Effect of different pH values adjusted by ammonia on the dielectric properties of CaCu3Ti4O12 ceramics prepared by a sol-gel method

Article abstract of DOI:10.1016/j.jallcom.2018.11.244

To investigate the effect of different pH values (pH = 1, 3, 5, 7) of a solution used to prepare CaCu₃Ti₄O₁₂ (CCTO) ceramics on the microstructure and dielectric properties of the resultant ceramics, the acid regulator in the solutions was replaced with an alkaline regulator, ammonia water, and CCTO ceramics were prepared through a sol-gel method. The data show that the dielectric constants of the as-prepared samples maintained relatively high values in the frequency range from 1 to 10⁷ Hz. In addition, the dielectric loss decreased to 0.028 at 10³ Hz for the sample prepared at pH = 5. Moreover, the results reveal that the grain size, porosity and content of the CuO phase at the grain boundaries of the CCTO ceramics varied with the pH value. Our results indicate that using an alkaline regulator to adjust the pH value of the synthesis solution is an effective method for reducing the dielectric loss of CCTO ceramics.

Full text of DOI:10.1016/j.jallcom.2018.11.244

Accepted Manuscript
Effect of different pH values adjusted by ammonia on the dielectric properties of
CaCu Ti O ceramics prepared by a sol-gel method
4 12
3
Jianhua Zhang, Jiecheng Zheng, Yecheng Li, Yuqian Liu, Wentao Hao, Lingyan Lin,
Yuanyuan Li, Jiancheng Song
PII:
S0925-8388(18)34376-7
DOI:
https://doi.org/10.1016/j.jallcom.2018.11.244
JALCOM 48470
Reference:
To appear in: Journal of Alloys and Compounds
Received Date: 18 March 2018
Revised Date: 3 November 2018
Accepted Date: 19 November 2018
Please cite this article as: J. Zhang, J. Zheng, Y. Li, Y. Liu, W. Hao, L. Lin, Y. Li, J. Song, Effect of
different pH values adjusted by ammonia on the dielectric properties of CaCu Ti O ceramics
4 12
3
prepared by a sol-gel method, Journal of Alloys and Compounds (2018), doi: https://doi.org/10.1016/
j.jallcom.2018.11.244.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
Effect of different pH values adjusted by ammonia on the
dielectric properties of CaCu Ti O ceramics prepared by a
3
4
12
sol gel method
-
a,
a
a
a
b,c
JIANHUA ZHANG *, JIECHENG ZHENG , YECHENG LI , YUQIAN LIU , WENTAO HAO ,
a
a
a
LINGYAN LIN , YUANYUAN LI , JIANCHENG SONG
a
Shanxi Key Laboratory of Coal Mining Equipment and Safety Control, National & Provincial
Joint Engineering Laboratory of Mining Intelligent Electrical Apparatus Technology, College of
Electrical and Power Engineering, Taiyuan University of Technology, No.79 Yingze West
Avenue, Wanbolin District, Taiyuan, Shanxi 030024, PR China
b
Key Lab of Advanced Transducers and Intelligent Control System, Ministry of Education,
Taiyuan University of Technology, Taiyuan 030024, PR China
c
College of Physics and Optoelectronics, Taiyuan University of Technology, Jinzhong, Shanxi
030：00, PR China
Address correspondence to the authorꢀ
*
E-mail: zhangjianhua@tyut.edu.cn

ACCEPTED MANUSCRIPT
Abstract
To investigate the effect of different pH values (pH = 1, 3, 5, 7) of a solution used
to prepare CaCu Ti O (CCTO) ceramics on the microstructure and dielectric
3
4
12
properties of the resultant ceramics, the acid regulator in the solutions was replaced
with an alkaline regulator, ammonia water, and CCTO ceramics were prepared
through a sol-gel method. The data show that the dielectric constants of the
as-prepared samples maintained relatively high values in the frequency range from 1
7
3
to 10 Hz. In addition, the dielectric loss decreased to 0.028 at 10 Hz for the sample
prepared at pH = 5. Moreover, the results reveal that the grain size, porosity and
content of the CuO phase at the grain boundaries of the CCTO ceramics varied with
the pH value. Our results indicate that using an alkaline regulator to adjust the pH
value of the synthesis solution is an effective method for reducing the dielectric loss
of CCTO ceramics.
Keywords: CaCu Ti O (CCTO); Sol-gel processes; pH value; Alkaline regulator;
3
4
12
Dielectric response.

ACCEPTED MANUSCRIPT
1. Introduction
Since CaCu Ti O (CCTO) ceramics were discovered in 2000 [1], extensive research has been
3
4
12
4
5
devoted to them because of their high dielectric constant (ε ) of approximately 10 –10 and
r
phase-transition stability over a wide temperature range from 100 to 400 K [2]. CCTO ceramics
also exhibit non-ohmic characteristics, which makes these ceramics promising for applications such
as energy storage capacitors and over-voltage protection devices [3, 4]. However, the relatively high
dielectric loss (tanδ) of CCTO ceramics limits their application in devices. Although the origin of
the promising properties of CCTO ceramics are not yet well understood [5], both the giant dielectric
constant and non-ohmic characteristics of CCTO have been tentatively attributed to an internal
barrier layer capacitor (IBLC) effect originating from extrinsic polarization at the grain boundaries
[
6, 7]. In this model, bulk grains behave as a semiconductor and grain boundaries exhibit insulating
behavior with high resistance. Generally, the high resistance of the grain-boundary layer
corresponds to a low conductance, which means a low tanδ. Therefore, the tanδ can be reduced by
methods that increase the grain-boundary resistance [8].
Thus far, modification of the preparation process has been demonstrated to be one of the most
effective methods for controlling the structure of the interfacial regions at the grain boundaries. In
addition, the dielectric properties of CCTO ceramics have been found to be highly sensitive to the
structure of the grain boundaries [9]. Among the methods for fabricating CCTO ceramics, the
sol-gel method has been commonly used because of the good chemical uniformity of the product,
low-temperature synthesis conditions and ease of controlling the components. The preparation of
CCTO ceramics via the sol-gel method involves several key processes, including solution
preparation, gel formation, pre-sintering and sintering [10]. Most of the previously related research
has been focused on the sintering process, whereas little has considered the effect of the pH value of
the solution on the dielectric performance of the products. In previous works, when certain elements
such as Sr, Mg and Pr were added by doping, the pH value was usually adjusted to 3 and only the

ACCEPTED MANUSCRIPT
effects of the sintering temperature and/or the dopants were investigated [10-14]. In another work,
the pH value of the solution was varied from 1.0 to 1.8 with an acidity regulator, nitric acid, and the
sample exhibited the highest ε when the pH was 1.6 [15]. Notably, the pH interval was narrow, only
r
between 1 to 3, when acidity regulators were used, and the effect of larger pH values on the
microstructure and dielectric properties of CCTO ceramics warrants further exploration. Therefore,
in this paper, CCTO ceramics were prepared through the sol-gel method. An alkaline regulator,
ammonia water, was used to adjust the pH of the solution from 1 to 7, and the influences of
different pH values on the microstructure and dielectric properties were investigated.
2. Experimental
CCTO precursor powders were prepared by the sol-gel method using solutions with different
pH values. The raw materials included Cu(NO ) ·3H O (99 wt%), Ca(NO ) ·4H O (99 wt%) and
3
2
2
3 2
2
Ti(OC H ) (99 wt%). Citric acid was used as a complexing agent. First, appropriate amounts of
4
9 4
Ca(NO ) ·4H O and Cu(NO ) ·3H O were dissolved in ethanol to form solution A and Ti(OC H )
9 4
3
2
2
3 2
2
4
was immixed into ethanol to form solution B. Solution B was then added into a citric acid solution
immediately with vigorously stirring by a magnetic stirrer to obtain solution C. The two solutions A
and C were mixed in a beaker under stirring, and the pH was adjusted to 1, 3, 5 or 7 by addition of
ammonia water to obtain a sol. The sol was subsequently heated to 80 °C for 12 h to prepare the dry
gel. The precursor powders were obtained by drying the gel in a crucible at 700 °C in air; the
organic matter was fired for 2 h at 750 °C. The as-prepared powders were pressed into 30 mm
diameter pellets under 200 MPa pressure, and the CCTO ceramics were finally fabricated by
sintering the pellets at 1050 °C for 12 h.
The phase structures of the sintered specimens were identified by X-ray diffraction with Cu Kα
radiation (MSAL–XD2). The fractured cross-sectional microstructures were observed by scanning
electron microscopy (SEM, TESCAN VEGA SBU) and Energy-dispersive X-ray spectroscopy
(EDS) was conducted using a spectrometer attached to the scanning electron microscope to analyze

ACCEPTED MANUSCRIPT
the chemical composition. Electronic backscattered micrographs were performed through the field
emission scanning electron microscopy (FE-SEM, JSM-6700F). To measure the dielectric
properties, both sides of the samples were coated by silver paste and fired at 620 °C to form
electrodes. The dielectric dispersion spectra and complex impedance spectra were collected using
7
an impedance analyzer (NOVOCONTROL Concept 80) in the frequency range from 1 to 10 Hz at
different temperatures. The current-voltage (J-E) characteristics were recorded using a precision
high-resistance meter (Keithley 6517b).
3. Results and Discussion
3.1 Structural studies
Fig. 1 shows the XRD patterns for calcinated CCTO powders with different pH values. The
XRD analysis confirms pseudo-cubic CCTO phase is clearly the main phase in all the samples by
comparison with the standard powder diffraction file database (JCPDF File No. 75-1149), while
CuO phase was also present in all the sample powders. When the pH value of the sol was 1, an
impurity phase of TiO appeared in the product, and both the CaTiO and TiO phases were found in
2
3
2
the sample prepared using a sol with pH = 7, suggesting that an excessively low or excessively high
sol pH value adversely affects the formation of the pure CCTO phase powders.
Fig. 2 shows the XRD patterns of CCTO ceramics prepared from sols with different pH values,
after the ceramics were sintered at 1050 °C for 12 h. The cubic perovskite CCTO phase as the main
phase and the CuO phase as the secondary phase were detected in all the samples. A comparison
with Fig. 1 reveals that the mixed phases CaTiO and TiO disappeared after the sintering process,
3
2
whereas the peaks of the CuO phase become more intense. Through the full width at half maximum
of XRD pattern, the lattice parameters for the samples prepared from sols with pH = 1, 3, 5 and 7
were calculated to be 7.389, 7.390, 7.392 and 7.392 Å, respectively. These lattice parameters of the
samples are not substantially different (<0.002 Å) from the value of 7.391 Å reported in the
literature [16].

ACCEPTED MANUSCRIPT
3.2 Morphology
Fig. 3 presents SEM and backscattered micrographs of the ceramic samples prepared from sols
with different pH values, after the samples were sintered at 1050 °C for 12 h. All the CCTO
ceramics are composed of particles with sizes that range from several to more than 10 ꢀm. In the
samples prepared from sols with pH ≤ 3, no grains of large size are present; by contrast, grains with
both large and small sizes are observed in the samples prepared from sols with pH ≥ 5. The average
grain sizes estimated using the program Nano-measurer are listed in Table 1. With increasing pH of
the sols, the grain size first decreases to the minimum (1.14 ꢀm for the sample prepared from the sol
with pH = 3) and then increases again. The SEM images show that the sample prepared from a sol
with pH = 5 shows a uniform grain microstructure with a grain size of approximately 4.39 ꢀm. We
attributed the wide range of grain sizes for these samples to the different pH values of the sols,
which may lead to different contents of the CuO phase at the grain boundary during the formation
of the CCTO ceramics [9, 17-19].
Table 1
Average grain size estimated using the Nano-measurer program, and the density measured by
Archimedes' method for CCTO ceramics prepared by sols with different pH values.
pH value of sol
1
3
5
7
Grain size (ꢀm)
2.49
4.50
1.14
4.54
4.39
4.60
4.41
4.09
3
Density (g/cm )
Table 1 also lists the densities of the samples prepared from sols with different pH values
measured by Archimedes' method. The density increases to the maximum for the sample prepared
from the sol with pH = 5 and then decreases to the minimum for the sample prepared from the sol
with pH = 7. The variation of the density may be related to the different concentrations of CuO at
the grain boundary as a result of the variation of the pH of the solution. During the sintering process,

ACCEPTED MANUSCRIPT
the CuO phase melts first; the liquid phase subsequently promotes grain growth and influences the
density of the samples [7, 20-22].
To confirm the concentration of CuO at the grain boundary, EDS for all samples was
conducted at A, B, C and D point, respectively. As shown in Fig. 2, the CuO phase exists in all the
sintered ceramic samples. Combined with the EDS data shown in Table 2, the results in Fig. 3
confirm that the content of CuO at the grain boundary first decreases and then increases with
increasing pH of the sol, indicating that pH of the sol plays an important role in determining the
CuO concentration at the grain boundaries. It also can find from the electron backscattered
micrographs of all the samples (Fig 3 a'–d') that no obvious second phase was observed except for
the grain boundaries, indicating all the secondary CuO phase existed at the grain boundaries.
Table 2
EDS analysis results for the marked points in Fig. 3.
Element/at.%
A
B
C
D
O
Ca
Ti
64.86
4.92
61.89
8.79
20.69
8.63
65.24
2.32
64.258
3.282
21.67
10.79
19.47
10.75
21.47
10.97
Cu
3.3 Complex impedance spectroscopy
To further study the effect of the pH value on the electrical properties of the CCTO ceramics,
the complex impedance properties of the samples were analyzed. In accordance with the IBLC
model, a series circuit of two RC parallel circuits can be used to illustrate the electrical
heterogeneity of the grains and grain boundaries in polycrystalline ceramics. The complex
impedance (Z*) of the equivalent circuit can be calculated by [7, 23]
Rgb
Rg
*
Z = Z′ −iZ′′ =
+
(1)
1
+ i
ω
R C
1+ iωR C
g g
gb gb

ACCEPTED MANUSCRIPT
Rgb
Rg
R Cg
Z′ =
+
(2)
(3)
2
2
1
+
(
ω
R C
)
1+
(
ω
)
gb gb
g







ω
R C
ω
R Cg
gb gb
g


Z′′ = Rgb
+ Rgb
2
2






1
+
(
ω
R C
)
1+
(
ω
R Cg
)
gb gb
g

where R and C are the resistance and capacitance, respectively, of the grain boundaries; and R
g
gb
gb
and C are the resistance and capacitance of the grains, respectively. The complex impedance
g
spectra for the CCTO ceramics at room temperature in the 0.01 Hz to 10 MHz frequency range are
'
shown in Fig. 4. According to Eq.(1), R and R can be obtained from the intercepts on the Z axis.
g
gb
The figure suggests that the R of the sample first decreases and then increases with the increase of
gb
the pH value, and the R of the sample with pH = 3 is the minimum. A comparison of the results of
gb
the EDS and impedance analysis shows that the trend in the Cu content at the grain boundary is
basically in line with the trend in R with the increasing pH value. This indicates that the pH value
gb
has a great influence on Rgb.
In general, the R and R follow the Arrhenius laws [24]:
g
gb
R = R exp(E / K T)
(4)
(5)
g
0
g
B
R = R exp(E / K T)
gb
0
gb
B
where R is the pre-exponential constant term, the values of E and E represent the activation
0
g
gb
energies at grain and grain boundaries, respectively, k is the Boltzmann constant, and T is the
B
absolute temperature. The E and E could be calculated from the slopes of the linear fitting results.
g
gb
In order to determine the E and E for all the samples, complex impedances spectra were
g
gb
conducted at different temperatures (not shown here). Fig 5 shows the dependence of lnR and lnRg
gb
on the 1000/T. With the rising of pH values, it can be found that the E for different samples
gb
increased to a maximum of 0.66 eV for the CCTO ceramics prepared with sol of pH=5 at first, and
then decreased. While the E varies with pH value, the E remains 0.081-0.089 eV, with negligible
gb
g
differences, which is consistent with other reported literature [25]. It can be inferred that different

ACCEPTED MANUSCRIPT
pH values for CCTO ceramics adjusted by ammonia has obvious impact on properties of grain
boundaries rather than grains, and appropriate pH value facilitates the improvement of Egb for
CCTO samples. Usually, high activation is associated with low tan
with pH=5 with the lowest tan would be expected, as described below in Section 3.4.
.3 Current–voltage characteristics
The current–voltage characteristics are also affected by R [26]. As shown in Fig. 6, nonlinear
δ, and the CCTO ceramic sample
δ
3
gb
J–E characteristics are observed for all the CCTO samples at room temperature. Breakdown electric
2
field (E ) is defined when the leakage current density is 10 mA/cm . The nonlinear coefficient (
α) of
b
all the samples are calculated from these curves via the equation [27]
log
(
I2 / I1
)
)
α
=
(6)
log
(
U /U1
2
where U and U are the voltages corresponding currents I and I , respectively;
α
is calculated
1
2
1
2
when I = 0.1 mA and I = 1 mA. The results are summarized in Table 3. Generally, a large R , a
1
2
gb
large number of grain boundaries, lower dc conduction and low porosity improve the E [28, 29].
b
According to the results in Fig. 6, the sample prepared from a sol with pH = 3 exhibits the lowest Eb,
and the smallest R even though it has the largest number of grain boundaries. Meanwhile, the
gb
sample prepared from a sol with pH = 5 exhibits the maximum E , while it behaves a larger R and
b
gb
the highest density. Although the sample prepared from a sol with pH = 7 has the largest R , the
gb
higher porosity and the small number of grain boundaries lead to a reduction of the E . In this work,
b
the obtained
α values are consistent with those of the original CCTO ceramics reported in the
previous literatures[18, 30].
Generally, the formation of the grain-boundary Schottky potential barrier is attributed to the
difference of electronic structures between grain and grain boundary: the chemical heterogeneity
can lower the local Fermi energy of grain boundary and can trap the flowing electrons till the Fermi
energy in grain and grain boundary become equal. This separation and accumulation of charges can

ACCEPTED MANUSCRIPT
produce a depletion layer at grain boundary, and the inner electric field in the depletion layer forms
the grain-boundary Schottky potential barrier. Hence, increasing the barrier height (
the hopping or migration of charge carriers over the potential barrier in CCTO ceramics more
difficult, indicating a larger can decrease the , improve the E ,and effectively restrain the
leakage conductance, which may be a critical factor responsible for the low tan
31, 32]. In present work, the pH causes the variation of the amount of Cu at grain boundary,
Ф) would make
Ф
α
b
δ
of CCTO ceramics
[
which can affect the Fermi energy at grain boundary and introduce electronic traps, which might
induce the variation of Ф. According to the J–E characteristics measured at 298–358 K (not shown
here), the
Ф of CCTO ceramics were calculated by the equations [33]:
2
1/2
J = AT exp(
β
E −
Φ
)/ k T
(7)
(8)
B


2
InJ = −
Φ
/ k T + In( AT )
0
B
where A and T are the Richardson constant and absolute temperature, respectively,
pre-exponential factor and Boltzmann constant, and
boundary, J is the current density at zero electric field. As shown in the inset of Fig. 6, a good
β
and k are the
B
Φ
is the Schottky barrier height of the grain
0
1
/2
linear relationship between lnJ and E can be obtained at the low electric field region in the range
98–338 K, which indicates that the J–E characteristics of the CCTO ceramics can be well
elucidated based on the field-assisted thermionic emission mechanism [34]. The was obtained at
.38, 0.55, 0.57, and 0.31 eV for the CCTO samples prepared from sols with different pH values.
of different samples increased with the rise of pH values at lower pH, then sharply dropped when
pH=7. The sample prepared from the sol with pH = 5 exhibits the largest , which means that its dc
2
Ф
0
Ф
Ф
conduction is suppressed and E will be improved, which is consistent with the analysis of
b
aforementioned E data. The results indicate that the pH change caused by adding ammonia greatly
b
effects on grain boundary properties.
Table 3 The εr and tan
δ values at room temperature and 1 kHz, nonlinear coefficient (α), and DC
breakdown electric field (E ) of CCTO ceramics prepared from sols with different pH values; the
b

ACCEPTED MANUSCRIPT
samples were sintered at 1050 °C for 12 h.
3
pH value
εr × 10 (1 kHz)
tan
δ
(1 kHz)
α
Eb (V/cm)
1272.79
494.97
1
3
5
7
3.81
3.37
4.46
4.16
7
2
.92
.73
0.052
0.030
0.028
0.354
2
8.70
1725.34
636.40
1
26.73
3
.4. Dielectric properties
The εr and tan
Notably, the pH value of the sol strongly influences not only the morphologies of the CCTO
ceramics but also their dielectric properties. The εr and tan of all the samples at room temperature
δ as functions of the frequency are shown in Fig. 7 and Fig. 8, respectively.
δ
and 1 kHz are listed in Table 3. The εr of all the CCTO ceramic samples decrease slowly over the
frequency range from 1 to 400 kHz, and a sharp decreased is observed at frequencies greater than
4
00 kHz, except for the sample prepared from a sol with pH = 7. Moreover, the εr decreases first
and then increases with increasing sol pH value. The sample prepared from a sol with pH = 3
exhibits the minimum εr, which is approximately 2500 in most of the frequency range of the test.
The maximum εr is observed for the sample prepared from the sol with pH = 7. However, it exhibits
poor frequency stability. The difference in εr is attributed to the various grain size distributions.
According to the IBLC model, the εr can be obtained by the equation [23, 35]
dg + d
gb
εr
=
εgb
(9)
dgb
where εgb is the dielectric constant of the grain boundary; and d and d are the sizes of the grain
g
gb
and grain boundary, respectively. The εr is closely related to d , as d is much larger than dgb.
g
g
Therefore, with increasing sol pH value, the εr changes because of the variation of the grain size,
which results in the minimum εr of the sample prepared from the sol with pH = 3 which has the
smallest grain size. This observation is consistent with the SEM results presented in Fig. 3 and
Table 1.
In Fig.8, tanδ decreases first and then increases with increasing pH value. Among the samples,

ACCEPTED MANUSCRIPT
the sample prepared from a sol with pH = 5 presents the lowest tan
δ, approximately 0.028 at room
4
temperature and 1 kHz, and its εr is maintained at approximately 2 × 10 . Generally, a lower tan
δ
is
closely correlated with a large number of grain boundaries, high density, low dc conduction of the
grain barrier layer and high R [36-41]. Therefore, the sample prepared from a sol with pH = 5,
gb
which possesses a high R , the largest observed density and the maximum barrier height, has the
gb
lowest tanδ.
4. Conclusions
CCTO ceramics were prepared by the sol-gel method from solutions with 4 different pH values,
where the pH of the solution was adjusted with ammonia water as an alkaline regulator. With
increasing sol pH value, both the εr and the tan of the obtained CCTO first decrease and then
δ
increase, exhibiting similar trends. The sample prepared from a sol with pH = 5 exhibits the lowest
4
tan
δ
of 0.028, and its εr
,
α
and E were 2 × 10 , 4.46 and 1725.34 V/cm, respectively. This material
b
may thus be a promising varistor material for over-voltage protection in electronic systems. Our
results demonstrate that adjusting the pH value of the sol using an alkaline regulator is an effective
method for reducing the tanδ of CCTO ceramics.
Acknowledgment
This work was financially supported by the National Natural Science Foundation of China
(Grant No. 51701136, 51602214, 51577123 and U1210112) and the National Science Foundation of
Shanxi (Grant No. 201701D221240, 201601D21092 and 201601D202052).
References
[
1] M.A. Subramanian, D. Li, N. Duan, B.A. Reisner, A.W. Sleight, J Solid State Chem, 151 (2000)
23-325.
3
[
[
2] M.A. Subramanian, A.W. Sleight, Solid State Sci, 4 (2002) 347-351.
3] C. Kant, T. Rudolf, F. Mayr, S. Krohns, P. Lunkenheimer, S.G. Ebbinghaus, A. Loidl, Phys Rev
B, 77 (2008) 5131.

ACCEPTED MANUSCRIPT
[
[
[
[
[
[
[
[
[
[
4] D.E. Kotecki, Integr Ferroelectr, 16 (1997) 1-19.
5] L. Wu, Y. Zhu, S. Park, S. Shapiro, G. Shirane, J. Tafto, Phys Rev B, 71 (2005) 014118.
6] T.B. Adams, D.C. Sinclair, A.R. West, Phys Rev B, 73 (2006) 094124.
7] D.C. Sinclair, T.B. Adams, F.D. Morrison, A.R. West, Appl Phys Lett, 80 (2002) 2153-2155.
8] R. Espinoza-González, E. Mosquera, Ceram Int, 43 (2017) 14659-14665.
9] D. Wang, G.P. Bierwagen, Prog Org Coat, 64 (2009) 327-338.
10] L. Sun, Z. Wang, W. Hao, E. Cao, Y. Zhang, H. Peng, J Alloys Compd, 651 (2015) 283-289.
11] L. Sun, R. Zhang, Z. Wang, E. Cao, Y. Zhang, L. Ju, J Alloys Compd, 663 (2016) 345-350.
12] D.L. Sun, A.Y. Wu, S.T. Yin, J Am Ceram Soc, 91 (2010) 169-173.
13] S. Vangchangyia, E. Swatsitang, P. Thongbai, S. Pinitsoontorn, T. Yamwong, S. Maensiri, V.
Amornkitbamrung, P. Chindaprasirt, J Am Ceram Soc, 95 (2012) 1497-1500.
14] E. Swatsitang, T. Putjuso, J Mater Sci: Mater El, 28 (2017) 18966–18976.
15] Y. Li, P. Liang, X. Chao, Z. Yang, Ceram Int, 39 (2013) 7879-7889.
[
[
[
[
[
16] J. Li, A.W. Sleight, M.A. Subramanian, Solid State Commun, 135 (2005) 260-262.
17] U.G. Akpan, B.H. Hameed, J Hazard Mater, 170 (2009) 520-529.
18] P. Leret, J.F. Fernandez, J.D. Frutos, D. Fernández-Hevia, J Eur Ceram Soc, 27 (2007)
3
901-3905.
[
[
[
[
[
19] C. He, S.W. Kim, D.S. Lee, J Control Release, 127 (2008) 189-207.
20] A.A. Felix, M.O. Orlandi, J.A. Varela, Solid State Commun, 151 (2011) 1377-1381.
21] S.F. Shao, J.L. Zhang, P. Zheng, C.L. Wang, Solid State Commun, 142 (2007) 281-286.
22] R. Schmidt, D.C. Sinclair, Chem Mater, 22 (2010) 6-8.
23] R. Schmidt, M.C. Stennett, N.C. Hyatt, J. Pokorny, J. Prado-Gonjal, M. Li, D.C. Sinclair, J Eur
Ceram Soc, 32 (2012) 3313-3323.
[
[
24] X. Huang, H. Zhang, M. Wei, Y. Lai, J. Li, J Alloys Compd, 708 (2017) 1026-1032.
25] J.A. Cortés, G. Cotrim, S. Orrego, A.Z. Simões, M.A. Ramírez, J Alloys Compd, 735 (2018)

ACCEPTED MANUSCRIPT
1
40-149.
26] L. Liu, L. Fang, Y. Huang, Y. Li, D. Shi, S. Zheng, S. Wu, C. Hu, J Appl Phys, 110 (2011)
94101.
27] P. Kum-onsa, P. Thongbai, B. Putasaeng, T. Yamwong, S. Maensiri, J Eur Ceram Soc, 35 (2015)
441-1447.
[
[
0
1
[
[
[
[
[
[
[
28] Z. Tang, K. Wu, Y. Huang, J. Li, Energies, 10 (2017) 1031.
29] E.A. Stefanescu, X. Tan, Z. Lin, N. Bowler, M.R. Kessler, Polymer, 51 (2010) 5823-5832.
30] J. Li, K. Wu, R. Jia, L. Hou, L. Gao, S. Li, Mater Design, 92 (2016) 546-551.
31] C.C. Wang, L.W. Zhang, Appl Phys Lett, 90 (2007) 323.
32] C.C. Wang, M.N. Zhang, K.B. Xu, G.J. Wang, J Appl Phys, 112 (2012) 323.
33] X. Zhao, L. Ren, L. Yang, S. Li, R. Liao, W. Li, J. Li, IEEE T Dielect El In, 24 (2017) 764-773.
34] K. Tsuji, W.T. Chen, H. Guo, W.H. Lee, S. Guillemetfritsch, C.A. Randall, J Appl Phys, 121
(2017) 064107.
[
35] Briz, V. Eacute, G. Gruener, J. Wolfman, K. Fatyeyeva, M. Tabellout, M. Gervais, F. Gervais,
Mater Sci Eng B, 129 (2006) 135-138.
[
[
[
[
[
[
36] G. Deng, P. Muralt, MRS Proceedings, 1256 (2010) 515-526.
37] T.B. Adams, D.C. Sinclair, A.R. West, Phys Rev B, 73 (2006) 094124.
38] J.L. He, J. Hu, Y.H. Lin, Sci China Technol Sc, 54 (2011) 2506-2510.
39] R. Schmidt, S. Pandey, P. Fiorenza, D.C. Sinclair, RSC Adv., 3 (2013) 14580-14589.
40] S. Ke, H. Huang, H. Fan, Appl Phys Lett, 89 (2006) 182904.
41] M. Li, D.C. Sinclair, A.R. West, J Appl Phys, 109 (2011) 084106.

ACCEPTED MANUSCRIPT
Figure Captions
Fig.1. XRD patterns of the as-prepared CCTO powders prepared via the sol-gel method from sols
with different pH values.
Fig.2. XRD patterns of CCTO ceramic samples prepared from sols with different pH values.
Fig.3. SEM and electronic backscattered images of CCTO ceramics prepared from sols with
different pH values: (a), (a') pH = 1; (b), (b') pH = 3; (c), (c') pH = 5; (d), (d') pH = 7.
Fig.4. Impedance spectroscopy plots for all the CCTO ceramics at room temperature.
Fig.5. Arrhenius plots of the relaxation behavior of CCTO ceramics with different pH values: (a)
grain boundaries; (b) grains.
Fig.6. Current–voltage characteristics of CCTO ceramic samples. The inset shows the calculation of
the
Ф in the temperature range from 298 to 338 K for all the CCTO ceramics.
Fig.7. εr as a function of the frequency measured at room temperature for all CCTO ceramics.
Fig.8. Frequency dependence of tanδ measured at room temperature for all CCTO ceramics.

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
Highlights:
ꢀ
The pH values of CCTO sol was adjusted by alkaline regulator from 1 to 7.
4
ꢀ
ε for the sample prepared from the sol with pH = 5 maintained 2.87× 10 at 1K
r
Hz.
ꢀ
ꢀ
The tanδ for the sample prepared from the sol with pH=5 was 0.028 at 1 kHz.
The tanδ decreasing may be attributed to the increasing R and lower porosity.
gb