A key to room-temperature ferromagnetism in Fe-doped ZnO: Cu

Article abstract of DOI:10.1063/1.1525885

The synthesis of ferromagnetic semiconductors, Zn_1-xFe_xO, at room temperature was reported. Additional Cu doping was used to achieve room-temperature ferromagnetism in the ferromagnetic semiconductor. The saturation magnetization at room temperature reached a value of 0.75 μ_B per Fe and a transition temperature of 550 K was obtained in Zn_0.94Fe_0.05Cu_0.01O.

Full text of DOI:10.1063/1.1525885

A key to room-temperature ferromagnetism in Fe-doped ZnO: Cu
S-J. Han, J. W. Song, C.-H. Yang, S. H. Park, J.-H. Park, Y. H. Jeong, and K. W. Rhie
Citation: Applied Physics Letters 81, 4212 (2002); doi: 10.1063/1.1525885
View online: http://dx.doi.org/10.1063/1.1525885
View Table of Contents: http://scitation.aip.org/content/aip/journal/apl/81/22?ver=pdfcov
Published by the AIP Publishing
Articles you may be interested in
Extrinsic origin of room-temperature ferromagnetism in Co-doped ZnO annealed in Zn vapor
Appl. Phys. Lett. 99, 082501 (2011); 10.1063/1.3624601
Robust room temperature ferromagnetism in Cu doped ZnO thin films
Appl. Phys. Lett. 99, 042504 (2011); 10.1063/1.3615714
A possibility to obtain room temperature ferromagnetism by transition metal doping of ZnO nanoparticles
J. Appl. Phys. 107, 053917 (2010); 10.1063/1.3329457
Where does the spin reside in ferromagnetic Cu-doped ZnO?
Appl. Phys. Lett. 91, 012501 (2007); 10.1063/1.2751587
Room Temperature Magnetooptics Of Ferromagnetic ZnO Doped With Transition Metals And Aluminum
AIP Conf. Proc. 893, 1189 (2007); 10.1063/1.2730323
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:
131.156.157.31 On: Mon, 24 Nov 2014 14:09:02

APPLIED PHYSICS LETTERS
VOLUME 81, NUMBER 22
25 NOVEMBER 2002
A key to room-temperature ferromagnetism in Fe-doped ZnO: Cu
S-J. Han, J. W. Song, C.-H. Yang, S. H. Park, J.-H. Park, and Y. H. Jeonga)
Department of Physics and Electron Spin Science Center, Pohang University of Science and Technology,
Pohang, 790-784, South Korea
K. W. Rhie
Department of Physics, Korea University, Chochiwon 339-700, South Korea
͑Received 20 May 2002; accepted 5 October 2002͒
Successful synthesis of room-temperature ferromagnetic semiconductors, Zn1ϪxFexO, is reported.
The essential ingredient in achieving room-temperature ferromagnetism in bulk Zn1ϪxFexO was
found to be additional Cu doping. A transition temperature as high as 550 K was obtained in
Zn0.94Fe0.05Cu0.01O; the saturation magnetization at room temperature reached a value of 0.75␮B per
Fe. A large magnetoresistance was also observed below 100 K. © 2002 American Institute of
Physics. ͓DOI: 10.1063/1.1525885͔
Diluted magnetic semiconductors ͑DMSs͒ have attracted
a great deal of attention because of the possibility of incor-
porating the magnetic degrees of freedom in traditional
semiconductors.1,2 DMSs combine their transport and/or op-
tical properties with magnetism, and thereby carry an enor-
mous potential of opening up a path to entirely new devices.
Until recently, ͑Ga,Mn͒As has been a representative DMS
with its moderately high Curie point ͑maximum TC
Ϸ110 K).3 However, an essential task in a realization of the
potential is to find a DMS with the Curie point above room
temperature.4–6
Recent theoretical works predicted ferromagnetism
above room temperature in a II–VI semiconductor ZnO, nor-
mally n-type, when doped with magnetic impurities.7,8 Since
ZnO is optically transparent, ferromagnetic ZnO would be a
transparent magnet as well. Despite intensive efforts on tran-
sition metal-doped ZnO thin films, the experimental results
did not converge on a definite conclusion; there are, for in-
stance, contradicting reports on Co-doped ZnO thin films.9,10
Even the successful report, in which Co-doped ZnO thin
films showed TC of about 300 K, was attached with the res-
ervation that the reproducibility was less than 10%.10 Sur-
veying the current situation with transition metal-doped
ZnO, one naturally comes to a suspicion that the inconsistent
results in thin film DMSs, within a group or among different
groups, might be due to sensitive dependence of thin films on
detailed process conditions such as target qualities, sub-
strates, growing temperatures, oxygen pressure, etc. These
thoughts motivated us to probe bulk samples rather than thin
films.
be used as an additional p-type dopant into naturally n-type
ZnO samples.11 The idea of additional Cu-doping in
Zn1ϪxFexO was highly successful and led us to a room-
temperature ferromagnetic DMS. In view of the inconsistent
film results, reproducibility was ascertained by measuring
several samples synthesized by the same procedures.
Polycrystalline samples were fabricated with the stan-
dard solid state reaction method in Ar-filled quartz tubes.
High purity ZnO(99.99ϩ%), FeO(99.9ϩ%), and
CuO(99.99ϩ%) powders were mixed thoroughly and pro-
cessed at 1170 K for 24 h. The single phase nature of
samples was checked by ␪–2␪ x-ray diffraction ͑XRD͒ using
a Cu K␣ source; substantial amount of exposure time was
allowed in the XRD scans to check even a minute amount of
a secondary phase. Magnetization, resistivity, Hall coeffi-
cient, and thermopower were measured by employing two
pieces of equipment, a magnetic property measurement sys-
tem ͑MPMS͒ and a physical property measurement system
͑PPMS͒, manufactured by Quantum Design. The supercon-
ducting quantum interference device magnetometer ͑MPMS͒
was equipped with a high temperature oven facility. X-ray
absorption spectroscopy ͑XAS͒ was carried out at the
Dragon beamline of the Synchrotron Radiation Research
Center in Taiwan.
Figure 1͑a͒ shows a typical powder XRD pattern of
Zn1ϪxFexO for xϭ0.07. All the peaks belong to the hexago-
nal lattice of ZnO, and no indication of a secondary phase is
found. A shift of XRD peak positions related to lattice spac-
ing changes was clearly observed when the concentration of
Fe was varied; the data refinement revealed that the shift was
caused by a variation of lattice spacing a as displayed in Fig.
1͑b͒. The linear expansion of the a-axis lattice spacing with
increasing x indicates that doped Fe atoms substitute for Zn
atoms in the lattice up to xϭ0.07 under the current process-
ing condition. The incorporation of Fe atoms into the lattice
was also evidenced from the XAS measurements, which
yielded the oxidation state of Fe to be mostly Fe2ϩ. On
further doping of Fe above xϷ0.1, the system enters a co-
existence region of the dominant hexagonal phase ͑Fe-doped
ZnO͒ and a minor cubic phase ͑FeO͒. It should be noted here
that the Fe solubility of bulk ZnO found in this work is
In this letter, we concentrate on Fe-doped ZnO bulk
samples. As described later, Fe-doping alone turned out to be
not sufficient for room temperature ferromagnetism in ZnO,
and a third element was required. As one member of the
transition metal group, Cu substitution may be considered as
magnetic doping if Cu substitutes for Zn2ϩ as Cu2ϩ. Cu-
doping, however, did not induce a significant change in the
magnetic property of ZnO films.9 On the other hand, Cu may
a͒
Author to whom correspondence should be addressed; electronic mail:
yhj@postech.ac.kr
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:
0003-6951/2002/81(22)/4212/3/$19.00 4212 © 2002 American Institute of Physics
131.156.157.31 On: Mon, 24 Nov 2014 14:09:02

Appl. Phys. Lett., Vol. 81, No. 22, 25 November 2002
Han et al.
4213
FIG. 3. The magnetization of Zn0.94Fe0.05Cu0.01O is plotted as a function of
temperature. The strength of the applied field was 500 Oe.
Figs. 2͑b͒, 2͑c͒, and 2͑d͒. The magnetization is greatly en-
hanced with Cu-doping, so that Ms at room temperature of
the sample with 1% Cu (0.75␮B /Fe) becomes 30 times
larger than that of the sample without Cu. The fact that Ms
increases systematically only with Cu contents virtually
eliminates the possibility of Fe ͑or iron oxide͒ clusters being
responsible for magnetization. The saturation occurs at
2 kOe, while the coercive field is about 20 Oe at room tem-
perature. The small coercivity of a polycrystalline sample
indicates the intrinsically soft nature of this material.
Now that we have established ferromagnetism in
Zn0.95ϪyFe0.05CuyO at room temperature, its Curie tempera-
ture is of interest. In order to determine TC of
Zn0.94Fe0.05Cu0.01O, which shows the largest Ms at 300 K in
Fig. 2͑c͒, its M was measured as a function of temperature at
a small field ͑500 Oe͒. The measured M, presented in Fig. 3,
clearly shows that a transition to a paramagnetic state occurs
near 550 K. It is noted that the transition is rather sharp and
appears to deviate from a mean field description. The transi-
tion temperature of 550 K is considerably higher than the
Curie temperature of most DMSs. More importantly, it is
high enough for the purpose of device applications at room
temperature.
FIG. 1. ͑a͒ Intensity of XRD vs 2␪, from a polycrystalline Zn1ϪxFexO
sample with xϭ0.07. All the peaks belong to the hexagonal structure of
ZnO. ͑b͒ Variation of the lattice constant a vs Fe concentration x.
higher than that of thin films.9 The low solubility of the films
seems to be related to the fact that it is the c-axis lattice
constant, not the a-axis which is constrained by a substrate,
that varies as the contents of Fe changes. This is a manifes-
tation of the difference between bulk and thin films, which
will be discussed elsewhere.12
We now turn to the magnetic properties of Zn1ϪxFexO.
When measured at room temperature, a sample with x
ϭ0.05 showed a maximum saturation magnetization (Ms).
However, the measured Ms of Zn0.95Fe0.05O (0.025␮B /Fe,
␮BϭBohr magneton͒ was far too small to be considered as a
room temperature ferromagnet. The situation is evident in
Fig. 2͑a͒, showing the magnetization–field (M –H) curve of
Zn0.95Fe0.05O sample at 300 K. We then attempted additional
Cu-doping into Zn0.95Fe0.05O as a means of rendering re-
spectable ferromagnetism to the system. According to Dietl
et al., hole doping is most effective in achieving carrier-
mediated ferromagnetism in semiconductors.7 Jun and Choi
showed that introduction of CuO into n-type ZnO increases
resistivity up to 1% mixture,11 which suggests that Cu can
In addition to ferromagnetism, transport properties are
of exceeding importance in DMS. Figure 4͑a͒ is the plot of
the resistivity of Zn0.94Fe0.05Cu0.01O at zero field as a func-
tion of temperature. The resistivity displays a typical semi-
conducting behavior. At room temperature the resistivity
play the role of
a p-type dopant. Cu-doping into
Zn0.95Fe0.05O via the standard ceramic method neither caused
any structure change nor induced a secondary phase up to
1% incorporation. A small amount of additional Cu-doping,
however, brought about drastic changes in M as illustrated in
has a value of 0.1 ⍀ cm. Magnetoresistance MRϵ ␳(H)
͓
͕
FIG. 2. Room temperature M –H curves of bulk Zn0.95ϪyFe0.05CuyO: ͑a͒
yϭ0, ͑b͒ 0.002, ͑c͒ 0.005, and ͑d͒ 0.01. The vertical scale ͑magnetic
moment/Fe͒ was converted from measured values, assuming homogeneous
FIG. 4. Transport properties of Zn0.94Fe0.05Cu0.01O: ͑a͒ The resistivity vs
temperature, ͑b͒ Magnetoresistance ͑MR͒ at various temperatures. MR is
defined as ␳(H)Ϫ␳(0) /␳(0). At 100 K, the MR is insignificant.
͓ ͔
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:
magnetic states.
131.156.157.31 On: Mon, 24 Nov 2014 14:09:02

4214
Appl. Phys. Lett., Vol. 81, No. 22, 25 November 2002
Han et al.
In conclusion, we achieved ferromagnetism in
Zn1ϪxFexO by Cu-doping, and the Curie temperature of 550
K and the saturation magnetization of 0.75␮B /Fe were ob-
tained for Zn0.94Fe0.05Cu0.01O.
Ϫ␳(0) /␳(0) was also measured at various temperatures.
͔
͖
The measured MR of a bulk sample is very large and posi-
tive, as shown in Fig. 4͑b͒. However, the MR of
Zn0.94Fe0.05Cu0.01O becomes insignificant above 100 K, even
if the system is still ferromagnetic. At present, we have no
understanding on this peculiar behavior.
This work was supported by the SRC program of KO-
SEF and the BK21 program of MOE. J.-H. Park thanks Dr.
H.-J. Lin and Dr. C.T. Chen of the SRRC in Taiwan for the
beam time.
In order to elucidate the origin of ferromagnetism, we
also measured the thermopower and the Hall coefficient of
Zn0.95Fe0.05O and Zn0.94Fe0.05Cu0.01O. Both samples yielded
negative values for thermopower as well as Hall coefficient,
and no anomalous Hall effect was observed. These results,
first of all, indicate that electrons are major carriers in the
samples. The number of carriers at room temperature esti-
mated from the Hall coefficients are 4.2ϫ1017/cm3 and 5.0
ϫ1017/cm3 for Zn0.94Fe0.05Cu0.01O and Zn0.95Fe0.05O, re-
spectively. XAS measurements also showed that the valence
1 J. K. Furdyna, J. Appl. Phys. 64, R29 ͑1988͒.
2 S. A. Wolf, D. D. Awschalom, R. A. Buhrman, J. M. Daughton, S. von
Molnar, M. L. Roukes, A. Y. Chtchelkanova, and D. M. Treger, Science
294, 1488 ͑2001͒.
3 H. Ohno, J. Magn. Magn. Mater. 200, 110 ͑1999͒.
4 G. A. Medvedkin, T. Ishibashi, T. Nishi, K. Hayata, Y. Hasegawa, and K.
Sato, Jpn. J. Appl. Phys. 39, L949 ͑2000͒.
5 Y. Matsumoto, M. Murakami, T. Shono, T. Hasegawa, T. Fukumura, M.
Kawasaki, P. Ahmet, T. Chikyow, S. Koshihara, and H. Koinuma, Science
291, 854 ͑2001͒.
state of Cu in Zn0.94Fe0.05Cu0.01O is Cu1ϩ, rather than Cu2ϩ
.
Thus, Cu ions play the role of acceptors and reduces the
number of electron carriers, although they are not able to
change the n-type character itself. It is then evident that the
ferromagnetism in Zn0.94Fe0.05Cu0.01O ͑which appears with
reduction in the number of electrons͒ is not mediated by
electrons. As Sato et al. suggested, it is very likely that a 3d
band is formed by doped Fe atoms, and ferromagnetism oc-
curs within this band via double exchange.8 The lack of
anomalous Hall effect is probably due to a weak coupling of
the conduction electrons, which dominate transport, to mag-
netic degrees of freedom.
6 M. L. Reed, N. A. El-Masry, H. H. Stadelmaier, M. K. Ritums, M. J. Reed,
C. A. Parker, J. C. Roberts, and S. M. Bedair, Appl. Phys. Lett. 79, 3473
͑2001͒.
7 T. Dietl, H. Ohno, F. Matsukura, J. Cibert, and D. Ferrand, Science 287,
1019 ͑2000͒.
8 K. Sato and H. Katayama-Yoshida, Jpn. J. Appl. Phys. 39, L555 ͑2000͒.
9 Z. Jin, T. Fukumura, M. Kawasaki, K. Ando, H. Saito, T. Sekiguchi, Y. Z.
Yoo, M. Murakami, Y. Matsumoto, T. Hasegawa, and H. Koinuma, Appl.
Phys. Lett. 78, 3824 ͑2001͒.
10 K. Ueda, H. Tabata, and T. Kawai, Appl. Phys. Lett. 79, 988 ͑2001͒.
11 S.-T. Jun and G.-M. Choi, J. Am. Chem. Soc. 81, 695 ͑1998͒.
12 J. W. Song, C. H. Yang, S-J. Han, and Y. H. Jeong ͑unpublished͒.
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:
131.156.157.31 On: Mon, 24 Nov 2014 14:09:02