ARTICLE IN PRESS

Physica B 405 (2010) 3259–3266

Contents lists available at ScienceDirect

Physica B
journal homepage: www.elsevier.com/locate/physb

Substrate-dependent structural and magnetic properties of Sr2FeMoO6

nanostructured double perovskite thin films
Deepak Kumar, Davinder Kaur n
Fuctional Nanomaterials Research Laboratory, Department of Physics and Centre of Nanotechnology, Indian Institute of Technology Roorkee, Roorkee 247 667, Uttrakhand, India

a r t i c l e in fo abstract

Article history: We report the synthesis of nanostructured Sr2FeMoO6 (SFMO) thin films on SrTiO3 (0 0 1), LaAlO3
Received 4 October 2009 (0 0 1) and MgO (0 0 1) substrates by pulsed laser deposition technique. The influence of substrate
Received in revised form nature on structural and magnetic properties of thin films has been systematically investigated. The
20 February 2010
films deposited on SrTiO3 and LaAlO3 were single crystalline with preferred (0 0 4) orientation without
Accepted 23 April 2010
any impurity peak, whereas the films deposited on MgO substrate exhibit impurity peaks of SrFeO3
along with (0 0 4) reflection of SFMO. The out of plane lattice parameter, ‘c’ and grain size have been
Keywords: shown to be influenced by nature of substrate. Further, the magnetic properties of the films, including
Double perovskites the coercive field, remnant magnetization, saturation magnetization and magnetoresistance were also
Pulsed laser deposition
observed to be strongly dependent on the choice of substrate. The investigations reveal an increase in
Magnetoresistance
magnetization with decrease in film–substrate lattice mismatch. The films on SrTiO3 substrate exhibit a
saturation magnetic moment of 3.24mB/f.u. at 5 K, with a Curie temperature larger than 320 K, whereas
the films deposited on MgO substrate are observed to have low magnetization, which is related to the
high proportion of antisite defects in these films. The magnitude of MR was observed to be smaller in
case of films deposited on SrTiO3 substrate and in addition the variation in MR with applied field was
found to be linear in these films, whereas the films on MgO substrate were found to exhibit higher value
of MR with two different regimes: the low-field region (Ho 20 kOe) with a steeper slope and the high-
field behavior with a less pronounced slope. The present study elucidates the effects of substrate nature
on magnetization and that the magnetoresistance behavior of SFMO thin films are useful in magnetic
sensors and spintronics application.
& 2010 Elsevier B.V. All rights reserved.

1. Introduction Polycrystalline bulk samples have been extensively investi-

gated for fundamental studies aimed to establish the structural
Recently, half metallic ferromagnet Sr2FeMoO6 (SFMO) has and fundamental properties of SFMO [2–13]. However, epitaxial
attracted much attention due to its large room temperature or highly oriented thin films of this compound are necessary for
magnetoresistance (MR) and high curie temperature of Tc  410 K application in spintronics, which has not proven to be easy. There
making it a potential candidate for spintronic device applications are several constraints on the growth parameters of the films
such as spin valves and magnetic tunnel junctions [1]. SFMO is an dictated by the difficulty of avoiding the segregation of spurious
ordered double perovskite (A2BB0 O6) with regular arrangement of phases and by the necessity of minimizing disorder in the double
corner sharing FeO6 and MoO6 octahedra, alternating along all perovskite A2BB’O6 lattice. It was observed that often the
three directions of the crystal [2–4]. The Sr cations reside in the magnetization is reduced due to antisite defects, which causes
cuboctahedral vacancies created by the corner sharing octahedral. the sample to locally separate into SrFeO3 and SrMoO3 areas that
The Fe3 + and Mo5 + ions are spatially distributed alternately in the in turn reduce the magnetization because SrFeO3 is antiferro-
B site of the perovskite structure. The antiferromagnetic coupling magnetic with TN ¼ 134 K and SrMoO3 is paramagnetic [13]. Due
of Fe3 + (S¼5/2) and Mo5 + (S¼ 1/2) ions leads to saturated to severe constraints and narrow space of deposition parameters
moment Ms per formula unit of 4mB. However the reported value for growth of high-quality films of these double perovskites, there
of Ms is smaller than this predicted value due to presence of are only few reports on SFMO thin films [14–19] as compared to
naturally grown antisite defects, where the partial Fe(Mo) ions are very large number of studies on polycrystalline bulk material.
misplaced at Mo(Fe) sites [5,6]. Among various thin film deposition techniques, PLD has been
recognized as a promising versatile technique for the deposition
of stoichiometric films of SFMO and other metal oxides at high
n
Corresponding author. Tel.: + 91 1332 285407; fax: +91 1332 273560. deposition rate [17–21]. Moreover, PLD has an added advantage
E-mail address: [email protected] (D. Kaur). of being a non-equilibrium process. Thin films with Tc higher than

0921-4526/$ - see front matter & 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.physb.2010.04.056
 ARTICLE IN PRESS
3260 D. Kumar, D. Kaur / Physica B 405 (2010) 3259–3266

300 K have been obtained by a careful selection of the deposition sectional Field Emission Scanning Electron Microscopy (FESEM)
parameters. In particular, the critical role played by the oxygen and was kept constant  200 nm for all the films. The magnetic
background pressure and by the substrate temperature has been properties of the films were measured using superconducting
clearly demonstrated. Oxygen pressure higher than 10  4 mbar quantum interference device (Quantum Design) in an applied
has been reported to be deleterious for the stabilization of single field of 77 T. Magnetoresistance was measured using a computer
phase SFMO, because the more stable SrMoO4 phase can nucleate controlled system made up of a Keithley-220 sourcemeter, a
[22]. Substrate temperature has been reported to affect the Keithley-2000 nanovoltmeter and a LakeShore-340 temperature
vertical unit cell size of SFMO films deposited by PLD, as at low controller by the conventional four-terminal technique with
substrate temperature the atomic mobility seems to be insuffi- magnetic field up to 110 kOe. The contacts over the sample were
cient to achieve a perfect arrangement of the double perovskite made by silver paint.
configuration. Asano et al. [18] observed that the films deposited
at low temperatures has semiconductor like behavior of resistiv-
ity, whereas those deposited at high temperatures were metallic. 3. Results and discussion
Westerburg et al. [16] observed small negative MR of  3% at very
large magnetic field of 80 kOe at 300 K in case of their PLD films. 3.1. Structural properties
Manako et al. [17] have also reported that SFMO films could
only be obtained in a narrow range of deposition temperature Fig. 1 shows the X-ray diffraction (XRD) pattern of SFMO
and oxygen partial pressure. Shinde et al. [19] have also reported thin films deposited on various substrates at fixed deposition
that best quality SFMO films could only be grown at particular temperature of 825 1C and pressure of 5  10  5 Torr. The samples
substrate temperature and oxygen partial pressure. Trolio et al. S1, S2 and S3 are the films deposited on SrTiO3, LaAlO3 and MgO
[22] have demonstrated the influence of laser fluence on structural substrate, respectively. The XRD pattern of films S1 and S2 showed
and magnetic properties of SFMO films. only diffraction peaks corresponding to SFMO (0 0 l) reflections
In the case of SFMO thin films, the role of substrates is also and the substrate, indicating that the films were oriented along
very important because it can limit the growth direction and the c-axis, perpendicular to the growth plane. However, the XRD
growth rate. There are reports on synthesis of SFMO thin films on pattern of sample S3 showed additional minute spurious peak at
different substrates such as SrTiO3 [22], MgO [23] and LaAlO3 [24]. 2y ¼42.91 due to small quantity of secondary phase, i.e. strontium
However, there are limited studies on systematic investigation of ferrate (SrFeO3  a 0o a o0.5), which has been reported to affect
influence of substrate on microstructure and magnetic properties the magnetic behavior of SFMO films. The strontium ferrate phase
of these films. In the present study we have tried to investigate develops when antisite disorder occurs in the double perovskite
the influence of substrate type such as SrTiO3, LaAlO3 and MgO on lattice [26,27]. We are able to observe (1 0 1) reflection [inset
growth and properties of SFMO thin films. The grain size, c-axis of Fig. 1] depicting the ordering of Fe/Mo ions at the B/B0 sites
lattice constant, saturation magnetization and magnetoresistance in these films. Further the intensity ratio of (1 0 1) and (0 0 4)
has been shown to be influenced by the nature of substrate. These reflections was calculated and antisite defects were determined
studies are useful in magnetic sensors and spintronics application. for samples S1, S2 and S3 in similar manner as reported by
Venimadhav et al. [46] The percentage of antisite defects
calculated by this method is presented in Table 2. It is clear that
2. Experimental the films grown on MgO substrate have less degree of B-site
ordering in comparison to films grown on LaAlO3 and SrTiO3
SFMO thin films were grown by pulsed laser deposition substrate. The inset of Fig. 1 shows the c-axis lattice parameter of
technique using a 15 mm diameter target. The details of process these films, calculated from XRD results. The value of the c-axis
setup are given elsewhere [25]. The targets were prepared as lattice constant was 8.023 Å (S1), 7.95 Å (S2) and 7.94 Å (S3).
polycrystalline powders by solid state reaction method. Stoichio- For all the films, the c-axis was observed to be elongated in
metric amounts of high-purity powders of SrCO3, Fe2O3 and MoO3 comparison to the bulk c-axis (7.879 Å) [28,29], which could be
were ground and calcined at 900 1C for 12 h. The resulted attributed to strain [30–32] originating from the film–substrate
powders were reground and pressed into pellets under a pressure lattice mismatch or cationic offstoichiometry [33] or to the
of about 80 MPa and then sintered at 1000 1C for 24 h in argon presence of disorder in the Fe and Mo sublattices [18,19].
atmosphere. The excimer laser KrF with wavelength of 248 nm The film–substrate lattice mismatch was calculated for all the
was used for deposition. The pulse repetition rate was 10 Hz with samples and is presented in Table 1. We note that SrTiO3 has
laser fluence of about 2–3 J cm  2. The target substrate distance the best lattice match with SFMO among these substrates with
was kept at 40 mm and working pressure was 5  10  5 Torr. lattice mismatch of  1.14%. The negative value indicates that
The deposition temperature and time was kept at 825 1C and the substrate lattice parameter is smaller than the film lattice
25 min, respectively. Single crystal SrTiO3 (0 0 1), LaAlO3 (0 0 1) parameter and the film will grow in a compressive strain
and MgO (0 0 1) wafers were used as substrates and were cleaned environment. SrTiO3 (0 0 1) is a cubic perovskite with a lattice
sequentially in acetone, methanol and deionized water prior to parameter c, such that when doubled 2c¼7.81 Å it is smaller
deposition. than the corresponding SFMO lattice constant c ¼7.879 Å.
The orientation and crystallinity of the films were investigated Therefore, SFMO films sustain an in-plane compressive strain
using Bruker AXS D-8 advanced diffractometer of Cu Ka (1.54 Å) when deposited on SrTiO3 substrate. On the other hand MgO
in y–2y geometry. To obtain a profile fitting with good signal, was observed to have maximum lattice mismatch of 6.41%.
polycrystalline silicon powder was used for instrumental correc- The positive lattice mismatch means that the substrate has a large
tion. The surface morphology of the films was studied by means of lattice parameter than the film lattice parameter and the film will
atomic force microscopy in semi-contact mode (AFM, NTMDT) grow in a tensile strain condition. The large value of lattice
with silicon nitride (Si3N4) tip of 10 nm radius. The root-mean- mismatch in case of films deposited on MgO substrate could
square (RMS) roughness of the surface was calculated from AFM probably be the reason for high density of antisite defects in these
scans over substrate areas of 3 mm  3 mm, and the RMS rough- films and is in agreement with the reported results [18]. Further
ness was calculated three times at a different spot for each using XRD data the crystallite size of the films was also calculated
sample. The thickness of the films was measured using cross- and was found to be 27.6, 16.0 and 12.7 nm for sample S1, S2 and
 ARTICLE IN PRESS
D. Kumar, D. Kaur / Physica B 405 (2010) 3259–3266 3261

8.04
(a) (004)
(101) 8.00

MgO
18 19 20 43 7.96

c (Å)
44 45 46
(101) (004)
7.92
18 19 20 44 45 46 47
7.88 Bulk SFMO
(101) (004)

(004)
Intensity (a.u.)

STO
S1 S2 S3
18 19 20 45 46 47

LAO
2θ (Degree)
S1

S2

* SrFeO3
+ SrMoO4 * S3
+

30 40 50 60 70
2θ (degree)

Fig. 1. X-ray diffraction patterns of SFMO thin films deposited on STO, LAO and MgO substrates. Inset (a) shows the (1 0 1) and (0 0 4) reflections and in the inset (b) we
have shown the c-axis lattice constants of S1, S2 and S3 samples.

Table 1
Calculated crystallite size, lattice parameter and microstrain for SFMO thin films deposited on MgO, LAO and STO substrates.

Substrate Lattice parameter ‘c’ Lattice mismatch (%) Crystallite size (nm) Surface Roughness (nm)

Substrate (Å) Film (Å) XRD AFM

SrTiO3 7.81 8.023  1.14 27.62 83.7 3.2

LaAlO3 7.642 7.95  3.27 15.95 48.5 7.6
MgO 8.406 7.94 6.41 12.71 36.4 11.3
SFMO 7.897    

S3, respectively. The small value of crystallite size observed in The data have been corrected for the background signal from the
sample S3 supports the presence of disorder in the Fe and Mo substrate using equation
sublattices. Mfilm ðHÞ ¼ Mtotal ðHÞwsubstrate H
The surface morphology of the films was studied using
atomic force microscopy (AFM) in tapping mode and is shown where wsubstrate is the susceptibility of the substrate, Mtotal the
in Fig. 2(a–c). It clearly depicts that the sample S1 is magnetization of (film+substrate) and H the magnetic field
homogeneous, smooth, and uniform with average roughness of applied parallel to the film surface. Fig. 3(a) and 3(b) show
3.2 nm while sample S2 and S3 show roughness of 7.6 and the M–H curves of S1, S2 and S3 at 5 and 300 K. Inset of Fig. 3(a)
11.3 nm, respectively. The grain size of the films was found to be shows the low-field region of the loop, showing coercivity (Hc) of
83.7 nm (S1), 48.5 nm (S2) and 36.4 nm (S3). The smallest grain  1163 Oe, and remanent magnetization of about (Mr)  0.88mB/
size observed in case of films deposited on MgO substrate was in f.u. for sample S3. The value of coercivity (Hc)  800 Oe and
confirmation with the XRD results. However, overall particle size remanent magnetization 0.91mB/f.u. was observed in case of
shown by AFM was much larger as compared with that calculated sample S2, whereas sample S1 showed Hc  716 Oe and Mr
from the XRD results (Table 1). This was because of the fact that  0.96mB/f.u. at 5 K. It is worth noting that the value of Hc for
the XRD gave the average mean crystallite size while AFM showed sample S3 was 1.6 times higher than sample S1, further indicating
agglomeration of the particles. The XRD and AFM data can be more disorder in case of sample S3. The ratio of Hc values
reconciled by the fact that smaller primary particles have a large measured at 5 and 300 K was 7.31 for sample S1, 6.25 for sample
surface free energy and would, therefore, tend to agglomerate S2 and 5.41 for sample S3 (Table 2).
faster and grow into larger grains. Assuming that the film density is close to bulk theoretical
value and film volume is only filled with the SFMO unit cells, the
value of saturation magnetization Ms at 5 K was 3.24mB/f.u. (S1),
3.2. Magnetic properties 2.83mB/f.u. (S2) and 2.21mB/f.u. (S3). Such a value is lower than
theoretical 4mB/f.u. expected on the basis of the ferromagnetic
The influence of substrate on magnetic properties of SFMO thin coupling between Fe3 + and Mo5 + configurations suggesting the
films was carried out using a SQUID magnetometer. The area presence of antisite defects in these films. The reported experi-
of the substrates used for the film deposition was 0.5  0.5 cm2. mental results on value of saturation magnetization (Ms) are very
 ARTICLE IN PRESS
3262 D. Kumar, D. Kaur / Physica B 405 (2010) 3259–3266

In most cases of thin films and bulk SFMO, the value of Ms is

shown to be less than the theoretical value. Since the ionic sizes of
Fe + 3 and Mo + 5 are similar, there is a finite concentration of mis-
site disorder in Sr2FeMoO6, which interchanges the position of Fe
and Mo sites in a random fashion. The most significant effect of
disordering is to reduce the net magnetization of the sample.
However, the nature and origin of this decrease in the magnetiza-
tion in the presence of disorder is still a matter of debate in the
literature [38,39]. There are two distinct ways that the net
magnetization may be reduced in Sr2FeMoO6 in the presence of
mis-site disorders. One possibility is that the disorder destroys the
specific spin arrangement of Fe and Mo sublattices without any
significant effect on the individual magnetic moments at these
sites. This can be achieved by transforming the ferromagnetic
coupling between some of the Fe sites to an antiferromagnetic
coupling. This view has been preferred by most in recent times,
under the assumption that Fe–O–Fe interactions, induced by the
mis-site disorder in place of Fe–O–Mo, will be antiferromagneti-
cally driven by the superexchange. Alternately, the magnetic
moments at each individual site may decrease due to the different
chemical environments induced by the disorder, without affecting
the nature of the spin order within the Fe and Mo sublattices.
The real situation may even be a combination of both these effects,
with a simultaneous reduction in the magnetic moments at
different sites as well as a change in the nature of the magnetic
coupling between different sites. In order to understand the nature
of magnetic exchange in double perovskites, X-ray magnetic
circular dichroism (XMCD) has been used to find the value of
magnetic moment on the B and B0 site of double perovskites. For
Sr2FeMoO6 single crystal, using XMCD measurements, Besse et al.
[35] have shown for the first time the spin moments of +3.05 and
 0.32mB for Fe3 + and Mo5 + ions, respectively, which gives a direct
confirmation of the ferrimagnetic ordering and also predicts that
a localized spin carried on Fe sites and a delocalized electron
of Mo are responsible for the half metallicity of Sr2FeMoO6
compound. This is in agreement with a kinetic energy driven
exchange model proposed by Sarma et al. [41], which explain the
strong ferromagnetic exchange in double perovskites despite the
large distance between the magnetic ions. Subsequently, Majewski
et al. [36,37] have also investigated the magnetic moment of the 5d
shell of the nonmagnetic W ion in the double perovskites Sr2CrWO6
and Ca2CrWO6 using XMCD. In both compounds a finite negative
spin and positive orbital magnetic moment was detected. These
studies suggest that there is a correlation between the magnitude
of the magnetic moment at the nonmagnetic ion and the magnetic
ordering temperature Tc in the double perovskites.
The relation between the antisite defect (of concentration x) and
saturation magnetization is given by Ms ¼4–8xmB/formula unit (f.u.).
This can be easily understood if one considers that the saturation
magnetization of perfect SFMO (4mB/f.u.) is calculated assuming an
antiferromagnetic coupling between Fe + 3 (3d5; S¼5/2) and Mo + 5
Fig. 2. AFM images of SFMO thin films deposited on (a) STO, (b) LAO and (c) MgO
substrates. (4d1; S¼1/2) spins as previously reported for double perovskite
systems [40], and that the interaction between two Fe magnetic
moments through the 2p oxygen orbitals is also antiferromagnetic.
different in thin films as well as bulk. Recently, Li et al. [34] have The value of antisite defect in our case was found to be highest
reported the value of Ms is less than 3.2mB/f.u. at 10 K in case of (22.37%) for S3 sample, 14.6% for S2 and lowest (9.5%) for S1 sample.
their bulk samples prepared by sol–gel method. For single crystal The results indicate that decrease in concentration of antisite defects
of SFMO, the reported value of Ms is 3.2mB/f.u. at 10 K [29]. and corresponding increase in magnetization are according to the
Westerburg et al. [16] have reported Ms value of 4mB/f.u. for SFMO substrate order MgO4LaAlO3 4SrTiO3. The differences in magnetic
thin films grown under Ar atmosphere and have observed the properties are likely arise from variations in the lattice–substrate
(1 0 1) peak indicative of a significant degree of B-site ordering. mismatch, which can induce different epitaxial strain states, sample
Trolio et al. [22] have reported saturation magnetization of 3.4mB/ inhomogeneities and phase separation. Since the lattice mismatch of
f.u. in case of films prepared by pulsed laser deposition on SrTiO3 SFMO/MgO (6.41%) is larger than those of SFMO/LAO ( 3.47%) and
substrate at low laser fluence, whereas Shinde et al. [19] have SFMO/STO ( 1.14%), the degree of lattice distortion and the amount
reported saturation magnetization of 3.28mB/f.u. in case of films of inhomogeneities in SFMO/MgO are expected to be larger than
prepared by a similar technique. those in the SFMO/LAO and SFMO/STO. The lattice distortion and
 ARTICLE IN PRESS
D. Kumar, D. Kaur / Physica B 405 (2010) 3259–3266 3263

4 3 Sample S2
Sample S1
3
2
2
Magnetization (μΒ/f.u.)

Magnetization (μΒ/f.u.)
5K 1 5K
1 300K 300K

0 0
1

Magnetization (μB/f.u.)
Magnetization (μB/f.u.)
1
-1
-1
0 0
-2
-2
-3 -1 -1
-800 -400 0 400 800 -800 -400 0 400 800
Magnetic Field (Oe) -3 Magnetic Field (Oe)
-4
-10000 -5000 0 5000 10000 -10000 -5000 0 5000 10000
Magnetic Field (Oe) Magnetic Field (Oe)

Sample S3
2

1 5K
Magnetization (μΒ/f.u.)

300K

0 1.0
Magnetization (μB/f.u.)

0.5

-1 0.0

-0.5

-2 -1.0
-1000 -500 0 500 1000
Magnetic Field (Oe)
-15000 -10000 -5000 0 5000 10000 15000
Magnetic Field (Oe)

Fig. 3. Magnetization hysteresis loops of (a) S1, (b) S2 and (c) S3 samples measured at 5 and 300 K. The magnetic field is applied parallel to the film plane. Inset shows the
data for the low-field regions.

Table 2
Various magnetic parameters of SFMO thin films deposited on MgO, LAO and STO substrates.

Sample Coercivity Hc (Oe) Ratio of Remnant Saturation Intensity Ratio I Antisite Defect (%) Magnetoresistance
Hc 5 K/300 K Magnetization Magnetization (1 0 1)/I (0 0 4) MR (%)
Mr (mB/f.u.) Ms (mB/f.u.)

5K 300 K 5K 300 K 5K 300 K From XRD FIM Model 70 K 200 K 300 K

S1 716 98 7.31 0.96 0.19 3.24 1.62 0.034 12 9.5  0.8  0.3  0.18
S2 800 128 6.25 0.91 0.18 2.83 1.32 0.025 17 14.6  4.7  3.1  1.5
S3 1163 215 5.41 0.88 0.17 2.21 1.1 0.017 24 22.37  16.5  12.9  3.6

sample inhomogeneities decrease the degree of ferromagnetic the magnetic moment was found to increase rapidly with the
long-range order, inducing the decrease in magnetization. decrease in temperature.
Fig. 4 shows the magnetization versus temperature of sample
S1, S2 and S3. The measurements were performed after field 3.3. Magnetoresistance
cooling the sample from 300 K down to 5 K with a magnetic field
H¼200 Oe parallel to the film surface. The magnetization curve
The magnetoresistance (MR) is defined as
for sample S1 reflects continuous decrease in magnetization with
temperature, which suggests Tc above 320 K. In case of sample S3, RðHÞRð0Þ
MRð%Þ ¼  100
no evidence of long-range ferromagnetic order was observed and Rð0Þ
 ARTICLE IN PRESS
3264 D. Kumar, D. Kaur / Physica B 405 (2010) 3259–3266

1.5
Sample S1 1.2 Sample S2

1.2
Magnetization (µB/f.u.)

Magnetization (µB/f.u.)
0.9

0.9 H = 200 Oe H = 200 Oe

0.6

0.6
0.3

0.3

0 50 100 150 200 250 300 350 0 50 100 150 200 250 300 350
Temperature (K) Temperature (K)

1.0
Sample S3

0.8
Magnetization (µB/f.u.)

0.6

H = 200 Oe
0.4

0.2

0 50 100 150 200 250 300 350

Temperature (K)

Fig. 4. Temperature-dependent magnetization of (a) S1, (b) S2 and (c) S3 samples measured from 5 to 320 K at a constant magnetic field of 200 Oe.

Where R(H) is the resistance of the sample at temperature T in the understood in terms of the existence of antiferromagnetic (AF)
presence of an applied magnetic field of strength H. MR of SFMO correlations at the antisite defects (i.e. between two consecutive
thin films with temperature on different substrates was measured Fe cations). This antiferromagnetism introduces a significant
using a four probe resistivity measurement at magnetic field up to spin disorder in the system, which in turn, leads to an increase
110 kOe. The area of the substrates used for the film deposition in the material resistivity, due to the enhanced spin-polarized
was 0.5  0.5 cm2. The magneoresistance (MR) measured for electron scattering. The application of an external magnetic field
samples S1, S2 and S3 at three different temperatures of 70, 200 suppresses this spin disorder to a great extent, lowering the
and 300 K is shown in Fig. 5. The MR magnitude for all the resistivity and thus inducing a remarkable magnetoresistive effect
samples at room temperature, compared to that seen at 70 K, is and is consistent with our MR results on sample S3.
somewhat reduced, being 3–5 times smaller, which could be due It was interesting to note that magnitude of MR in case of films
to spin thermal fluctuations being dominant at high temperature. deposited on SrTiO3 substrate (S1) at all measured temperatures,
The values of magnetoresistance at 70 K for samples S1, S2 and S3 i.e. 70, 200 and 300 K, was smaller (Table 2) and in addition to the
were  0.8%,  4.7% and  16.5%, respectively (Table 2). MR was variation in MR with applied field was almost linear, which is in
observed to be minimum for films deposited on SrTiO3 substrate agreement with the reported results on epitaxial films [19]. On
(sample S1) and maximum for films deposited on MgO substrate the other hand, in case of films deposited on MgO substrate (S3)
(sample S3) with higher concentration of disorder defects. the magnitude of MR was largest at all measured temperatures
There are three basic mechanisms that give rise to a high and the variation in magnetoresistance with applied magnetic
magnetoresistive response in SFMO material: the extrinsic field showed two different regions: the low-field region (Ho20
tunnel magnetoresistance (TMR), accounting for a large fraction kOe) with a steeper slope and the high-field behavior with a less
of the observed magnetoresistance in the polycrystalline SFMO pronounced slope without showing any sign of saturation up to
bulk, mainly at low temperatures and low fields [41]; the intrinsic the highest magnetic field (110 kOe). We note the present type
magnetoresistive effect associated with the ferromagnetic of MR in case of sample S3, i.e. sharp increase in the magnitude of
transition, appearing also in single crystals and enhanced at MR at low applied field and then a slow increase at large magnetic
temperatures near Tc, and also a third component related to field has also been observed in polycrystalline SFMO [1] and some
the presence of antisite defects. This latter contribution can be other half-metallic oxides [42,43], as well as in the granular alloys
 ARTICLE IN PRESS
D. Kumar, D. Kaur / Physica B 405 (2010) 3259–3266 3265

0 0
STO STO
-2
-2
-4
LAO
Magnetoresistance (%)

LAO -4

Magnetoresistance (%)
-6

-8 -6

-10
-8
-12
-10
-14
T = 70 K MgO T = 200 K
-16 -12 MgO

-18 -14
0 2 4 6 8 10 12 0 2 4 6 8 10 12
Magnetic Field (T) Magnetic Field (T)

0.0
STO
-0.5

-1.0
Magnetoresistance (%)

LAO
-1.5

-2.0

-2.5

-3.0
T = 300 K MgO
-3.5

-4.0
0 2 4 6 8 10 12
Magnetic Field (T)

Fig. 5. Magnetoresistance of SFMO thin film on STO, LAO and MgO substrates at (a) 70 K, (b) 200 K and (c) 300 K.

containing some transition metal [44,45]. The origin of this low- 4. Conclusion
field MR in SFMO has been interpreted in terms of tunnelling; MR
at the grain boundaries. In this process, the hopping of the spin- In conclusion, highly crystalline SFMO thin films with (0 0 4)
polarized electron between microcrystalline grains (each with orientation were grown on different substrates using pulsed laser
their own magnetic domains) is critically affected by the relative deposition technique. Both structural and magnetic properties of
angle of the magnetization directions, and hence may be SFMO films were shown to be influenced by choice of substrate.
controlled by an external magnetic field via the domain-rotation The investigations reveal an increase in magnetization with
process. We speculate on the basis of the observed features, that a decrease in film–substrate lattice mismatch. The films deposited
similar process is dominant in case of SFMO films on MgO on SrTiO3 substrate with best lattice match with SFMO were
substrate. Fig. 6 shows the variation in MR with temperature for found to exhibit a maximum saturation magnetic moment,
samples S1, S2 and S3. Maximum change in values of MR was whereas the films deposited on MgO substrate with large misfit
observed in case of films deposited on MgO substrate. We obtain were observed to have low magnetization, which is related to the
values of MR (70 K), MR (200 K) and MR (300 K) as large as high proportion of antisite defects in these films. The magnitude
16.5%,  12.9% and  3.6%, respectively. We note that a steep of MR was observed to be smaller in case of films deposited on
low-field MR curve against the field was observed not only at 70 K SrTiO3 substrate and in addition the variation in MR with applied
but also at room temperature 300 K. Such a low-field-sensitive field was found to be linear in these films, whereas the films
MR, as well as its gradually increasing magnitude with decrease in on MgO substrate were found to exhibit higher value of MR with
temperature, signals that the spin-dependent scattering is not due two different regimes: the low-field region (Ho20 kOe) with a
to the bulk phenomenon arising from the thermal fluctuation of steeper slope and the high-field behavior with a less pronounced
spins, but occurs at the grain boundaries or at the magnetic slope. Such a low-field-sensitive MR, as well as its gradually
domain boundaries. increasing magnitude with decrease in temperature, suggests that
 ARTICLE IN PRESS
3266 D. Kumar, D. Kaur / Physica B 405 (2010) 3259–3266

0 [10] H. Sakuma, T. Taniyama, Y. Kitamoto, Y. Yamazaki, J. Appl. Phys. 93 (2003)

2816.
[11] K.-I. Kobayashi, T. Okuda, Y. Tomioka, T. Kimura, Y. Tokura, J. Magn. Magn.
-2
Mater. 218 (2000) 17.
[12] D.D. Sarma, E.V. Sampathkumaran, Sugata Ray, R. Nagarajan, Subham
-4 Majumdar, Ashwani Kumar, G. Nalini, T.N. Guru Row, Solid State Commun.
114 (2000) 465.
Magnetoresistance (%)

-6 [13] T. Saitoh, M. Nakatake, A. Kakizaki, H. Nakajima, O. Morimoto, S. Xu,

Y. Moritomo, N. Hamada, Y. Aiura, Phys. Rev. B 66 (2002) 035112.
[14] D. Sanchez, M. Garcia-Hernandez, N. Auth, G. Jacob, J. Appl. Phys. 96 (2004)
-8 2736.
STO [15] A. Venimadhav, F. Sher, J.P. Attfield, M.G. Blamire, J. Magn. Magn. Mater. 269
-10 LAO (2004) 101.
MgO [16] W. Westerburg, D. Reisinger, G. Jakob, Phys. Rev. B 62 (2000) R767.
[17] T. Manako, M. Izumi, Y. Konishi, K.-I. Kobayashi, M. Kawasaki, Y. Tokura, Appl.
-12 Phys. Lett. 74 (1999) 2215.
[18] H. Asano, S.B. Ogale, J. Garrison, A. Orozco, Y.H. Li, E. Li, V. Smolyaninova,
-14 C. Galley, M. Downes, M. Rajeswari, R. Ramesh, T. Venkatesan, Appl. Phys.
Lett. 74 (1999) 3696.
[19] S.R. Shinde, S.B. Ogale, R.L. Greene, T. Venkatesan, K. Tsoi, S.-W Cheong,
-16
A.J. Millis, J. Appl. Phys. 93 (2003) 1605.
[20] D. Kaur, J. Jesudasan, P. Raychaudhuri, Solid State Commun. 136 (2005) 369.
-18 [21] D. Kaur, K.V. Rao, Solid State Commun. 128 (2003) 391.
50 100 150 200 250 300 [22] A. Di Trolio, R. Larciprete, A.M. Testa, D. Fiorani, P. Imperatori, S. Turchini,
N. zema, J. Appl. Phys. 100 (2006) 013907.
Temperature (K) [23] R.P. Borges, S. Lhostis, M.A. Bari, J.J. Versluijs, J.G. Lunney, J.M.D. Coey,
M. Besse, J.P. Contour, Thin Solid Films 5 (2003) 429.
Fig. 6. Variation in magnetoresistance with temperature for films deposited on [24] X.B. Zhu, J.M. Dai, X.H. Li, B.C. Zhao, S.M. Liu, W.H. Song, Y.P. Sun, Mater. Lett.
STO, LAO and MgO substrates. 59 (2005) 2366.
[25] A. Kaushal, D. Kaur, Sol. Energy Mater. Sol. Cell 93 (2009) 193.
[26] D.Y. Kim, Jin Soo Kim, Bae Ho Park, Jeon-Kook Lee, Jang Hee Kim, Je Hyun lee,
the spin-dependent scattering is not due to the bulk phenomenon joonyeon Chang, Hi-Jung Kim, Appl. Phys. Lett. 84 (2004) 5037.
arising from the thermal fluctuation of spins, but occurs at the [27] M. Garcia-Hernandez, J.L. Martinez, M.J. Martinez-Lope, M.T. Casais,
grain boundaries or at the magnetic domain boundaries. J.A. Alonso, Phys. Rev. Lett. 86 (2001) 2443.
[28] B. Garcia-Landa, C. Ritter, M.R. Ibarra, J. Blasco, P.A. Algarabel, R. Mahendiran,
J. Garcia, Solid State Commun. 110 (1999) 435.
[29] Y. Tomioka, T. Okuda, Y. Okimoto, R. Kumai, K.-L. Kobayashi, Phys. Rev. B 61
Acknowledgement (2000) 422.
[30] P. Majewski, S. Geprags, A. Boger, M. Opel, L. Alff, R. Gross, J. Magn. Magn.
Mater. 1154 (2005) 290.
The financial support provided by Ministry of Communications [31] J.B. Philipp, P. Majewski, D. Reisinger, S. Geprags, M. Opel, A. Erb, L. Alff,
and Information Technology (MIT), India, under Nanotechnology R. Gross, Acta Phys. Pol. A 105 (2004) 7.
[32] J.B. Philipp, P. Majewski, L. Alff, A. Erb, R. Gross, T. Graf, M.S. Brandt, J. Simon,
Initiative Program with Reference no. 20(11)/2007-VCND and T. Walther, W. Mader, D. Topwal, D.D. Sarma, Phys. Rev. B 68 (2003)
DRDO ARMREB, New Delhi, is highly acknowledged. The authors 144431.
are also thankful to Prof. V. Vankataraman, IISc Bangalore, for [33] M. Besse, F. Pailloux, A. Barthélémy, K. Bouzehouane, A. Fert, J. Olivier,
O. Durand, F. Wyczisk, R. Bisaro, J.-P. Contour, J. Cryst. Growth 241 (2002)
providing research facilities for 11 T magnetic field measure- 448.
ments. The author Deepak Kumar is thankful to CSIR for award of [34] X.H. Li, Y.P. Sun, W.J. Lu, R. Ang, S.B. Zhang, X.B. Zhu, W.H. Song, J.M. Dai, Solid
senior research fellowship. State Commun. 145 (2008) 98.
[35] M. Besse, V. Cros, A. Barthelemy, H. Jaffres, J. Vogel, F. Petroff, A. Mirone,
A. Tagliaferri, P. Bencok, P. Decorse, P. Berthet, Z. Szotek, W.M. Temmerman,
References S.S. Dhesi, N.B. Brookes, A. Rogalev, A. Fert, Europhys. Lett. 60 (2002) 608.
[36] P. Majewski, S. Geprägs, A. Boger, M. Opel, A. Erb, R. Gross, G. Vaitheeswaran,
V. Kanchana, A. Delin, F. Wilhelm, A. Rogalev, L. Alff, Phys. Rev. B 72 (2005)
[1] K.-I. Kobayashi, T. Kimura, H. Sawada, K. Terakura, Y. Tokura, Nature 395 132402.
(1998) 677 (London). [37] P. Majewski, S. Geprags, O. Sanganas, M. Opel, R. Gross, F. Wilhelm,
[2] K.I. Kobayashi, T. Kimura, Y. Tomioka, H. Sawada, K. Terakura, Y. Tokura, Phys. A. Rogalev, L. Alff, Appl. Phys. Lett. 87 (2005) 202503.
Rev. B 59 (1999) 11 159. [38] H.Q. Yin, J.-S. Zhou, R. Dass, J.P. Zhou, J.T. McDevitt, J.B. Goodenough, J. Appl.
[3] F.K. Patterson, C.W. Moeller, R. Ward, Inorg. Chem. 2 (1963) 196. Phys. 87 (2000) 6761.
[4] F.S. Galasso, F.C. Douglas, R.J. Kasper, J. Chem. Phys. 44 (1966) 1672. [39] Ll. Balcells, J. Navarro, M. Bibes, A. Roig, B. Martinez, J. Fontcuberta, Appl.
[5] J.M. Greneche, M. Venkatesan, R. Suryanarayanan, J.M.D. Coey, Phys. Rev. B 63 Phys. Lett. 78 (2001) 781.
(2001) 174403. [40] A.W. Sleight, J.F. Weiher, J. Phys. Chem. Solids 33 (1972) 679.
[6] J. Linden, T. Yamamoto, M. Karppinen, Y. Yamauchi, T. Pietari, Appl. Phys. Lett. [41] D.D. sarma, Curr. Opin. Solid State Mater. Sci. 5 (2001) 261.
76 (2000) 2925. [42] H.Y. Hwang, S.W. Cheong, Nature 389 (1997) 942.
[7] A.P. Douvalis, M. Venkatesan, P. Velasco, C.B. Fitzgerald, J.M.D. Coey, J. Appl. [43] H.Y. Hwang, S.W. Cheong, Science 278 (1997) 1607.
Phys. 93 (2003) 8071. [44] D. Vanderbilt, Phys. Rev. B 41 (1990) 7892.
[8] O. Chmaissem, R. Kruk, B. Dabrowski, D.E. Brown, X. Xiong, S. Kolesnik, [45] K. Lassonen, A. Pasquarello, R. Car, C. Lee, D. Vanderbilt, Phys. Rev. B 47
J.D. Jorgensen, C.W. Kimball, Phys. Rev. B 62 (2000) 14197. (1993) 10142.
[9] D.D. Sarma, Sugata Ray, K. Tanaka, M. Kobayashi, A. Fujimori, P. Sanyal, [46] A. Venimadhav, M.E. Vickers, M.G. Blamire, Solid State Commun. 130 (2004)
H.R. Krishnamurthy, Dasgupta, Phys. Rev. Lett. 98 (2007) 157205. 631.