High temperature ferromagnetism in Ni-doped in2O3 and indium-tin oxide

Article abstract of DOI:10.1063/1.2220529

Observation of high temperature ferromagnetism in Ni-doped In ₂O₃ and indium-tin-oxide (ITO) samples prepared by a solid state synthesis route is reported. Both Ni-doped compounds showed a clear ferromagnetism above 300 K with the magnetic moments of 0.03-0.06 μ _B/Ni and 0.1μ_B/Ni at 300 and 10 K, respectively. Ni-doped In₂O₃ samples showed a typical semiconducting behavior with a room temperature resistivity of ρ ～ 2 Ω cm, while Ni-doped ITO samples were metallic with p ～ 2 × 10^-2 Ω cm. Analysis of different conduction mechanisms suggested that variable range hopping model explains our p-T data for the Ni-doped In₂O ₃ sample the best.

Full text of DOI:10.1063/1.2220529

High temperature ferromagnetism in Ni-doped In 2 O 3 and indium-tin oxide
Germanas Peleckis, Xiaolin Wang, and Shi Xue Dou
Citation: Applied Physics Letters 89, 022501 (2006); doi: 10.1063/1.2220529
View online: http://dx.doi.org/10.1063/1.2220529
View Table of Contents: http://scitation.aip.org/content/aip/journal/apl/89/2?ver=pdfcov
Published by the AIP Publishing
Articles you may be interested in
Structure–magnetic property relationship in transition metal ( M = V , Cr , Mn , Fe , Co , Ni ) doped Sn O 2
nanoparticles
J. Appl. Phys. 103, 07D141 (2008); 10.1063/1.2836797
Electrical transport properties of ferromagnetic Ga x Cr 1 x N thin films
Appl. Phys. Lett. 89, 142105 (2006); 10.1063/1.2357603
Low temperature transport properties of Ni-doped Co Ge 1.5 Se 1.5
J. Appl. Phys. 100, 036101 (2006); 10.1063/1.2214221
Room-temperature ferromagnetism in highly-resistive Ni-doped TiO 2
Appl. Phys. Lett. 88, 092505 (2006); 10.1063/1.2179607
Observation of ferromagnetism above 900 K in Cr–GaN and Cr–AlN
Appl. Phys. Lett. 85, 4076 (2004); 10.1063/1.1812581
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
139.184.30.132 On: Mon, 01 Sep 2014 09:58:41

APPLIED PHYSICS LETTERS 89, 022501 ͑2006͒
High temperature ferromagnetism in Ni-doped In2O3 and indium-tin oxide
Germanas Peleckis, Xiaolin Wang,a͒ and Shi Xue Dou
Institute for Superconducting and Electronic Materials, University of Wollongong, Wollongong,
New South Wales 2522, Australia
͑Received 7 February 2006; accepted 22 May 2006; published online 10 July 2006͒
Observation of high temperature ferromagnetism in Ni-doped In2O3 and indium-tin-oxide ͑ITO͒
samples prepared by a solid state synthesis route is reported. Both Ni-doped compounds showed a
clear ferromagnetism above 300 K with the magnetic moments of 0.03–0.06␮B/Ni and 0.1␮B/Ni
at 300 and 10 K, respectively. Ni-doped In2O3 samples showed a typical semiconducting behavior
with a room temperature resistivity of ␳ϳ2 ⍀ cm, while Ni-doped ITO samples were metallic with
␳ϳ2ϫ10−2 ⍀ cm. Analysis of different conduction mechanisms suggested that variable range
hopping model explains our ␳-T data for the Ni-doped In2O3 sample the best. © 2006 American
Institute of Physics. ͓DOI: 10.1063/1.2220529͔
Since the discovery of the giant magnetoresistance
͑GMR͒ effect in metallic multilayer structures,1 much effort
has been put into developing and finding new materials that
utilize the spin of electrons. Usually diluted magnetic semi-
conductors ͑DMSs͒ are formed when magnetic transition
metal ions, such as 3d elements, are doped into the host
lattice of a semiconductor. Theoretical predictions by Dietl
et al.2 have boosted the search for ferromagnetism in the
transition metal doped ZnO. The obtained results are very
controversial, where some groups claimed that this system is
ferromagnetic,3–5 while others have found no ferromag-
netism in TM doped ZnO.6–8 Recently, high temperature fer-
romagnetism was observed in Co-doped SnO2 with large
magnetic moment ϳ7.5␮B/Co.9 However, this giant mag-
netic moment was found only in as prepared samples. Due to
the small solubility of magnetic ions in the host semiconduc-
tor, ferromagnetism probably is induced by the clusters of
dopants formed during the sample preparation procedure.10
If one considers integrating electronic, magnetic, and
photonic properties in the new generation devices, the mate-
rials suitable for such devices should have high tunability of
charge carriers, high carrier mobility, optical transparency,
high abundance of the consisting elements, and low impact
on the environment. In2O3 is one of the most promising can-
didates for such tasks. It is a well known semiconducting
oxide already used practically worldwide. When In2O3 is
doped with Sn, a so called indium-tin oxide ͑ITO͒ is formed.
High conductivity of ITO is attributed to the high carrier
concentration which is caused by incorporation of Sn ions
into In2O3 host lattice and oxygen nonstoichiometry. ITO has
a cubic bixbyite structure with the lattice parameter a
=10.118 Å.11 Electrical conductivity ͑␴͒ for best quality ITO
samples was found to be ϳ104 S/cm.12
et al.15 showed that Ni is distributed uniformly throughout
the film, which suggests that clustering of Ni particles in the
samples is very unlikely. Hence, this fact gives a firm ground
for assumption that observed ferromagnetism in their
samples is rather intrinsic property of material. It is impor-
tant to note that laser ablation technique used in Ref. 15 is a
synthesis technique with nonequilibrium conditions. There-
fore samples, prepared under equilibrium conditions, exhib-
iting same properties would confirm observed magnetic phe-
nomena. In this letter we report on the observation of room
temperature ferromagnetism in Ni-doped In2O3 and ITO.
Polycrystalline
samples
of
In2−xNixO3
and
In2−xNixSn0.02O3 ͑1% Sn doping͒ were prepared by a conven-
tional solid state synthesis technique. In2O3 and SnO2 were
factory prepared chemicals ͑high purity: 99.99%, Aldrich͒.
Black NiO was prepared after decomposition of Ni͑OH͒
2
͑high purity: 99.99%, Aldrich͒ in tube furnace at 300 °C for
2 h in air. Starting materials were weighed and mixed in a
mortar in corresponding ratios to obtain nominal chemical
compositions for the final products. Mixed powders were
calcined in an argon atmosphere at 850 °C for 12 h. After
that, the reacted powders were reground, pressed into the
pellets, and fired at 950 °C for 12 h in an argon atmosphere.
The phases and crystal structure of samples were analyzed
by means of x-ray diffraction technique ͑XRD͒ using Cu K␣
irradiation ͑Mac Science; M03 XHF22͒. Figure 1 shows the
XRD patterns of the pulverized pellets of In1.95Ni0.05O3 and
In1.95Ni0.05Sn0.02O3. The observed diffraction patterns corre-
spond to that of cubic In2O3 phase. Lattice parameter a
calculated from Rietveld refinements of the x-ray diffracto-
grams showed some slight deviation from the literature data
of a=10.118 Å, i.e., a͑In1.95Ni0.05O3͒=10.121 Å and
a͑In1.93Ni0.05Sn0.02O3͒=10.119 Å.
Transport properties were investigated using four point
technique utilizing Physical Property Measuring System
͑PPMS, Quantum Design͒. Measurements were performed
in a temperature range from 350 down to 5 K. Figure 2
represents the dependence of the electrical resistivity ͑␳͒ on
temperature ͑T͒ for both samples. Ni-doped In2O3 shows a
characteristic semiconducting behavior, while Ni-doped ITO
clearly exhibits a metallic behavior with much lower ␳. The
absolute value of ␳ at 300 K of In1.95Ni0.05O3 was found
to be ␳=2 ⍀ cm, which is in a good agreement with pub-
lished data for Ni-doped In2O3 thin films.15 At 300 K
In1.93Ni0.05Sn0.02O3 sample has ␳=2.02ϫ10−2 ⍀ cm. A
Recently, Yoo et al.13 and He et al.14 have reported both
bulk and thin film samples of
a diluted magnetic
semiconductor—Fe and Cu codoped In2O3 oxide. The solu-
bility of Fe ions in the host compound was found to be
around 20%, which is very high. In addition to that Ni-doped
In2O3 thin films were also found to be ferromagnetic at room
temperature,15 although the doping level of Ni ͑ϳ6 wt %͒
was not as high as for samples reported in Ref. 13. Hong
a͒
Author to whom correspondence should be addressed; electronic mail:
xiaolin@uow.edu.au
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
0003-6951/2006/89͑2͒/022501/3/$23.00 89, 022501-1 © 2006 American Institute of Physics
139.184.30.132 On: Mon, 01 Sep 2014 09:58:41

022501-2
Peleckis, Wang, and Dou
Appl. Phys. Lett. 89, 022501 ͑2006͒
FIG. 1. X-ray powder diffraction patterns of pulverized In1.95Ni0.05O3 and
In1.93Ni0.05Sn0.02O3 pellets. The inset shows reference peaks for In2O3 phase
from ICDD Card No. 06-0416.
rather high resistivity of our Ni-doped ITO sample could be
explained by probable porosity of the sample due to the
preparation conditions ͑Ar atmosphere͒. Furthermore, as we
can see from the inset in Fig. 2, there is a minimum point
T=123 K in the ␳-T curve for Ni-doped ITO sample. One
should consider two different mechanisms in two tempera-
ture ranges. When TϾ123 K, the temperature dependent
electrical transport mechanism of the system should have
negative temperature coefficient of resistivity ͑TCR͒. Then,
if TϽ123 K, the TCR changes its value to +1. It is hard to
explain what happens and why the sign of TCR changes.
Ederth et al.16 have argued that at temperatures above 123 K
metallic behavior in electrical resistivity ͑␳͒ curves can be
caused due to the sintering of conducting clusters, which
lead to formation of conducting paths through the sample.
Below T=123 K the resistivity is dominated by ionized im-
purity scattering ant thus TCR becomes positive.
FIG. 3. Fittings of logarithm of electrical resistivity ͑r͒ vs temperature ͑T͒
for sample In1.95Ni0.05O3 applying ͑a͒ variable range hopping and ͑b͒ nearest
neighbor hopping conduction models.
Various electrical transport models were applied in order
to understand which transport mechanism explains the ␳-T
data of In1.95Ni0.05O3 sample the best. Firstly hopping con-
duction models were investigated. Generally hopping con-
duction can be represented by Eq. ͑1͒,
1/q
B3
␳ = B exp
,
͑1͒
ͩ ͪ
ͫ ͬ
2
T
where B1, B2, and q are constants, and T is temperature. A
conduction mechanism dominated by hopping between lo-
calized states distributed randomly in the material ͓variable
range hopping ͑VRH͔͒ has a temperature dependence de-
scribed by Eq. ͑1͒ with q=4. When q=1 localized states are
distributed in periodic array ͓nearest neighbor hopping
͑NNH͔͒. Secondly, a granular-metal ͑GM͒ system model was
considered. In such system metallic grains are separated by
thin insulating layers and q=2. Lastly, we have tried to fit
our data with regard to degenerate semiconductor ͑DS͒
model, which is expressed by Eq. ͑2͒,
␳͑t͒ ϰ t−3/2
.
͑2͒
For all of the mechanisms mentioned the fitted curves should
be linear. Our investigation revealed that the broadest tem-
FIG. 2. Electrical resistivity ͑␳͒ vs temperature ͑T͒ for In1.95Ni0.05O3 ͑open
circle͒ and In1.93Ni0.05Sn O3 ͑open square͒. The inset represents magnified
This article is copyrighted as0i.n02dicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
view of ␳-T curve for Ni-doped ITO sample.
perature range where ␳-T curve can be linearly fitted corre-
139.184.30.132 On: Mon, 01 Sep 2014 09:58:41

022501-3
Peleckis, Wang, and Dou
Appl. Phys. Lett. 89, 022501 ͑2006͒
ferromagnetic states. On the other hand, M-H loops at 300 K
͑inset Fig. 4͒ showed that more conductive Ni-doped ITO
sample has lower saturation magnetization, which contra-
dicts the above mentioned model. It might be possible that
lower portion of Ni ions contributes to the effective magnetic
moment of the system. Other models, such as bound mag-
netic polarons ͑BMPs͒ and double exchange interaction,
were also considered. In the former case, high conductivity
of our samples contradicts requirements for BMP model to
work. In the latter case, the probability of Ni being in mul-
tivalent state is very small, thus eliminating double exchange
interaction mechanism from the list of choices as well.
In summary, polycrystalline Ni-doped In2O3 and ITO
samples were prepared. Both compounds were found to be
ferromagnetic in the whole temperature range with saturation
magnetizations ͑Ms͒ at 300 K, Ms=0.06␮B/Ni and Ms
=0.03␮B/Ni for In1.95Ni0.05O3 and In1.93Ni0.05Sn0.02O3, re-
spectively. In1.95Ni0.05O3 transport properties measured
FIG. 4. Magnetization ͑M͒ as a function of temperature ͑T͒ for samples
Ni-doped In2O3 ͑open circle͒ and Ni-doped ITO ͑open square͒ under
2000 Oe applied magnetic field. Insets: M-H loops of the same samples at
300 K ͑left͒ and 10 K ͑right͒.
showed a typical semiconducting behavior with ␳
͑300 K͒
=2 ⍀ cm, while Ni-doped ITO was found to be metallic with
␳
=2.06ϫ10−2 ⍀ cm. Analysis of the different conduc-
͑300 K͒
sponds to that when q=4, i.e., VRH model ͓Fig. 3͑a͔͒. Other
curves showed large deviation from the linear fits in the same
temperature range, e.g., NNH model ͓Fig. 3͑b͔͒. Thus, VRH
conduction mechanism explains transport properties of our
Ni-doped In2O3 sample the best.
tion models suggested that electrical transport in our Ni-
doped In2O3 sample is best explained by variable range hop-
ping ͑VRH͒ conduction mechanism.
One of the authors ͑X.L.W.͒ thanks the support from the
Australian Research Council under Discovery Grant No.
DP0 558 753. Another author ͑G.P.͒ thanks the Australian
Government and the University of Wollongong for providing
IPRS and UPA scholarships for his Ph.D. studies.
The magnetization ͑M͒ versus temperature ͑T͒ data for
Ni-doped In2O3 and ITO samples are shown in Fig. 4. The
zero field cooled ͑ZFC͒ measurements were done during
warming in a field of 2000 Oe ͑Quantum Design, MPMS
XL͒. Both samples were found to be ferromagnetic in the
whole temperature range. The insets in Fig. 4 show magne-
tization ͑M͒ versus applied magnetic field ͑H͒ at 300 and
10 K. The maximal applied magnetic field varied from
2 to 5 T. Both samples show clear saturation magnetization
͑Ms͒ at 300 K indicating high temperature ferromagnetism in
the system. Observed Ms at 300 K was Ms=0.06␮B/Ni and
Ms=0.03␮B/Ni for In1.95Ni0.05O3 and In1.93Ni0.05Sn0.02O3, re-
spectively. The obtained Ms value for Ni-doped In2O3 is
smaller than Ms=0.7␮B/Ni at 300 K reported for Ni-doped
In2O3 thin films made by pulsed laser deposition ͑PLD͒.15
On the other hand, our Ms values at 300 K for both com-
pounds are well comparable to those reported for Mn-doped
ITO thin films17 and Co-doped In2O3 thin films,18 Ms
=0.08␮B/Mn and Ms=0.05␮B/Co, respectively. Investiga-
tion of M-H loops of the samples at 10 K showed that for
Ni-doped In2O3 M-H loop is still saturated and Ms
=0.1␮B/Ni, while magnetic response of Ni-doped ITO
sample does not have this trend anymore.
1M. N. Baibich, J. M. Broto, A. Fert, F. N. Van Dau, and F. Petroff, Phys.
Rev. Lett. 61, 2472 ͑1988͒.
2T. Dietl, H. Ohno, F. Matsukara, J. Cibert, and D. Ferrand, Science 287,
1019 ͑2000͒.
3K. Ueda, H. Tabata, and T. Kawai, Appl. Phys. Lett. 79, 988 ͑2001͒.
4K. P. Bhatti, S. Chaudhary, D. K. Pandya, and S. C. Kashyap, Solid State
Commun. 136, 384 ͑2005͒.
5C. B. Fitzgerald, M. Venkatesan, J. G. Lunney, L. S. Dorneles, and J. M.
D. Coey, Appl. Surf. Sci. 247, 493 ͑2005͒.
6A. S. Risbud, N. A. Spaldin, Z. Q. Chen, S. Stemmer, and R. Seshadri,
Phys. Rev. B 68, 205202 ͑2003͒.
7J. Alaria, P. Turek, M. Bernard, M. Bouloudenine, A. Berbadj, N. Brihi, G.
Schmerber, S. Colis, and A. Dinia, Chem. Phys. Lett. 415, 337 ͑2005͒.
8S. Kolesnik, B. Dabrowski, and J. Mais, J. Appl. Phys. 95, 2585 ͑2004͒.
9S. B. Ogale, R. J. Choudhary, J. P. Buban, S. E. Lofland, S. R. Shinde, S.
N. Kale, V. N. Kulkarni, J. Higgins, C. Lanci, J. R. Simpson, N. D.
Browning, S. D. Sarma, H. D. Drew, R. L. Greene, and T. Venkatesan,
Phys. Rev. Lett. 91, 077205 ͑2003͒.
10J. H. Kim, H. Kim, D. Kim, A. E. Yhm, and W. K. Choo, J. Eur. Ceram.
Soc. 24, 1847 ͑2004͒.
11P. Nath and R. F. Bunshah, Thin Solid Films 80, 63 ͑1980͒.
12K. Utsumi, O. Matsunaga, and T. Takahata, Thin Solid Films 334, 30
͑1998͒.
Different magnetic mechanisms might be responsible for
observed magnetic phenomena in our samples.19 Since our
samples are highly conductive, the most “appropriate” model
that would somehow support and explain the magnetism in
our samples is carrier induced magnetism. It is known that
carriers are donated by oxygen vacancies that act as shallow
donors. Our samples were prepared under argon atmosphere,
suggesting and thus supporting our speculation that the
system must be rich in oxygen vacancies and thus itinerant
carriers. Itinerant carriers coupled to local moments lead to
the well-known Zener–Ruderman-Kittel-Kasuya-Yoshida
͑RKKY͒ mechanism of indirect magnetic interaction be-
tween the magnetic impurities with consequent induction of
13Y. K. Yoo, Q. Xue, H. C. Lee, S. Cheng, X. D. Xiang, G. F. Dionne, S. Xu,
J. He, Y. S. Chu, S. D. Preite, S. E. Lofland, and I. Takeuchi, Appl. Phys.
Lett. 86, 042506 ͑2005͒.
14J. He, S. Xu, Y. K. Yoo, Q. Xue, H. C. Lee, S. Cheng, X. D. Xiang, G. F.
Dionne, and I. Takeuchi, Appl. Phys. Lett. 86, 052503 ͑2005͒.
15N. H. Hong, J. Sakai, N. T. Huong, and V. Brizé, Appl. Phys. Lett. 87,
102505 ͑2005͒.
16J. Ederth, P. Johnsson, G. A. Niklasson, A. Hoel, A. Hultåker, P. Heszler,
C. G. Granqvist, A. R. van Doorn, M. J. Jongerius, and D. Burgard, Phys.
Rev. B 68, 155410 ͑2003͒.
17J. Philip, N. Theodoropoulou, G. Berera, and J. S. Moodera, Appl. Phys.
Lett. 85, 777 ͑2004͒.
18N. H. Hong, J. Sakai, N. T. Huong, and V. Brizý, J. Magn. Magn. Mater.
302, 228 ͑2006͒.
19M. J. Calderon and S. Das Sarma, e-print cond-mat/0603182.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
139.184.30.132 On: Mon, 01 Sep 2014 09:58:41