High dielectric permittivity behavior in cu-doped CaTiO3

Article abstract of DOI:10.1111/j.1551-2916.2009.03254.x

Cu-doped CaTiO₃-based polycrystalline ceramics have been prepared by the conventional solid-state sintering. Our results indicate that the dielectric constant can be enhanced greatly by increasing the Cu-doped content, which show weak frequency

Full text of DOI:10.1111/j.1551-2916.2009.03254.x

J. Am. Ceram. Soc., 92 [11] 2776–2779 (2009)
DOI: 10.1111/j.1551-2916.2009.03254.x
r 2009 The American Ceramic Society
ournal
J
High Dielectric Permittivity Behavior in Cu-Doped CaTiO3
Bo Cheng,z Yuan-Hua Lin,w,z Hao Yang,z Jinle Lan,z Ce-Wen Nan,z Xi Xiao,y and Jinliang Hey
zState Key Laboratory of New Ceramics and Fine Processing, Department of Material Science and Engineering,
Tsinghua University, Beijing 100084, China
yDepartment of Electrical Engineering, Tsinghua University, Beijing 100084, China
Cu-doped CaTiO3-based polycrystalline ceramics have been
In this work, we prepared Cu-doped CaTiO3 polycrystalline
ceramics, and observed high dielectric permittivity, and low loss
tangent behavior. The dielectric constant can be tuned by the
addition of Cu, and shows good temperature and frequency
stability. Our results indicate that it is a promising material for
capacitors and memory devices.
prepared by the conventional solid-state sintering. Our results
indicate that the dielectric constant can be enhanced greatly by
increasing the Cu-doped content, which show weak frequency
and temperature dependence. The fitted activation energy is
almost same (B0.10 eV) as the Cu-doped content is 0.4–0.6,
which may be ascribed to the first ionization of the oxygen va-
cancies. The origin of the high dielectric permittivity observed in
these Ca1ꢀxCuxTiO3-based ceramics should be attributed to the
interfacial polarization mechanism, and can be well described by
the percolation theory with fcꢁ0.27 and sꢁ0.74.
CaCO3, CuO, and TiO2 powders were used as the raw mate-
rials (all reagents are analytical purity), and Ca1ꢀxCuxTiO3
ceramics (x 5 0, 0.1, 0.4, 0.5, and 0.6, abbreviated as CTO-0,
CTO-1, CTO-2, CTO-3, and CTO-4, respectively) were prepared
by a solid-state reaction sintering method. The starting materials
were weighted as the above nominal composition and milled for
6 h. The dried mixture powders were presintered at 9001C for 2 h,
and then the precursor powders were pressed to green pellets (12
mm in diameter) with polyvinyl alcohol binder. Finally, the pel-
lets were sintered at 11001C for 3 h in air. The phase composi-
tions of these as-sintered samples were measured by the X-ray
diffraction (XRD) equipment (Rigaku D-Max 3A, Suginami-ku,
Tokyo, Japan, CuKa radiation). Scanning electron microscopy
(SEM) equipped with X-ray energy-dispersive spectrometer
(EDS) and electron back scatter diffraction (EBSD) was used
to study the microstructure and the composition of the samples.
The ceramic samples were polished and pasted by silver paste on
both sides, and then treated at 6001C for 30 min to form the
electrodes. The dielectric response of the specimens was mea-
sured using a HP 4194A gain-phase analyzer (Santa Clara, CA)
over a frequency range from 100 Hz to 1 MHz and at an oscil-
lation voltage of 1 V. These measurements were performed in the
temperature range from ꢀ120 to 350 K. Each measured temper-
ature was kept constant with an accuracy of 71 K.
Figure 1 show XRD patterns of the CTO-based samples,
which indicate that two main phases CTO and CCTO can be
observed. The relative intensity of the characteristic peaks of the
CCTO phase increases with the doped concentration of Cu.
SEM images and related EDS analysis results (not shown here)
also show that ceramics are composed of CTO and CCTO grain
particles.
As shown in Fig. 2, with the addition of a little amount of Cu
in the pure CTO ceramic, the dielectric constant increases ob-
viously, especially for CTO-4 ceramic, and can be enhanced up
to about 40 times higher than that of pure CTO ceramics at
1 kHz. Of interest to note, the dielectric losses still remain low
(o3% at 1 kHz), and show weak frequency dependence for
various CTO-based ceramic samples. Another intriguing fea-
ture of the Cu-doped CTO-based samples is the very weak
dependence of the dielectric constants on temperature as shown
in Fig. 3, and these samples show excellent temperature stability
on the dielectric property over a wide-temperature range, which
is of technological importance for applications in a reproducible
electronic device.
ITH the miniaturization and integration of electronic
devices, high-permittivity dielectric materials have
W
attracted considerable attention for both scientific under-
standing and their numerous technological applications such
as capacitors and memory devices.1 Usually, high-dielectric
properties are observed in the perovskite ferroelectric or re-
laxor oxides, e.g., Pb(Zr,Ti)O3 and Pb(Mg,Nb)O3.2,3 How-
ever, both kinds of materials show strong temperature
dependence due to the ferroelectric phase transition, and
most of such perovskite oxides contain Ba/Pb, which are in-
dispensable to modern electronic devices. Surface and internal
barrier layer capacitors based on semiconductive perovskite
such as (Ba,Sr)TiO3 ceramics have also been attractive.4 How-
ever, these capacitors are of poor reproducibility and complex
processing (involving high temperatures, reducing atmo-
spheres, and limited diffusion of oxygen and dopant ions
along the grain boundaries), and inherit a strong variation in
electrical properties with temperature and frequency, which is
unfavorable for many applications.
Recently, a lead-free perovskite-like oxide CaCu3Ti4O12
(CCTO),5,6 and Li, Ti co-doped NiO (LTNO)7,8 ceramics have
been reported to possess an extraordinarily high dielectric con-
stant of B104–105 at room temperature, which is almost con-
stant over a wide temperature and frequency range. However,
both of CCTO and LTNO materials have not yet been obtained
applications because of their large dielectric loss around room
temperature (40.10 at 1 kHZ). Therefore, creation of dielectric
materials with a lead-free, high dielectric constant, low loss tan-
gent, and good stability in wide temperature and frequency
ranges are highly desired.
X. M. Chen—contributing editor
Manuscript No. 26127. Received April 14, 2009; approved June 1, 2009.
This work was supported by the Ministry of Science and Technology of China through a
973-Project under Grant No. 2009CB623303, NSF of China (50737001, and 50677029), and
NSF of Beijing.
In order to further understand the physical nature of the
dielectric behavior in this CTO system mentioned above, we
measured the frequency dependence of dielectric permittivity
at various temperatures shown in Fig. 4(a). Generally, the
wAuthor to whom correspondence should be addressed. e-mail: linyh@mail.tsinghua.
edu.cn
2776

November 2009
Communications of the American Ceramic Society
2777
4500
4000
3500
3000
2500
2000
1500
1000
500
0.30
@ 1KHz
0.25
0.20
0.15
0.10
0.05
0
100
150
200
250
300
350
Temperature (K)
Fig. 3. Temperature dependence of the dielectric constant and dielec-
tric loss for various Ca1ꢀxCuxTiO3-based ceramic samples.
where Ea is the activation energy, t0 represents the preexponen-
tial factor, and kB is the Boltzmann constant.
Fig. 1. X-ray diffraction patterns of various Cu-doped CaTiO3 ceramic
samples. CCTO, CaCu3Ti4O12.
As shown in Fig. 4(b), a good linear regression of log t vs 1/T
can be obtained, which yields Ea 5 0.101 eV. Similarly, we can
also get the other two samples’ activation energy (Ea) 5 0.103 eV
dielectric relaxation can be represented by a Debye relaxation
relation as the following9:
es ꢀ e
1
eꢂðoÞ ¼ e0ie00 ¼ e1 þ
(1)
130 K
150 K
200 K
240 K
300 K
320 k
3000
2500
2000
1500
1000
500
1 ꢀ iot
CTO-3
es ꢀ e
1
e0ðoÞ ¼ e1 þ
(2)
2
1 þ ðotÞ
where e0 is the real part of the dielectric constant, e00 is the imag-
inary part of dielectric constant, o is the angular frequency, e0
and eN are static dielectric constant and permanent dielectric
constant, respectively, t is the relaxation time.
Therefore, the relaxation time can be obtained from the fitted
dielectric permittivity curves by Eq. (2). As shown in the inset of
Fig. 4(b), the rapid decrease of t with increasing temperature can
be observed, which is suggestive of an increasing dipole density
and a faster polarization process. Actually, the variation of re-
laxation time t with 1/T can be fitted with the Arrhenius law:
0
10k
100k
1M
Frequency (Hz)
t ¼ t0 expðEa=kBTÞ
(3)
–4.0
–4.5
–5.0
–5.5
–6.0
–6.5
–7.0
CTO-3
E
= 0.101 eV
a
2
3
4
5
6
7
8
9
Fig. 4. (a) Frequency dependence of for CTO-based ceramics measured
at different temperatures. The solid lines represent the calculated values
by a Debye model. (b) The log of the relaxation time t vs 1/T. The
symbols are the experimental points and solid lines are the fitted line by
Eq. (3). CTO, Ca1ꢀxCuxTiO3.
Fig. 2. Frequency dependence of dielectric constant of various samples.
CCTO, CaCu3Ti4O12
.

2778
Communications of the American Ceramic Society
Vol. 92, No. 11
and 0.100 eV for CTO-2 and CTO-4, respectively. As reported
previously,10–12 in the perovskite oxides, the activation energy
for the first ionization of oxygen vacancies is about 0.10 eV, near
our fitted activation energy for the conduction in the measured
temperature region, which indicates that the first ionization of
the oxygen vacancies should be responsible for the conduction
for the CCTO of this system.
1300
1280
1260
1240
1220
1200
1180
1160
1140
It is generally believed that pure CTO is an incipient ferro-
electric with orthorhombic (space group Pcmn) structure in a
broad temperature range, and its dielectric constant is about 100
at room temperature. However, as shown in Fig. 3, the dielectric
constant of the CCTO–CTO composite ceramics increases with
CCTO concentration. For example, the dielectric constant of the
CTO-4 ceramic (CCTO volume ratio B30%) is about 40 times
higher than that of the pure CTO ceramic. Why the addition of
Cu can improve the dielectric properties greatly in these CTO-
based ceramics? Actually, a lot of previous experiments demon-
strated that CCTO ceramic is electrically heterogeneous and
consists of semiconducting or conducting grains with insulating
grain boundaries.13,14 Two plausible models were also estab-
lished to explain CCTO semiconducting grain, one based on
oxygen loss, and the other based on cation nonstoichiometry.15
Formation of CCTO semiconducting grain is explained by Cu
oxidation process. During the heating process, Cu21 becomes
unstable and reduced to Cu11, so Ti41 substitute in Cu site to
maintain oxidation state of this Cu site. Therefore, the ionic
930 931 932 933 934 935 936 937 938
BE
2400
2200
2000
1800
1600
1400
1200
1000
800
formula of CCTO can be expressed as Ca21(Cu Cu11
21
1-3x 2x
Tix41)3Ti441O12 at high temperature. And during the cooling
process, Cu11 is oxidized to Cu21. In this time, electrons move
into Ti 3d band and finally forms Ca21(Cu Tix
)
3
21
1ꢀx
41
31 41
(Ti Ti )O12 in low temperature. So Ti31 represents charge
6x 4ꢀ6x
carier and exhibits semiconductive grain16,17 (as shown in
Fig. 6). Recently, Chung et al.18 measured directly the local
conduction behaviors in the CCTO ceramic by probe force mi-
croscopy, and found that CCTO grains have high conductivity
at room temperature (Bseveral hundreds O ꢃ cm). Therefore,
such a large increase in the dielectric constant should be caused
by the existence of a large number of high-conductivity CCTO
particles in parallel and in very close proximity, but blocked by
the insulating CTO phase. Some researchers have described
the dielectric behavior of the composite system with a conduct-
ing phase dispersed in an insulating matrix. The dielectric be-
havior in these composites can be well described by percolation
theory19–21 by the following equation:
456
458
460
462
BE
464
466
468
Fig. 6. The XPS patterns of the C–CTO-3 sample. (a) Cu (b)Ti. CTO,
Ca1ꢀxCuxTiO3.
where e is the dielectric constant of CTO–CCTO composites,
e0 is dielectric constant of CTO, ffiller is the volume fraction of
the conductive CCTO particles, fc is the percolation threshold,
and s is a critical exponent.
As shown in Fig. 5, a best fit of Eq. (4) to our experimental
data near the percolation threshold is performed. The results
indicate that the experimental values of the dielectric constant
are in good agreement with Eq. (4), with fcꢁ0.27 and sꢁ0.74.
Such a large increase in the dielectric constant is caused by the
existence of a large number of conducting CCTO particles in
parallel and in very close proximity, but blocked by thin barriers
of the dielectric CTO material. Normally, the interfacial polar-
ization will contribute on the high-dielectric behavior for a ma-
terial consisting of two phases (one is insulating and another is
conductive).22,23 Therefore, this high dielectric behavior in Cu-
doped CTO ceramics is also associated with interfacial polar-
ization. As the conductive CCTO volume fraction is low, the
CCTO particles are dispersed into the CTO ceramic matrix, and
thus the dielectric behavior of theses CCTO–CTO composites is
dominated by the CTO ceramic matrix. With increasing the
CCTO volume fraction, the CCTO particles form large particle
clusters and even continuous clusters, and thus percolation
effects will be effective. As the composition becomes almost
pure CCTO phase, the dielectric constant can reach 17550 at
1 kHz, which means that percolation effects will become the
boundary-layer capacitor effects as fc-1.
ꢀ
ꢁꢀs
fc ꢀ ffiller
e ¼ e0
(4)
fc
2500
2000
1500
1000
500
@ 1KHz
30
0
0
In conclusion, Cu-doped CaTiO3-based polycrystalline ce-
ramics exhibit high dielectric permittivity and low-loss tangent
behavior. Analysis of the ceramic microstructure and composi-
tion analyses indicate that the obtained ceramics are composed
of two phases, i.e., CTO and CCTO. The dielectric constant is
almost independent on temperature and frequency in a wide
5
10
15
20
25
Fig. 5. Dielectric constant of CTO-based ceramic samples as a function
of volume fraction of CCTO with logarithmic plot of Eq. (1). The inset
is the fitted line of lge vs lg(fcꢀfCCTO). CCTO, CaCu3Ti4O12
.

November 2009
Communications of the American Ceramic Society
2779
11C. C. Wang and L. W. Zhang, ‘‘Oxygen-Vacancy-Related Dielectric Anomaly
in CaCu3Ti4O12: Post-Sintering Annealing Studies,’’ Phys. Rev. B, 74, 024106 (2006).
12Z. Wang, X. M. Chen, L. Ni, Y. Y. Liu, and X. Q. Liu, ‘‘Dielectric Relax-
ations in Ba(Fe1/2Ta1/2)O3 Giant Dielectric Constant Ceramics,’’ Appl. Phys. Lett.,
90, 102905 (2007).
range. The observed high-dielectric behavior can be explained
by the percolation effects and interfacial polarization. The re-
sults indicate that such CTO-based ceramics will provide po-
tential application for microelectronic devices.
13D. C. Sinclair, T. B. Adams, F. D. Morrison, and A. R. West, ‘‘CaCu3Ti4O12
:
One-Step Internal Barrier Layer Capacitor,’’ Appl. Phys. Lett., 80, 2153–5 (2002).
14Y.-H. Lin, J. Cai, M. Li, C.-W. Nan, and J. He, ‘‘Grain Boundary Behavior in
Varistor-Capacitor TiO2-Rich CaCu3Ti4O12 Ceramics,’’ J. Appl. Lett., 103, 074111
(2008).
References
1S. M. Spearing, ‘‘Materials Issues in Microelectromechanical Systems
(MEMS),’’ Acta. Mater., 48, 179–96 (2000).
15M. Li, A. Feteira, D. C. Sinclair, and A. R. West, ‘‘Influence of Mn Doping on
the Semiconducting Properties of CaCu3Ti4O12 Ceramics,’’ Appl. Phys. Lett., 88,
232903 (2006).
2J. Carreaud, C. Bogicevic, B. Dkhil, and J. M. Kiat, ‘‘Dielectric Evidences of
Core-Shell-like Effects in Nanosized Relaxor PbMg1/3Nb2/3O3,’’ Appl. Phys. Lett.,
92, 242902 (2008).
16J. Li, M. A. Subramanian, H. D. Rosenfeld, C. Y. Jones, B. H. Toby, and A.
W. Sleight, ‘‘Clues to the Giant Dielectric Constant of CaCu3Ti4O12 in the Defect
Structure of SrCu3Ti4O12,’’ Chem. Mater., 25 [16] 5223–5 (2004).
17J. Li, A. W. Sleight, and M. A. Subramanian, ‘‘Evidence for Internal Resistive
Barriers in a Crystal of the Giant Dielectric Constant Material: CaCu3Ti4O12,’’
Solid State Commun., 135, 260–2 (2005).
3B. G. Kim, S. M. Cho, T. Y. Kim, and H. M. Jang, ‘‘Giant Dielectric Permit-
tivity Observed in Pb-Based Perovskite Ferroelectrics,’’ Phys. Rev. Lett., 86, 3404–
6 (2001).
4N. Setter and R. Waser, ‘‘Electroceramic Materials,’’ Acta. Mater., 48, 151–78
(2000).
18S.-Y. Chung, I.-D. Kim, and S.-J. L. Kang, ‘‘Strong Nonlinear Current–Volt-
age Behaviour in Perovskite-Derivative Calcium Copper Titanate,’’ Nat. Mater., 3,
774–8 (2004).
5A. P. Ramirez, M. A. Subramanian, M. Gardel, G. Blumberg, D. Li, T. Vogt,
and S. M. Shapiro, ‘‘Giant Dielectric Constant Response in a Copper-Titanate,’’
Solid State Commun., 115, 217–20 (2000).
19C.-W. Nan, ‘‘Physics of Inorganic Inhomogeneous Materials,’’ Prog. Mater.
Sci., 37, 1–116 (1993).
6C. C. Homes, T. Vogt, S. M. Shapiro, S. Wakimoto, and A. P. Ramirez, ‘‘Op-
tical Response of High-Dielectric-Constant Perovskite-Related Oxide,’’ Science,
293, 673–6 (2001).
20D. M. Grannan, J. C. Garland, and D. B. Tanner, ‘‘Critical Behavior of the
Dielectric Constant of a Random Composite near the Percolation Threshold,’’
Phys. Rev. lett., 46, 375–8 (1981).
7J. Wu, C.-W. Nan, Y.-H. Lin, and Y. Deng, ‘‘Giant Dielectric Permittivity
Observed in Li and Ti Doped NiO,’’ Phys. Rev. Lett., 89, 217601 (2002).
8Y.-H Lin, M. Li, C.-W Nan, J. F. Li, J. Wu, and J. L. He, ‘‘Grain and Grain
Boundary Effects in High-Permittivity Dielectric NiO-Based Ceramics,’’ Appl.
Phys. Lett., 89, 032907 (2006).
21C. Pecharroman, F. Esteban-Bategon, and J. S. Moya, ‘‘New Percolative
BaTiO3-Ni Composites with a High and Frequency-Independent Dielectric Con-
stant (er 5 80000),’’ Adv. Mater., 13, 1541–4 (2001).
22V. V. Novikov1 and K. W. Wojciechowski, ‘‘Frequency Dependences of Di-
electric Properties of Metal-Insulator Composites,’’ Phys. Solid State, 44, 2055–62
(2002).
9J. R. Macdonald, Impendance Spectroscopy, pp. 191–205. John Wiley & Sons
Inc., New York, 1987.
10J. Daniels, K. H. Hardtl, D. Hennings, and R. Wernicke, ‘‘Defect Chemistry
and Electrical Conductivity of Doped Barium Titanate Ceramics,’’ Philips Res.
Rep., 31, 487–543 (1976).
23C. Wang, Q. F. Fang, and Z. G. Zhu, ‘‘Enhanced Dielectric Properties of
Low-Temperature Sintered SrBi2Nb2O9/Ag Composites,’’ Appl. Phys. Lett., 80,
3578–80 (2002).
&