Microstructural, magnetic and transport properties of La0.5Pr0.2Pb0.3-xSrxMnO3 manganites

Microstructural, magnetic and transport properties of La0.5Pr0.2Pb0.3-xSrxMnO3 manganites

Accepted Manuscript Microstructural, magnetic and transport properties of La0.5Pr0.2Pb0.3−xSrxMnO3 manganites M.-L. Craus, A. Kh. Islamov, E.M. Anitas...

753KB Sizes 0 Downloads 63 Views

Accepted Manuscript Microstructural, magnetic and transport properties of La0.5Pr0.2Pb0.3−xSrxMnO3 manganites M.-L. Craus, A. Kh. Islamov, E.M. Anitas, N. Cornei, D. Luca PII: DOI: Reference:

S0925-8388(14)00030-9 http://dx.doi.org/10.1016/j.jallcom.2014.01.002 JALCOM 30333

To appear in: Please cite this article as: M.-L. Craus, A.K. Islamov, E.M. Anitas, N. Cornei, D. Luca, Microstructural, magnetic and transport properties of La0.5Pr0.2Pb0.3−xSrxMnO3 manganites, (2014), doi: http://dx.doi.org/10.1016/j.jallcom.

2014.01.002

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.

Microstructural, magnetic and transport properties of La0.5 Pr0.2 Pb0.3−x Srx MnO3 manganites M.-L. Crausa,b , A. Kh. Islamova,c , E. M. Anitasa,d,∗, N. Corneie , D. Lucaf a Joint Institute for Nuclear Research, Dubna, Russia Institute of Research and Development for Technical Physics, Iasi, Romania c Laboratory of Advanced Research of Membrane Proteins, Moscow Institute of Physics and Technology, Dolgoprudniy, Russia d Horia Hulubei National Institute of Physics and Nuclear Engineering, Bucharest-Magurele, Romania e “Al. I. Cuza” University, Chemistry Department, Iasi, Romania f “Gh. Asachi” Technical University, Faculty of Materials Science and Engineering, Iasi, Romania b National

Abstract The most interesting and studied materials for practical applications of colossal magnetorsistance effect are rare earth manganites with general formula RMnO3 (where R is a rare/alkaline earth element). The coexisting of competing phases in manganites, such as metallic ferromagnetic, charge ordered, antiferromagnetic insulating and ferromagnetic insulating phases, determines an important change of magnetic and transport properties with the microstructure. In this paper, we report on the correlations between microstructural, magnetic and transport properties at temperatures ranging from 263 to 343 K of La0.5 Pr0.2 Pb0.3−x Srx MnO3 manganites synthesized by ceramic technology. The microstructure is studied using X-ray diffraction (XRD), small-angle X-ray scattering (SAXS) and small-angle neutron scattering (SANS) at Sr concentrations x = 0.00, 0.05, 0.15 and 0.20. SAXS and SANS data show the formation of magnetic nanodomains in the mosaic blocks, at temperatures higher than Curie temperature TC . SANS data reveal the shape and concentration of magnetic nanodomains, and their dependency on temperature. The La0.5 Pr0.2 Pb0.3−x Srx MnO3 manganites crystallize as cubic structure P m¯3m (x = 0.00 and x = 0.05) or as rhombohedral structure R¯3c (x = 0.15 and x = 0.20). We found that transport phenomena at temperatures higher than TC are greatly influenced by nanodomains concentration and their shape. We show that about room temperature manganites with x = 0.05 and x = 0.15 have large intrinsic resistance maxima, generating an important magnetoresistance effect. Keywords: Manganites, X-ray diffraction, small-angle X-ray scattering, small-angle neutron scattering, magnetic structure, transport properties 1. Introduction The RMnO3 manganites, where R is a rare/alkaline earth element, exhibit together with a huge change of the resistivity at application of a magnetic field, a very rich phase diagram, where several competing phases coexist, such as metallic ferromagnetic, charge ordered, antiferromagnetic insulating and ferromagnetic insulating

phases. The variation of the transport and magnetic properties is strongly influenced by the manganites structure at nano- and microscales [1], which in turn, is determined by the nature of the rare/alkaline earth elements.

author Email addresses: [email protected] (M.-L. Craus), [email protected] (A. Kh. Islamov), [email protected] (E. M. Anitas), [email protected] (N. Cornei), [email protected] (D. Luca)

For certain types of manganites the transport characteristics and/or magnetic properties are greatly influenced by other factors, such as the doping degree or the intensity of the applied magnetic field. For example, the doping degree in A1−x Srx MnO3 manganites induces a strong dependence between the structural changes, the tolerance factor and Curie temperature TC [2]. The structural changes are connected with the change of the band-

Preprint submitted to Journal of Alloys and Compounds

January 8, 2014

∗ Corresponding

width w ∝ cos(1/2(π − ∠MnOMn))/d3.5 MnO , where ∠MnOMn is the angle of Mn − O − Mn bond and dMnO is the length of Mn − O bond, and implicitly is proportional with the transfer probability of the 4+ eg electron between Mn3+ and cations [2]. √ Mn The tolerance factor t = dAO / 2dMnO , where dAO is the average length of A − O bonds, plays an important role for crystalline structure [2]. The substitution of Ca with Sr in La0.7 Ca0.3−x Srx MnO3 , induces a variation of structural parameters and also a change of the transition temperature from metallic to insulator state [3]. These manganites are characterized by a nanoscale phase separation, which leads to the appearance of two competing phases in an electronic inhomogeneous compound. The magnetic contrast between the ferromagnetic metallic phase, on one hand, and the ferromagnetic insulating and antiferromagnetic insulating phases, on another hand, allowed Saurel et al. to observe the dependence of size and shape of magnetic domains with the intensity of the applied magnetic field at low temperatures [4]. The relationship between nano- and microstructural characteristics, concentration of magnetic phase and the intensity of the applied magnetic field, on the physical properties of manganites have been only partially elucidated. Near transition from metallic to insulator state of Pr0.7 Ca0.3 MnO3 manganites, Saurel et al. have found a large range of heterogeneities, by using SANS experiments, and have shown that the insulator to metal state transition is induced by the percolation of ferromagnetic domains [5]. Viret et al. carried out polarized small angle neutron scattering studies on Pr0.67 Ca0.33 MnO3 monocrystal and observed at low temperature a charge ordered phase, an orbital disordered weakly magnetic phase, and below 110 K a ferromagnetic phase [6]. The disordered magnetic phase is characterized by the presence of magnetic filaments, which grow when a magnetic field is applied [6]. Small-angle neutron scattering (SANS) studies confirmed the presence of a short range order, implicitly the presence of magnetic clusters or nanodomains at temperatures higher than TC for Pr0.55 Ca0.45 Mn1−y Cry O3 (y = 0.0, 0.03 and 0.06), La0.7 Pb0.3 Mn1−x Fex O3 (0 < x < 0.2) and (Nd1−x Tbx )0.55 Sr0.45 MnO3 (x = 0.6 and x = 0.7) systems [7, 8, 9]. The aim of this paper is to establish a correlation between the concentration of magnetic phase, the average size and shape of magnetic domains and crystalline characteristics, on the transport and magnetic properties at temperatures higher than

TC , for La0.5 Pr0.2 Pb0.3−x Srx MnO3 (at Sr concentration x = 0.00, 0.05, 0.15 and 0.20) manganites synthesized by ceramic technology. To this end we have used X-ray diffraction (XRD), small-angle X-ray scattering (SAXS) and small-angle neutron scattering at temperatures ranging from 263 to 343 K. The nano- and microstructural data are used to explain the variation of resistance with temperature and Sr concentration in a very low magnetic field. 2. Experimental methods The La0.5 Pr0.2 Pb0.3−x Srx MnO3 manganites were obtained by ceramic technology, starting from a suitable mixture of La2 O3 (Alfa Aesar, 99.9 %), Pr6 O11 (Aldrich, 99.9 %), PbO2 (Aldrich, 99.999 %) and MnO2 (Aldrich, 99 %) at FLNP JINR, Dubna, Russia. The samples were treated at 800 ◦ C for 15 hours, grounded and finally sintered at 1250 ◦ C in air, in a closed tube of quartz, to avoid lead evaporation. All the samples were investigated by XRD at FLNP - JINR Dubna in order to determine the phase composition, lattice constants, oxygen and rare earth atoms positions in the unit cell, tolerance factor, average size of mosaic blocks and lattice microstrains. Magnetization measurements as a function of temperature were performed with a Foner type magnetometer between 77 and 320 K. Transport phenomena were investigated using four probe method, at NIRDTP Iasi, Romania, in a very small magnetic field (H = 55 Oe). SAXS and SANS measurements were performed in the temperature range from 263 to 343 K in the absence of external magnetic field. SAXS experiments were carried out on a Rigaku X-ray instrument with high-speed Cu rotating anode (SMAXS3000 Point SAXS system, at MIPT, Dolgoprudniy, Russia) using a standard transmission configuration. An X-ray wavelength of λ = 1.54 ˚ A was used, resulting a momentum transfer Q in the range of 0.025 − 0.5 ˚ A−1 , where Q = (4π/λ) sin(θ/2) and θ is the scattering angle. SANS experiments were performed at the time - of - flight YuMO spectrometer situated at high flux pulse IBR-2 reactor, JINR, Dubna, Russia [10]. The experiments were carried out at sample-to-detector distances of 5.28 and 13.04 m, resulting in a Q range of 0.007 − 0.3 ˚ A−1 . The diameter of the sample in the beam was 14 mm. The measured neutron scattering spectra were corrected for the transmission and thickness of the sample, background scattering on the film substrate 2

and on the vanadium reference sample using SAS software [11], yielding a neutron scattering intensity in absolute units of cm−1 .

and therefore decreasing the tolerance factor t. We attribute the non monotonous variation in magnetization to the change of structure from cubic (x = 0.00 and x = 0.05; t = 1.000) to a rhombohedral one (x = 0.15 and x = 0.20; t < 1.000) (Fig. 1). This structural transition is accompanied by a sudden increase of remanent magnetization due to the change of exchange interaction between Mn cations via O anions. On another hand, within the same structure the lowfield magnetization decreases with the increase of average size of A sites, in agreement with variation of remanent magnetization with tolerance factor, as shown in Ref. [12]. TC has a maximum for the Sr concentration x = 0.15, which indicates a maximum of the interaction between the magnetic moments associated to two Mn neighboring cations (Fig. 1).

3. Results and discussions 3.1. Structure According to XRD data obtained at room temperature, the samples contain only perovskite phase: (1) P m¯3m, with La/Pr/Sr or Pb on (1b) sites (1/2, 1/2, 1/2), Mn on (1a) sites (0, 0, 0) and O on (3d) sites (1/2, 1/2, 0) for x = 0.00 and x = 0.05, or (2) R¯3c, with La/Pr/Sr or Pb on (6a) sites (0, 0, 0.25), Mn on (6b) sites (0, 0, 0) and O on (18e) (X, 0, 0.25), for x = 0.15 and x = 0.20. The cell parameters, average size of mosaic blocks, microstrains, distances MnO and tolerance factor have been obtained through the profile matching stage of the Rietveld refinement (Tabs. 1 and 2). The data show an increase of unit cell with Sr concentration in the samples within each type of structure (Tab. 1).

10.0

x=0.00 x=0.05

, (uem/g)

7.5

Table 1: Variation of the lattice constants (a, b, c), average size of mosaic blocks (D) and microstrains (²) for La0.5 Pr0.2 Pb0.3−x Srx MnO3 manganites.

x

a = b (˚ A)

c (˚ A)

D (˚ A)

²

0.00 0.05 0.15 0.20

3.880 3.896 5.513 5.524

3.880 3.896 13.355 13.385

315 551 600 480

0.0010 0.0028 0.0005 0.0007

x=0.15 x=0.20

5.0

2.5

0.0 100

x

p (µB /f.u.)

dMnO (˚ A)

VA

t

3.03 3.13 3.17 2.80

1.9402 1.9462 1.9538 1.9634

0.022 0.024 0.025 0.016

1.000 1.000 0.995 0.963

T, (K)

300

400

Figure 1: Variation of magnetization (σ) with temperature (T ) at four values of the Sr concentration (x) for La0.5 Pr0.2 Pb0.3−x Srx MnO3 manganites at H = 55 Oe.

Table 2: Variation of the molar magnetization (p), MnO distances (dMnO ), voids concentration on A and B sites (VA = VB ) and tolerance factor (t) for La0.5 Pr0.2 Pb0.3−x Srx MnO3 manganites.

0.00 0.05 0.15 0.20

200

3.3. SAXS and SANS measurements We have performed two types of SAXS experiments. First, we have considered variation of temperature at a given Sr concentration, and second, variation of Sr concentration at a given temperature. Figure 2 shows the SAXS data for x = 0.00 and three temperatures (upper panel) and at T = 283 K and four Sr concentrations (lower panel). In both cases, at low values of the momentum transfer (Q < 0.1 ˚ A−1 ), one can observe the scattering intensity following a power-law behavior with scattering exponent −4 (Porod law), while at high momentum transfer (Q > 0.1 ˚ A−1 ), the background

3.2. Magnetic measurements Zhou et al. [12] have shown that Pr substitution in La0.7−x Prx Sr0.3 MnO3 and La0.7−x Prx Ca0.3 MnO3 increases both the remanent magnetization and coercive field with decreasing the average radius of A sites hrA i, 3

is attained. Since X-ray patterns are not sensitive to magnetic moments, the power-law dependence of SAXS corresponds to scattering from manganite crystallites.

The Porod-law region corresponds to the nuclear scattering of the manganite granular structure of the powder, and is independent of the temperature. The specific surface is related to the scattering intensity through [14]:

x = 0.00

10

I(Q) = 2πρ2 Q−4 Scr /Vcr = 2πρ2 Q−4 3/Rcr ,

T = 283K

1

where ρ = 3.34 ∗ 1010 cm−2 is the scattering length 2 3 density, Scr = 4πRcr and Vcr = 4π/3Rcr are surface area and respectively, the volume of manganite crystallites. Then, using Eq. (2) the average radius of crystallites are 4.0, 4.6, 3.4 and 3.8 µm for samples with x = 0.00, 0.05, 0.15 and, respectively x = 0.20. However, the scattering at high Q region (which is absent in SAXS data) changes with the temperature, representing the contribution of the magnetic domains. We are interested here in SANS at high Q, since in this regime the scattering intensity is associated with magnetic scattering. By subtracting the contribution of nuclear scattering from total scattering we obtain scattering from magnetic domains. This allows us to obtain the volume fraction of the magnetic phase (φ), using the Porod invariant [14] Z Qmax 2 2π ρ2mag φ = I(Q)Q2 dQ, (3) 3 Qmin

T = 313K T = 333K

I(Q),

(a.u.)

10

10

0

Q

-4

-1

10 10

10

T = 283K x = 0.00

Q 10

-1

1

x = 0.05

-4

x = 0.15 x = 0.20

0

-1

10

-1

Q, (

Å

-1

)

where ρ2mag = 3.24 ∗ 1020 cm−4 is the magnetic contrast factor for the manganties [4, 5] and the factor 2/3 is needed in order to average the scattering of non-oriented magnetic objects. Figure 4 shows the magnetic scattering for samples with x = 0.05, 0.15 and 0.20 at various temperatures in the range 263−343 K. Since the scattering for x = 0.00 shows a qualitatively similar behavior, the corresponding SANS intensities are not shown here but their corresponding volume fractions are given in Fig. (5a). The scattering exponent α corresponding to magnetic scattering (Fig. 5b) allows us to describe the approximate shape of magnetic nanodomains. For x = 0.00 and x = 0.05 samples, their values are found in the range 1 < α < 1.6, while for x = 0.15, 1 < α ≤ 2.2. A value of α = 1 is specific to long 1D rod-like magnetic domains, α = 2 corresponds to flat 2D disk-like domains, and 2 < α < 3 corresponds to scattering from 3D mass fractal structures [15]. The volumes of magnetic nanodomains at various Sr concentrations and temperatures were calculated

Figure 2: SAXS intensities for La0.5 Pr0.2 Sr0.3−x Pbx MnO3 manganites. (a) x = 0.00 and various temperatures; (b) T = 283 K and various Sr concentrations.

SANS technique can be used to determine the contribution of the magnetic scattering without the additional use of the external magnetic field and polarized neutrons [13]. Figures 3a and 3b show representative SANS curves at x = 0.00 and x = 0.15, each concentration being measured at four different temperatures. As compared with X-ray scattering, SANS data show a different behavior due to appearance of supplemental contribution of magnetic scattering. A general characteristic of SANS curves for all Sr concentrations is the existence of two distinct regions of scattering. For Q < 0.06 ˚ A−1 the scattering intensity follows a Porod-law, while for Q > 0.06 ˚ A−1 the scattering intensity decays, on a double logarithmic scale, as a simple power-law I(Q) ∝ Q−α

(2)

(1)

where the scattering exponent α take values between 1 and 2.2. 4

T = 263K

2

by use of ATSAS package (GNOM and DAMMIF programs) [16, 17] (Fig. 5c). In order to extract the information about the characteristic sizes of the magnetic nanodomains we fitted our data using the cylinder model of radius R and height H given by [18]

T = 293K

10

-4

-4

Q

Q

0

10

-1.54

-1.25

Q

Z

Q

-1

(cm )

I(Q) = A 0

I(Q),

T = 343K

T = 323K

2

-4

-4

Q

Q

0

10

-1

-1

Q

Q

-2

10

-2

-1

10

-2

10

(a)

-1

10

Q, (

Å

-1

10

)

T = 263K

T = 293K

2

10

-4

-4

Q

-2

Q

Q

I(Q),

-1

(cm )

Q -2.2

0

10

-2

10

T = 343K

T = 323K

2

10

-4

-4

Q

Q

0

10

-1.4

-1

Q

Q

-2

10

-2

10 (b)

-1

10

-2

-1

10

Q, (

Å

-1

Λ21 (QR

p

1 − x2 )S 2 (QHx/2)dx + B,

(4) where Λ1 (t) = 2J1 (t)/t, S(t) = sin(t)/t and J1 (t) is the spherical Bessel function [18]. We found that for x = 0.00, 0.05 and 0.20 at T = 343 K, we have α = 1. Using Eq. 4 we obtain that the magnetic nanodomains have a 1D rod-like shape with the radius R in the range 2.5−5 ˚ A and the height H in the range 40 − 60 ˚ A. For these concentrations the general characteristic is that a decreasing of temperature determines a slight increasing of the scattering exponent α (Fig. 5b), which indicates that the rod-like magnetic structures expand to form wormlike structures. For temperatures lower than 263 K we may expect an increase of scattering exponent α and formation of higher dimensionality magnetic structures. However, an interesting behavior can be observed for x = 0.15 sample, where we obtain the scattering exponent α ≈ 2 already for T = 293 K, which indicates the formation of 2D disk-like structures. Using again Eq. (4) we find the disk thickness 5 ˚ A which is insensitive to the disk radius R. The latter one is difficult to determine with good accuracy due to insufficient data at low Q values. In order to overcome this problem and to estimate the radius of disks we used magnetic domains evaluated by DAMMIF program [17] and divided by the disk thickness (5 ˚ A). At the same temperature T ' 293 K we obtained the radius p R of magnetic disk about 170 ˚ A, where R = V /πH and the volumes V were taken from data in Fig. (5c). At T = 263 K the radius increases up to 230 ˚ A. Therefore, decreasing the temperature from 293 K to 263 K and taking into account that the height (for α ≈ 1) / thickness (for α ≈ 2) remains constant at about 5 ˚ A and the lattice constants, given in Table 1, are a = b ≈ 5.5 ˚ A and c = 13 ˚ A, it turns out that the 1D magnetic domains grow in the “c” direction of the unit cell. Although the intensity at Q → 0 does not change significantly, a more detailed description would include the possibility of formation, for α = 2.2, of 3D magnetic domains of mass fractal type with fractal dimen-

-2

10

10

1

10

)

Figure 3: SANS intensities for La0.5 Pr0.2 Sr0.3−x Pbx MnO3 at various temperatures, for samples with: (a) x = 0.00, and (b) x = 0.15.

5

sion 2.2 (loosely branched magnetic domains which go out of the disk plane). At an even lower Qscale (≤ 1.5 ∗ 10−2 ˚ A−1 ) these structures form microscopic magnetic domains below TC , and respectively superparamagnetic domains for T > TC (Fig. 4b). The microscopic magnetic domains have a radius bigger than 0.5 µm, as indicated by the Porod law I(Q) ∝ Q−4 at low Q values (Fig. 4b). 10

0

283K

-1.3

10

ρ=

263K

x = 0.2

Q 10

rectly shows the influence of the nano and macroscopic magnetic domains. For a given temperature the resistance of the samples should increase when concentration of magnetic (nano)domains decreases or when the defects concentration increase. Qualitatively, the resistivity ρ increases when the number of electrons which participates at the conduction decreases, according to [19]

1/3

303K 343K

-2

Q, ( 10

Å

-1

10

)

-1

2

-4

x = 0.15

10

-2

0

Q

I(Q),

(cm

-1

)

Q

10

-2

10

10

10

10

10

-2

Q, (

Å

-3.5

1

Q

-1

)

10

-1

x = 0.05 -1.6

0

Q

-1

-2

10

-2

Q, (

Å

-1

)

10

(5)

where a ≈ Vcell is the lattice constant, C is the Mn4+ fraction, e is the electron charge, ²0 is the dielectric constant of the vacuum, ²r is the relative dielectric constant of the material, and Vcell is the unit cell volume. The resistance of the samples are determined by their chemical composition, the crystalline/magnetic structure and the concentration of magnetic domains, with metallic or insulating behavior. Figure 6 shows the variation of the resistance with temperature at all four values of Sr concentrations. We can clearly distinguish several regions which provide information about the component phases of the system. All samples are characterized by the appearance of large maxima (extrinsic component of resistance) at Tex (Fig. 6), which is a result of the presence of crystalline regions with a large concentration of defects. Taking into account magnetization data in intense magnetic fields and supposing a collinear arrangement of magnetic moments, we calculated the cation distribution and the voids concentration on A and B sites (Tab. 2). We suppose that the concentration of the voids is the same on both type of sites. There is a small variation of the voids concentration with Sr concentration, which does not influence the resistance of the samples. From the width of diffraction maxima we obtained using FullProf code [20], the average size of the mosaic blocks and microstrains (Tab. 1). The average size of magnetic domains in a very low magnetic field is related to the size of crystallites and to defects concentration. The average size of mosaic blocks varies between about 320 and 600 ˚ A diameter, with a maximum for x = 0.15, which corresponds also to a minimum of microstrains (Tab. 1). Performing a comparison between the microstrains and the resistance of the samples at temperatures around 250 − 270 K we observe that: 1) large extrinsic resistance are associated to cubic

323K

-1

ahkT e2 ²0 ²r C

-1

Figure 4: Magnetic scattering intensities for La0.5 Pr0.2 Sr0.3−x Pbx MnO3 with x = 0.20, 0.15 and 0.05 at various temperatures (263 − 343 K).

Formation of disk-like and 3D mass fractal structures for other Sr concentrations (x = 0.00, 0.05 and x = 0.20) may be possible, but at much lower temperatures. 4. Transport phenomena The data obtained for variation of resistance with temperature at relatively low magnetic field, indi6

0.05

phase, characterized by large microstrains (Fig. 6 and Tab. 1); 2) smaller extrinsic resistance are associated with rhombohedral phase, characterized by small values of microstrains (Fig. 6 and Tab. 1). At higher temperatures the resistance behavior is dominated by the intrinsic component for the samples with x = 0.05 and x = 0.15 and are determined by the microstrains values and the magnetic phase concentration (Figs. 5a and 6). In agreement with the observed values of magnetization and SANS data concerning nanomagnetic phase concentration, we concluded that a large amount of magnetic metallic phase of the samples with x = 0.15 and x = 0.20 (rhombohedral phase) is formed by large magnetic domains. Macroscopic magnetic domains contribute to the remanent magnetization, but have no contribution to SANS data in the investigated Q-range. This is in agreement with the difference between the resistance of cubic and, respectively, rhombohedral phases. At lower temperatures, in the region between 77 K and the lower metal-insulator transition temperature (Tex ) the manganites have a metallic behavior. We consider that the main sources which contribute to the appearance of large maximum of the resistance are: 1) the increase of resistance of metallic regions with temperature; 2) the decrease of magnetic phase concentration and the increase of insulator phase concentration with temperature and 3) the decrease of resistance of insulator phase with temperature. This model can be considered also for the intrinsic maxima at Tin = 293 K, where for x = 0.05 and x = 0.15 we obtain very pronounced maxima of the intrinsic component of the resistance, which arise due to the crystalline regions with the largest average volume of coherent domains (Tab. 1, Fig. 6) and due to the volume concentrations of largest magnetic nanodomains (Fig. 5a). On the other hand, we observed a net difference between the resistance of the cubic samples, which have a maximum value of resistance about 60 Ω, and those with the rhombohedral structure, where was observed a maximum of resistance at about 5 Ω (Fig. 6). For the sample with x = 0.15 we can associate this behavior with the presence of large magnetic domains about room temperature (Fig. 1). The sample corresponding to x = 0.20 has a much smaller magnetic nanodomains concentration and a larger concentration of insulator phase (Figs. 1 and 5a) as compared with the sample corresponding to x = 0.15. To the difference in the behavior of the resistance

x=0.00 x=0.05 x=0.15 x=0.20

0.04

0.03

260

280

260

280

(a)

300

320

340

300

320

340

320

340

T, (K)

2.0

1.5

1.0

(b)

T, (K)

6

10

5

V, (Å3)

10

4

10

3

10

(c)

260

280

300

T, (K)

Figure 5: Variation with temperature T and Sr concentration x, in La0.5 Pr0.2 Sr0.3−x Pbx MnO3 manganites, of: (a) volume fraction φ of the nanomagnetic phase, obtained from Eq. (3); (b) scattering exponent α, obtained from fitting with Eq. (1); (c) volume of magnetic clusters V , obtained using DAMMIF program.

7

can contribute also the shape of the magnetic nanodomains: for the sample with x = 0.05 they are mainly 1D rod-like, while for x = 0.15 they are 2D disk-like (Fig. 5b). The concentration of rod-like magnetic nanodomains for x = 0.05 is larger than that corresponding to the samples with x = 0.00 and x = 0.20 (Fig. 5a). The volume of the magnetic domains ant their shape determine the appearance of a narrow maximum for the sample corresponding to x = 0.05. The variation of the resistance with temperature between 240 − 300 K is in agreement with SANS data concerning nanomagnetic phase concentration and the magnetization data in the same range of temperatures (Figs. 5a and 1). The resistance of the sample with x = 0.20 decreases in the range 200 − 300 K due to decrease of the insulator phase resistance with temperature. 60 50 40

x=0.0 x=0.05

magnetic nanodomains and their 1D rod-like shape determine the appearance of a narrow maximum. We expect that the extension of the measured spectra at Q < 0.007 ˚ A−1 in SANS experiments would lead to an even better agreement between the SANS and magnetic and transport measurements. The magnetic measurements give the total magnetic moment including together nano- and macromagnetic domains. However, from the performed SANS measurements we obtain the magnetic characteristics and the structure corresponding to the nanomagnetic domains, but only few qualitative data concerning micromagnetic domains. 5. Conclusions The correlations between the microstructural, magnetic and transport properties of La0.5 Pr0.2 Pb0.3−x Srx MnO3 manganites synthesized by ceramic technology have been investigated using XRD, SAXS, SANS, resistance and magnetic measurements at Sr concentrations x = 0.00, 0.05, 0.15, 0.20, and at temperatures ranging from 263 to 343 K. XRD data show that, with increasing the Sr concentration, the structure changes from cubic space group P m¯3m (x = 0.00 and x = 0.05) to rhombohedral space group R¯3c (x = 0.15 and x = 0.20) and the maximum average size of the mosaic blocks is attained at x = 0.05, and respectively at x = 0.15. The transition from cubic to rhombohedral structure leads to a change of the exchange interaction. At the transition point, the remanent magnetization suddenly increases. Within the same structure the remanent magnetization decreases with increasing Sr concentration. SAXS and SANS data show the formation in the crystallites blocks of magnetic nanodomains, at temperatures higher than TC , their concentration ranging from 3 up to 5 %. Excepting regions near Tin , for cubic samples the resistance decreases with increasing Sr concentration. For rhombohedral samples the resistance increases with increasing Sr concentration. For the sample with x = 0.05 crystallizing in the P m¯3m space group, we observe a narrow maximum of the resistance at Tin ≈ 293 K. This maximum arises due to the large concentrations of the 1D rod-like nanomagnetic domains. A more interesting behavior was observed for the sample with x = 0.15 crystallizing in the R¯3c space

T

H = 0

in

T

x=0.15

ex

x=0.20

30

R

, (

)

20

T

ex

5

T T

in

ex

3

1 100

200

300

T, (K) Figure 6: Variation of resistance R with temperature and Sr concentration x. Tex and Tin represent the transition temperatures from metallic to insulator state for extrinsic and, respectively, intrinsic component of the resistance.

The SANS measurements show a decrease of average volume and concentration of magnetic nanodomains with temperature. For the sample with x = 0.15 the concentration of nanomagnetic phase remains practically constant until about 300 K, when it begins to decrease monotonously. Starting at 280 K it takes place a decrease of average size of the magnetic domains, implicitly of the average size of the metallic nanophase, which contribute to the quick increase of the resistance with temperature. The decrease of the resistance beginning with 293 K takes place due to the decrease of the concentration of the metallic phase and the increase of the insulator phase (Fig. 6). For x = 0.05, the volume of the 8

group. We observed a narrow maximum of the resistance at Tin ≈ 293 K. By decreasing the temperature, the nanomagnetic domains “grow” from randomly oriented 1D rod-like structures into 2D disk-like and, for the lowest temperatures, into 3D mass fractal structures and finally into microscopic magnetic 3D domains. A maximum of the average nanomagnetic domain volume and magnetic volume fraction with Sr concentration was observed at x = 0.15. The maximum of the resistance is due to high concentration of 2D disk-like nanomagnetic domains. However, for x = 0.05 the narrow peak is about one order higher than for x = 0.15. This behavior arises due to the fact that increasing the Sr concentrations is accompanied by the transition of nanomagnetic domains from 1D rod-like to 2D disk-like type.

[12] J. P. Zhou, J. T. McDevitt, J. S. Zhou, H. Q. Yin, J. B. Goodenough, Y. Gim, Q. X. Gia, Appl. Phys. Lett. 75 (1999) 1146–1148. [13] M. L. Craus, E. M. Anitas, N. Cornei, A. K. Islamov, V. Garamus, Sol. Stat. Phenom. 190 (2012) 121–124. [14] L. A. Feigin, D. I. Svergun, Structure Analysis by Small Angle X Ray and Neutron Scattering, Plenum, New York, 1987. [15] J. E. Martin, A. J. Hurd, J. Appl. Cryst. 20 (1987) 61–78. [16] D. I. Svergun, J. Appl. Cryst. 25 (1992) 495–503. [17] D. Franke, D. I. Svergun, J. Appl. Cryst. 42 (2009) 342–346. [18] A. Guinier, G. Fournet, Small-angle scattering of Xrays, John Wiley & Sons, New York, 1955. [19] C. Zener, Phys. Rev. 82 (1951) 403–405. [20] J. Rodriguez-Carvajal, CPD Newsletter 26 (2001) 12.

Acknowledgements The authors acknowledge support from JINR IFIN-HH projects, from JINR grant No. 13-302-02 and thank to ONEXIM group for the possibility to use SAXS experimental facilities at MIPT. References References [1] M. Salamon, M. Jaime, Rev. Mod. Phys. 73 (2001) 583–628. [2] Y. Tokura, Y. Tomioka, J. Magn. Magn. Mater. 200 (1999) 1–23. [3] M. Ziese, Rep. Prog. Phys. 65 (2002) 143–249. [4] D. Saurel, A. Brulet, A. Heinemann, C. Martin, S. Merconne, C. Simon, Phys. Rev. B 73 (2006) 094438–1– 094438–9. [5] D. Saurel, C. Simon, A. Pautrat, C. Martin, C. Dewhurst, A. Brlet, Phys. Rev. B 82 (2010) 054427–1– 054427–7. [6] M. Viret, F. Ott, J. P. Renard, H. Glttli, L. PinsardGaudart, A. Revcolevschi, Phys. Rev. Lett. 93 (2004) 217402–1–217402–4. [7] C. Castellano, A. Martinelli, M. Ferretti, M. R. Cimberle, C. Mondelli, J. Alloys Comp. 542 (2012) 63–67. [8] J. Guttierez, F. J. Bermejo, N. Veglio, J. M. Barandiar´ an, P. Romano, C. Mondelli, M. A. Gonz´ ales, A. P. Murani, J. Phys. Cond. Matt. 18 (2006) 9951–9966. [9] S. M. Yusuf, J. M. D. Teresa, P. A. Algarabel, J. Blasco, M. R. Ibarra, A. Kumar, C. Ritter, Physica B 385-386 (2006) 401–404. [10] A. I. Kuklin, A. K. Islamov, V. I. Gordeliy, Neutron News 16(3) (2005) 16–18. [11] A. G. Soloviev, T. M. Solovieva, A. V. Stadnik, A. K. Islamov, A. I. Kuklin, JINR Communication P10-200386 (2003).

9

Highlights: * We synthesized a series of pure La0.5Pr0.2Pb0.3-xSrxMnO3 manganites using ceramic technology. * The crystalline structure depends on the Sr concentration. * Small-angle neutron scattering reveals the presence of magnetic nanodomains at temperatures higher than Curie temperature. * The crystalline structure, the shape and the concentration of magnetic nanodomains are correlated with transport phenomena.