La0.6Sr0.35Ca0.05CoO3 (M0, M1 and M2) samples at a magnetic field of 0.05 T. (b) The dM/dT vs. T plots.
11
Figure 4: The field dependence of the magnetization for the La0.6Sr0.35Ca0.05CoO3 (M2) sample measured at different temperatures around TC.
12
Figure 5: Arrott plot isotherms of µ0H/M vs. M2 at different temperatures for the La0.6Sr0.35Ca0.05CoO3 (M2) sample.
13
14
Figure 6: Magnetic entropy change versus temperature for the La0.6Sr0.35Ca0.05CoO3 (M0, M1 and M2) samples for various magnetic applied field changes.
15
Figure 7: The master curve behavior of the curves as a function of the rescaled temperature.
16
Figure 8: Temperature dependence of the parameters A(T) and B(T) of the La0.6Sr0.35Ca0.05CoO3 (M2) sample.
17
Table 1: Crystallographic data for the La0.6Sr0.35Ca0.05CoO3 compound with different milling times, 0 h (M0), 1 h (M1) and 2 h (M2), from the Rietveld refinement of X-ray diffraction (XRD) data.
62
tm D(XRD) (h) (nm)
D(SEM) (µm)
a (Å)
b (Å)
M0
0
26
1.326
5.4389
5.4389 13.2411 339.2102 1.45
M1
1
11
1.047
5.4696
5.4696 13.2818 344.1079 1.57
M2
2
9
0.933
5.4750
5.4750 13.2942 345.1099 1.90
Sample
c (Å)
V (Å3)
18
19
Scanning electron micrographs (SEM) of La0.6Sr0.35Ca0.05CoO3 (M0, M1 and M2) samples.
20