Room Temperature Giant Dielectric Tunability Effect in Bulk LuFe2O4

Abstract

We report the extreme sensitivity of dielectric permittivity to applied dc bias electric field in bulk LuFe2O4. A small bias field of 50 V/cm can greatly reduce the dielectric permittivity in the vicinity of room temperature, which is in strong contrast to conventional ferroelectric materials where a large electric field of the order of tens of kV/cm is required. This giant dielectric tunability effect within a broad temperature interval around room temperature is very promising for tunable device applications. The possible origins of this giant effect are discussed.

Full text

Room temperature giant dielectric tunability effect in bulk LuFe2O4
Chang-Hui Li, Xiang-Qun Zhang, Zhao-Hua Cheng, and Young Suna兲
State Key Laboratory of Magnetism and Beijing National Laboratory for Condensed Matter Physics,
Institute of Physics, Chinese Academy of Sciences, Beijing 100080, People’s Republic of China
共Received 26 March 2008; accepted 16 April 2008; published online 6 May 2008兲
We report the extreme sensitivity of dielectric permittivity to applied dc bias electric field in bulk
LuFe2O4. A small bias field of 50 V /cm can greatly reduce the dielectric permittivity in the vicinity
of room temperature, which is in strong contrast to conventional ferroelectric materials where a
large electric field of the order of tens of kV/cm is required. This giant dielectric tunability effect
within a broad temperature interval around room temperature is very promising for tunable device
applications. The possible origins of this giant effect are discussed. © 2008 American Institute of
Physics.关DOI: 10.1063/1.2920775兴
Ferroelectric materials with a high dielectric tunability,
i.e., a strong dependence of their dielectric permittivity on
the applied dc bias electric field, have potential usefulness
for many devices.1–4The dielectric tunability nis defined as
the ratio of the dielectric permittivity ␧of the material at zero
electric field to its permittivity at an applied electric field E0,
n=␧共0兲/␧共E0兲. The relative tunability 共nr兲is defined by
nr=
␧共0兲−␧共E0兲
␧共0兲.共1兲
The appropriate level of tunability for tunable device appli-
cations 共nr艌0.3兲is practically achievable in a number of
ferroelectric materials such as Sr1−xBaxTiO3,Cd
2Nd2O7, and
K共Ta1−xNbx兲O3.1,2,4,5However, a high tunability in conven-
tional ferroelectric materials usually requires a large dc elec-
tric field in the order of tens of kV/cm. For bulk materials in
the form of ceramics or crystals, the high dc electric field
means that a very high dc voltage of the order of kilovolts is
desired for efficient tuning, which apparently limits the ap-
plicability of bulk materials in tunable devices. Besides, con-
ventional ferroelectric materials usually exhibit a high but
strongly temperature dependent tunability in the vicinity of
the ferroelectric transition temperature. As a result, a high
tunability only retains in a narrow temperature interval. Fur-
thermore, ferroelectrics with a high tunability at room tem-
perature have been rarely found. In this letter, we report our
exciting finding in a bulk LuFe2O4sample which exhibits a
high dielectric tunability in quite low bias electric fields over
a broad temperature interval around room temperature.
LuFe2O4belongs to the family of the rare earth–iron
oxide, RFe2O4.6The crystal structure consists of the alternate
stacking of triangular lattices of rare earth elements, iron and
oxygen. An equal amount of Fe2+ and Fe3+ coexists at the
same site in the triangular lattice. Compared to the average
Fe valence of 2.5+, Fe2+ and Fe3+ ions are considered as
having an excess and a deficiency of half an electron, respec-
tively. Thus the interaction between Fe2+ and Fe3+ is accom-
panied by a frustration on the triangular lattice. This frustra-
tion manifests itself as a three-dimensional charge ordered
state below 330 K.7,8The strong magnetic interactions be-
tween localized Fe moments develop as a two-dimensional
ferrimagnetic ordering below 240 K.8In 2005, Ikeda et al.
first reported the ferroelectricity in bulk LuFe2O4.9It has
been generally recognized that the ferroelectricity in
LuFe2O4is correlated with the charge ordering of Fe
ions.9–11 The spontaneous polarization is generated by a polar
arrangement of the ordered Fe-3dcharges because the cen-
ters of Fe2+ and Fe3+ ions do not coincide. Such ferroelec-
tricity, termed as “electronic ferroelectricity,” is in contrast to
conventional ferroelectrics in which the displacement of cat-
ion and anion pairs plays an essential role.
Although there have been some studies on the dielectric
properties of LuFe2O4,12,13 the dielectric tunability has not
been investigated so far. In this work, we have studied the
dielectric tunability of LuFe2O4as a function of temperature
and ac field frequency. The experiments were performed on a
ceramic sample of LuFe2O4which was prepared by a solid
state reaction method. A stoichiometric mixture of high-
purity Lu2O3共99.99%兲,Fe
2O3共99.9998%兲, and Fe 共99.99%兲
metal powders was well ground. The pelletized samples were
sintered at 1100 ° C in evacuated quartz tubes for 48 h. Pow-
der x-ray diffraction showed that the samples are single
phase. The sample for dielectric measurements was prepared
by applying silver electrodes to the polished surfaces of a
thin pellet with a thickness of 1.0 mm and attaching an elec-
trode to each face by using silver paint. The dielectric re-
sponse was measured by using a NF ZM2353 LCR meter at
various frequencies 共10–200 kHz兲with an excitation of 1 V
from 100 to 320 K. Direct current bias 共DCB兲voltage was
supplied by a Keithley 2400 source meter. The temperature
during the dielectric measurements was controlled by a
Quantum Design superconducting quantum interference de-
vice.
Figure 1共a兲shows the temperature dependence of dielec-
tric permittivity of LuFe2O4under zero bias field. The per-
mittivity shows a small value close to zero at temperatures
well below the ferroelectric transition. With increasing tem-
perature, it shows a sudden rise and reaches a large value
共⬎10 000兲around room temperature. As observed in previ-
ous reports, our sample also exhibits a strong frequency-
dependent dielectric response 共dielectric dispersion兲. The
sharp transition shifts to higher temperatures with increasing
frequency. This dispersion is also clear in the dielectric loss
response. As shown in Fig. 1共b兲, the peak of loss tangent,
tan
␦
, monotonically shifts to higher temperatures with in-
creasing frequency of the ac field. The large dielectric dis-
persion in LuFe2O4was interpreted in terms of the electron
a兲Author to whom correspondence should be addressed. Electronic mail:
youngsun@aphy.iphy.ac.cn.
APPLIED PHYSICS LETTERS 92, 182903 共2008兲
0003-6951/2008/92共18兲/182903/3/$23.00 © 2008 American Institute of Physics92, 182903-1
Author complimentary copy. Redistribution subject to AIP license or copyright, see http://apl.aip.org/apl/copyright.jsp

hopping 共fluctuation兲on iron ion which causes the motion of
ferroelectric domain boundary.9
Figure 2shows the dielectric permittivity of LuFe2O4at
100 kHz as a function of temperature under zero and a 5 V
DCB voltage. A large difference between two curves appears
above 250 K as the permittivity is greatly suppressed in a
5 V dc bias voltage, indicating a high tunability. This result
is striking because a 5 V dc bias voltage applied on a 1 mm
thick sample only generates a small bias electric field of
50 V/cm. Normally, such a small field would cause negli-
gible changes in dielectric permittivity for conventional fer-
roelectrics since the required bias field for noticeable
changes is in the order of kV/cm. To confirm this extreme
sensitivity of ␧on bias electric field, we further measured ␧
as a function of dc bias voltage with a cycle from positive to
negative 5 V at 300 K. As shown in the inset of Fig. 2, the
permittivity quickly changes with either positive or negative
dc bias voltages. The hysteresis observed in the field depen-
dence of ␧usually indicates the ferroelectric ordering and the
presence of domains.14,15 A 5 V bias voltage is able to sup-
press ␧from 12 000 to 5800, more than 50%. Therefore,
these results unambiguously demonstrate a giant dielectric
tunability effect in bulk LuFe2O4.
In order to gain the details of the variation of ␧with
DCB and ac frequency, we performed a series of measure-
ments of permittivity as a function of DCB at various tem-
peratures and frequencies. Figure 3presents the representa-
FIG. 2. 共Color online兲Influence of dc bias voltage on the dielectric permit-
tivity of LuFe2O4measured at 100 kHz. The inset shows the variation of
permittivity as a function of positive and negative dc bias voltages at 300 K.
FIG. 1. 共Color online兲Temperature dependence of 共a兲dielectric permittivity
and 共b兲dielectric loss tangent of LuFe2O4at various frequencies from
10 to 200 kHz.
FIG. 3. 共Color online兲Dielectric permittivity of LuFe2O4as a function of dc
bias field at various temperatures between 200 and 310 K. For comparison,
all data are normalized by the initial permittivity in zero dc bias field, ␧共0兲,
at each temperature. The data shown here are collected at three typical ac
frequencies of 共a兲10 kHz, 共b兲100 kHz, and 共c兲200 kHz.
182903-2 Li et al. Appl. Phys. Lett. 92, 182903 共2008兲
Author complimentary copy. Redistribution subject to AIP license or copyright, see http://apl.aip.org/apl/copyright.jsp

tive results at three frequencies, 10, 100, and 200 kHz. For
comparison, the data are normalized by the initial permittiv-
ity in zero DCB, ␧共0兲. At low temperatures where the tun-
ability is small, ␧almost linearly varies with increasing bias
field. At higher temperatures, the variation shows apparent
curvature, with a slow change in initial low fields, a fast
change in the intermediate field, and a slow down in high
fields. The permittivity is more sensitive to dc bias field at
lower frequency. At 10 kHz, a small dc bias field of
50 V/cm is able to suppress the permittivity by 80% at
240 K. Figure 4shows the relative tunability nr, defined in
Eq. 共1兲, under a 50 V /cm bias field as a function of tempera-
ture at different frequencies. Unlike conventional ferroelec-
tric materials where the tunability is strongly temperature
dependent and shows a sharp peak in a narrow temperature
range, the tunability in LuFe2O4exhibits a broad peak so that
a high tunability retains in a broad temperature interval. Such
a broad operating temperature interval around room tempera-
ture is very promising for applications. With increasing fre-
quency, the maximum value of nrdecays, while the broad
peak shifts to higher temperatures. Interestingly, although nr
has a strong frequency dependence at low temperature, it is
nearly the same around room temperature.
In addition to a high tunability, a low dielectric loss is
also necessary for tunable device applications. The loss tan-
gent in our sample is around 0.2–0.5 at room temperature
and increases in a dc bias field. This seems to be the only
disadvantage of bulk LuFe2O4for tunable applications.
However, the dielectric loss depends on many extrinsic fac-
tors such as defects.1A much lower loss is usually found in
high-quality single crystals. Moreover, there have been some
strategies to effectively reduce the dielectric loss, for ex-
ample, by making ferroelectric-dielectric composites and
layered structures.1Therefore, the relative high dielectric
loss in the polycrystalline LuFe2O4sample may not become
a significant obstacle for practical applications.
The physical origin of the giant dielectric tunability ef-
fect in bulk LuFe2O4is still a puzzle. On one side, this giant
effect could be intrinsically related to the spectacular elec-
tronic ferroelectricity in LuFe2O4. In general, the reduction
of dielectric permittivity under a dc bias field is mainly due
to the suppression of polarization fluctuation. In conven-
tional ferroelectrics where the polarization arises from cation
and anion pairs, this process involves the displacement of
ions and lattice distortion, which usually requires a high en-
ergy. As a result, a high electric field is required to induce a
noticeable reduction in permittivity. In contrast, the elec-
tronic ferroelectricity in LuFe2O4arises from a polar ar-
rangement of the ordered Fe-3dcharges instead of cation and
anion pairs. Therefore, the polarization fluctuation in
LuFe2O4is directly correlated with the charge fluctuation of
3-delectrons of iron. The energy required for charge motion
of 3-delectrons could be much lower than the case of ion
displacement. Consequently, a small electric field could
sharply reduce the charge fluctuation on Fe ions, leading to a
large reduction in permittivity. On the other side, we cannot
exclude extrinsic contributions to this giant effect at this
stage. It has been proposed that the Schottky barriers at the
electrode interfaces could introduce a remarkably strong
change in dielectric constant at very small applied fields.16
We have repeated the experiments on several samples and
found that the dielectric tunability depends on the thickness
of samples and the quality of electrodes,17 which indicates
that the Schottky barriers introduced by electrodes could
play a role. Although the present experiments unambigu-
ously demonstrate the giant dielectric tunability at low fre-
quencies 共f艋200 kHz兲, it is not clear at this stage whether
this giant effect would remain in microwave frequencies.
Unfortunately, we are currently unable to measure the dielec-
tric response at higher ac frequencies due to the limitation of
apparatus. The dielectric properties of LuFe2O4in micro-
wave frequencies, as well as the underlying mechanism of
this giant effect, certainly deserve further studies.
The authors are grateful to Mr. Shuo Liang and Professor
J. R. Sun for the help in dielectric response measurements.
This work was supported by the National Key Basic Re-
search Program of China 共2007CB925003兲and the National
Natural Science Foundation of China 共50721001兲.
1A. K. Tagantsev, V. O. Sherman, K. F. Astafiev, J. Venkatesh, and N.
Setter, J. Electroceram. 11,5共2003兲.
2O. G. Vendik, E. K. Hollmann, A. B. Kozyrev, and A. M. Prudan, J.
Supercond. 12, 325 共1999兲.
3K. M. Johnson, J. Appl. Phys. 33, 2826 共1962兲.
4A. Feteira, D. C. Sinclair, I. M. Reaney, Y. Somiya, and M. T. Lanagan, J.
Am. Ceram. Soc. 87, 1082 共2004兲.
5Z. Yu, C. Ang, R. Guo, and A. S. Bhalla, Appl. Phys. Lett. 81, 1285
共2002兲.
6N. Kimizuka, E. Muromachi, and K. Siratori, in Handbook on the Physics
and Chemistry of Rare Earths, edited by K. A. Gschneidner and L. Eyring
共Elsevier Science, Amsterdam, 1990兲, Vol. 13.
7Y. Yamada, S. Nohdo, and N. Ikeda, J. Phys. Soc. Jpn. 66,3733共1997兲.
8Y. Yamada, S. Kisuda, S. Nohdo, and N. Ikeda, Phys. Rev. B 62, 12167
共2000兲.
9N. Ikeda, H. Ohsumi, K. Ohwada, K. Ishii, T. Inami, K. Kakurai, Y.
Murakami, K. Yoshii, S. Mori, Y. Horibe, and H. Kito, Nature 共London兲
436, 1136 共2005兲.
10Y. Zhang, H. X. Yang, C. Ma, H. F. Tian, and J. Q. Li, Phys. Rev. Lett. 98,
247602 共2007兲.
11H. J. Xiang and M. H. Whangbo, Phys. Rev. Lett. 98, 246403 共2007兲.
12K. Yoshii, N. Ikeda, Y. Matsuo, Y. Horibe, and S. Mori, Phys. Rev. B 76,
024423 共2007兲.
13J. Y. Park, J. H. Park, Y. K. Jeong, and H. M. Jang, Appl. Phys. Lett. 91,
152903 共2007兲.
14H. Li, W. Si, A. D. West, and X. X. Xi, Appl. Phys. Lett. 73,190共1998兲.
15J. Yang, J. Chu, and M. Shen, Appl. Phys. Lett. 90, 242908 共2007兲.
16N. Biškup, A. de Andrés, J. L. Martines, and C. Perca, Phys. Rev. B 72,
024115 共2005兲.
17The thickness and electrode dependence of the dielectric response and
tunability in LuFe2O4will be reported in details elsewhere.
FIG. 4. 共Color online兲Temperature dependence of relative tunability nr
under a 50 V /cm dc bias field at various frequencies between 10 and
200 kHz.
182903-3 Li et al. Appl. Phys. Lett. 92, 182903 共2008兲
Author complimentary copy. Redistribution subject to AIP license or copyright, see http://apl.aip.org/apl/copyright.jsp