Vol.2, No.7, 688-693 (2010) Natural Science
http://dx.doi.org/10.4236/ns.2010.27085
Synthesis, structural characterization and formation
mechanism of giant-dielectric CaCu3Ti4O12 nanotubes
Nirupam Banerjee, Saluru Baba Krupanidhi*
Materials Research Center, Indian Institute of Science, Bangalore, India; [e-mail]; [e-mail];
*Corresponding Author:
Received 5 March 2010; revised 10 April 2010; accepted 18 April 2010.
ABSTRACT
A capillary-enforced template-based method
has been applied to fabricate calcium copper
titanate (CaCu3Ti4O12, CCTO) nanotubes (di-
ameter ~200 nm) by filling sol-gel CCTO pre-
cursor solution into the nanochannels of po-
rous anodic aluminum oxide (AAO) templates,
subsequent heating for phase formation and fi-
nally the removal of nano-channel templates by
applying basic solution. X-ray diffraction (XRD),
field emission scanning electron microscopy
(FE-SEM) and transmission electron micros-
copy (TEM) equipped with Energy-dispersive
X-ray spectroscopy (EDX) have been employed
to characterize the morphology, structure, and
composition of as-prepared nanotubes. XRD
and selected-area electron diffraction (SAED)
investigations demonstrated that postannealed
(750○C for 1 h) CCTO nanotubes were poly-
crystalline with a cubic pseudo-perovskite cry-
stal structure. The FE-SEM and TEM results
showed that CCTO nanotubes were of uniform
diameter (~200 nm) throughout their length.
High resolution TEM (HRTEM) analysis confirm-
ed that the obtained CCTO nanotubes are made
of randomly aligned nano-particles 5-10 nm in
size. EDX analysis demonstrated that stoichi-
ometric CaCu3Ti4O12 was formed. The possible
formation mechanism of CCTO nanotubes in the
AAO template is discussed.
Keywords: Nano-Tubes; Sol-Gel; Templated
Growth; High-K-Dielectrics
1. INTRODUCTION
Subsequent to the discovery of carbon nanotubes [1] by
Iijima et al. many research groups around the globe have
become involved in the development of one-dimensional
nano-structures such as nano-tubes, nano-rods, nano-
wires etc. of various functional materials due to their
important role in mesoscopic physics and nanoscale de-
vice fabrication, as they can be utilized as both nano-
scale device elements and interconnections while retain-
ing unique properties due to size confinement in the ra-
dial direction [2-6]. Recently efforts have been made to
synthesize and understand the growth of different nanos-
tructures of the functional materials such as complex
dielectric and ferroelectric oxides because of their prom-
ising applications for pyroelectric detectors, piezoelec-
tric actuators and transducers, nonvolatile memory, ul-
trasonic devices, piezoelectric ink-jet printers etc. [7-11].
Nanotubes of various materials have been synthesized
through various methods, among which sol-gel template
technology provides a versatile technique for synthesizing
highly ordered and controlled with the same structure and
properties at large-scale production of one dimensional
nanostructures, which is the basic building block of nan-
odevices. The nanochannel of template can be viewed as a
reactor, in which one-dimensional nanostructure of the
desired material can be synthesized in which the shape
and size of the nanostructures depend only on the dimen-
sion of the pores present in the template.
High dielectric constant materials are immensely use-
ful for miniaturization of electronic devices which is a
recent goal of our technological progress. In this sce-
nario ferroelectric materials are on demand due to their
high dielectric constant values. Hernandez et al. have
fabricated BaTiO3 (BT) and PbTiO3 (PT) nanotubes us-
ing template approach [12]. Using the same synthetic
approach Zhang et al. have fabricated highly ordered
Pb(Zr0.53Ti0.47)O3 (PZT) nanowires which were found to
exhibit a significant piezoresponse and proposed for
potential device applications [13]. Recently, Singh et al.
reported the fabrication of Ba0.6Sr0.4TiO3 (BST) nano-
tubes [14]. But one serious drawback in these well
known ferroelectric materials possessing higher dielec-
tric constant values is these are Pb or Ba containing ma-
terials which are severely toxic to human bodies and

N. Banerjee et al. / Natural Science 2 (2010) 688-693
689
689
therefore are non-environment friendly materials. One
more important point is that the dielectric constant (ε)
for most of these ferroelectric ceramics has strong tem-
perature dependence and shows rapid change near ferro-
electric transition temperature which makes them ineffi-
cient to act as a good ceramic capacitor material. In the
first of this millennium Subramanian et al. [15] have
discovered a new pseudo-perovskite CaCu3Ti4O12 (CC
TO) possessing a very high value of dielectric constant
(~10,000 for bulk and ~105 for single crystals) and mod-
erately lower dielectric loss (~0.1). The dielectric con-
stant remains reasonably invariant over a wide tempera-
ture range from 100 to 400 K below 1 MHz. Hence the
material is capable of potential industrial application for
miniaturizing electronic devices. To a surprise the mate-
rial is non-ferroic and shows a centrosymmetric structure
down to 35 K [16]. Both intrinsic and extrinsic mecha-
nisms were employed to explain this unusual phenome-
non. Most of the studies have taken the help of Internal
Barrier Layer Capacitor (IBLC) model consisting of
semiconducting grains and insulating grain boundaries to
explain this enormously large dielectric const [17-18].
whereas Lunkenheimer et al. believe it is arising from
the electrode polarization effects [19]. Although there
are few reports on the synthesis of nano-particles of this
giant dielectric material [20], fabrication of its nano-
structures is rare in literature. In this article we are re-
porting the synthesis of gaint dielectric CCTO nanotubes
of about 200 nm diameter by means of a sol-gel method
utilizing nano-porous anodic aluminum oxide (AAO)
templates, structural characterization of the nano-tubes
and high resolution electron microscopic (HRTEM)
studies on the nano-tube walls to reveal the crystalline
structures.
2. EXPERIMENTAL DETAILS
CaCu3Ti4O12 precursor-solution was prepared through
conventional sol-gel route whreby metallic-acetates na-
mely copper-acetate monohydrate (Cu(CH3COO)2.H2O
> 99%, Aldrich) and calcium acetate monohydrate
(Ca(CH3COO)2.H2O > 99%, Aldrich) were used as me-
tal-ion precursors. Appropriate amount of these metal-
lic-acetates, corresponding to 1:3 ratios of Ca:Cu metal
ions were dissolved in glacial acetic acid which was used
as solvent. Titanium isopropoxide (TIP) (Ti-{OCH-
(CH3)2}4 > 99.9%, Aldrich) was used as Ti-source and
was dropwise added in appropriate amount (i.e., amount
corresponding 1:3:4 stoichiometric ratios of Ca:Cu:Ti
metal ions) to glacial acetic acid with a gas-tight syringe
while stirring the solution continuously with the help of
a magnetic stirrer. It is a well known phenomenon that
transition metallic alkoxides are unstable due to the
presence of extremely reactive alkoxide groups and are
prone to hydrolysis. Hence acetylacetone (CH2(CH3CO)2,
Merck) was added to the solution for stabilization of TIP.
Next these two solutions were mixed together to form
the final precursor solution which was adjusted to a
concentration 0.1 M (in terms of CCTO) and was sub-
sequently refluxed for one hour at 160○C for homogene-
ity. The final precursor solution was transparent, dark
blue in color and was stable for long time without any
precipitation.
Nanopourous Anodic Aluminum Oxide (AAO) mem-
branes (pore diameter: ~200 nm, and 60 µm thickness)
were purchase from Whatman®, were used as the tem-
plates for synthesis of CCTO nanotubes. Before using,
templates were cleaned for 20 min in an ultrasonic bath
using solvents of different polarity such as double dis-
tilled water, acetone, hexane, toluene and were annealed
at 200oC for 2 h in vacuum to make free of any existent
impurities. Next the AAO templates were immersed in
the precursor solution for 2 h and were dried at 150○C
for 1 h. These templates containing the precursor were
heated in air at 750oC for 1 h with the help of a thermal
annealing furnace for the desired phase to form. CCTO
nanotubes were isolated on removal of the AAO tem-
plates by immersing them in 6 M NaOH solution at
room temperature for 24 h. To obtain higher amount of
nanotubes, the base solution was diluted in several steps
with distilled water and finally with ethanol. The iso-
lated nanotubes were collected by centrifugation. The
structural studies on the as prepared CCTO nanotubes
were carried out using X-ray powder diffraction (XRD)
using a Philips PW3710 diffractometer (Cu-K radiation,
30 kV and 20 mA, λ = 1.5406 Å).The morphology of the
CCTO nanotubes was studied by scanning electron mi-
croscope (SEM) (Quanta 200). To reduce the charging
effects on insulating AAO templates and CCTO nano-
tubes during SEM studies, a very thin layer of gold was
sputtered for conductivity enhancement. The X-ray stru-
ctural data were corroborated by the transmission elec-
tron microscopic (TEM, Tecnai F30) studies, operating
at an acceleration voltage of 200 kV.
3. RESULTS AND DISCUSSIONS
Since the size and uniformity of the product nanotubes
solely depend on the used nanoporous AAO template, its
features namely the uniformity of size of the pores in the
template were studied by SEM. Figure 1(a) and Figure
1(b) show typical scanning electron micrographs of
top-view and lateral-views of an empty AAO template
respectively. It can be noticed that the pores size (di-
ameter) in the AAO template were in the range of 190-
220 nm and pores are well-uniformly distributed over

N. Banerjee et al. / Natural Science 2 (2010) 688-693
690
the template matrix in an approximately hexagonal-
shaped “honey-comb” like structure. Figure 1(c) shows
SEM image of CCTO nano-tubes after almost 90% re-
moval of the AAO template with NaOH solution. Figure
1(d) features a side view of some of the prepared nano-
tubes. The micrographs clearly show that as synthesized
CCTO nanotubes were hollow, roughly parallel to each
other, and vertically oriented on the AAO template to
form an array. The average diameter of the nano-tubes
are in the range of 160-180 nm, little smaller than the
expected value of 200 nm. This contraction of nano-
tubes from the original dimension of the pores is proba-
bly due to the shrinkage experienced by CCTO nano-
tubes on densification and dehydration associated with
thermal treatment. The diameter is quite uniform for
most of the nano-tubes indicating the superiority of this
method for large scale homogeneous nanotube synthesis.
To examine crystallinity and phase-purity of as-pre-
pared nano-tubes obtained after annealing at 700oC,
X-ray diffraction studies were carried out. Figure 2
shows the X-ray diffraction pattern of template free
CCTO nano-tubes, which were collected and gathered as
centrifuge residue after removal of NaOH solution and
subsequent washing with different solvent mediums for
several times. The XRD pattern clearly indicates the
crystallinity of the nano-tubes. All the peaks in the dif-
fractogram can be well-indexed according to the pseudo-
perovskite phase of CCTO (JCPDS card No. 75-2188).
Absence of any impurity peaks proves the phase-purity.
The orientations and the intensity ratios of the CCTO
nanotubes matched well with the cubic pseudo-perovs-
Figure 1. SEM images. (a) top-view of AAO template; (b) lateral-view of AAO template; (c) CCTO nanotubes after > 90%
removal of AAO templates with the help of NAOH solution; (d) lateral view of some of the CCTO nano-tubes.

N. Banerjee et al. / Natural Science 2 (2010) 688-693
691
691
Figure 2. X-Ray Diffraction pattern of prepared CCTO nano-
tubes annealed at 700oC after removal of AAO templates.
kite phase of bulk CCTO.
Figure 3(a) shows TEM image of a CCTO nano-tube
after complete removal of AAO template by dissolution in
6 M NaOH solution. The difference in contrast between
the middle and the edges of the tube in Figure 3(a) clearly
indicates its hollow structure. Enhanced scattering at edges
of the tube makes it darker than the middle portion in the
TEM picture. The diameter of the nano- tube is almost
uniform throughout its length which is approximately
equal (slightly smaller) to channel diameter of the applied
template. The length of these nano- tubes was found to be
shorter than the thickness (60 µm) of the AAO templates.
The possible reason may be the inefficient filling upto the
pore-depth in the template and also breakage of some tubes
to shorter ones during intense ultra-sonic dispersion treat-
ment. Figure 3(b) re- presents corresponding SAED pat-
tern which clearly proves the polycrystalline nature of the
nano-tubes. The rings in the SEAD pattern corresponds to
(422), (400), (013), (220) and (200) crystal planes of cubic
pseudo- perovskite phase of CCTO. The d-spacings were
calculated from the diameter of the bright circular rings
observed in the electron diffraction pattern which were
found to be in good agreement with that calculated from
the XRD data (JCPDS card No. 75-2188) and the data
available in the literature.
The micro-structure of the CCTO nano-tubes were
further investigated with the help of High Resolution
Transmission Electron Microscope (HRTEM) studies to
get a clear insight of the features. Figure 3(c) shows a
typical HRTEM image of the wall of a CCTO nano-tube
prepared with the help of AAO templates. The HRTEM
image clearly shows that the nano-tube is made up with
a number of CCTO well-resolved nano-particles that are
randomly oriented in the wall. The nano-particles com-
prising the walls of nano-tubes of other oxide materials
have been previously observed and reported in the lit-
erature [14,21]. The size of the constituent particles is in
the range of 5-10 nm which is in good agreement with
the previously reported data of these functionally com-
plex oxide nano-structures [14]. The observation of few
unresolved spaces in the HRTEM image suggests the
presence of few pores in the wall of CCTO nano-tubes.
The lattice spacing values (d220, d400) as obtained from
the HRTEM image closely resembles to that calculated
on the basis of XRD and SEAD results.
Energy dispersive X-ray (EDX) spectrum of the CCTO
nano-tubes after complete removal of AAO template is
shown in the Figure 3(d). No emission peaks from Al,
and other elements from the etching NaOH solution, re-
agent, and template appeared in this spectrum. The ab-
sence of any impurity peaks in the EDS spectrum and the
close resemblance of the atomic ratios of Ca, Cu and Ti to
the nominal composition in CaCu3Ti 4O12 prove the ele-
mental and phase purity of prepared CCTO nanotubes.
CCTO nanotubes were grown by sol-filling in the
AAO template with the help of capillary force. The sol
was drawn up into and filled the pores of the AAO tem-
plate. Air in the pores and vapor from the sol were
evaporated from the top surface of the AAO template.
The possible formation mechanism of CCTO nanotubes
inside the pores of AAO template is proposed here.
Schematics of the possible formation mechanism of
CCTO nanotube inside the pores of AAO template is
shown in Figure 4 (growth process of CCTO nanotube).
Initially, the template was put on the CCTO precursor
solution as shown in Figure 4 because the solution is
easier to fill into the pores of the template due to the
capillary force. As the solvent evaporates from the sur-
face and the concentration is enriched at the top of the
pores, precipitation or gelation occurs first at the top of
the pores on the AAO template surface exposed to the air
and subsequently proceeds downward along the entire
wall of the pores. The precursor solution consists of sol
nanoparticles or nanoclusters homogeneously dispersed
in the solvent. Such nanoclusters, which develop surface
charges and form a double layer, can respond to an ex-
ternally applied electric field or other nanoclusters [22].
When the surface of the AAO template is brought into
contact with a sol, a surface charge and a double layer
are formed on the surface of the AAO template. There-
fore, when sols are drawn into the pore channels of the
AAO template, if the surface charges of nanoclusters in
a sol and the surface of the AAO template are opposite,
there will be an electrostatic attraction force, and the

N. Banerjee et al. / Natural Science 2 (2010) 688-693
692
Figure 3. TEM image of (a) a CCTO nano-tube after complete removal of AAO template; (b) HRTEM image of a single
nano-tube surface revealing that it is comprised of a number of randomly oriented particles; (c) corresponding SEAD pat-
tern and (d) EDS spectrum of CCTO Nano-tubes.
Figure 4. A schematic diagram of the possible growth mechanism of CCTO nanotubes in the AAO template.

N. Banerjee et al. / Natural Science 2 (2010) 688-693
693
693
nanoclusters will preferentially deposit on the surface of
the pore channel, resulting in the formation of nano-
tubes.
4. CONCLUSIONS
In summary, CCTO nanotubes have been prepared suc-
cessfully by sol-gel method using the closely packed
porous nanochannel alumina templates. The crystallinity
and phase purity of the CCTO nanotubes were con-
firmed via XRD and SAED analysis. EDX analysis cor-
roborated the stoichiometric CaCu3Ti4O12 formation.
The walls of the nanotubes were found to be made up of
nanoparticles as confirmed by the HRTEM studies. The
nanoparticles embedded in the wall were found in the
range of 5-10 nm. This facile method of preparing the
CaCu3Ti4O12 nanotubes at a large scale might be impor-
tant for many applications in nanodevices.
5. ACKNOWLEDGEMENTS
N. B. thanks to Int. PH.D program, Indian Institute of Science for
financial assistances and Institute Nano-Science Initiative for micro-
scopic facilities.
REFERENCES
[1] Iijima, S. (1991) Helical microtubules of graphitic carbon.
Nature, 354(6348), 56-58.
[2] Hu, J., Odom, T.W. and Lieber, C.M. (1999) Chemistry
and physics in one dimension: Synthesis and properties of
nanowires and nanotubes. Accounts of Chemical Research,
32(5), 435-445.
[3] Atzke, G.R.P., Rumeich, F.K. and Nesper, R. (2002)
Based on oxide nanotubes and nanorods − anisotropic
building blocks for future nanotechnology. Angewandte
Chemie, 114(14), 2554-2571.
[4] Goldberger, J., He, R., Zhang, Y., Lee, S., Yan, H., Choi,
H. and Yang, P. (2003) Single-crystal gallium nitride
nanotubes, Nature, 422(6932), 599-602.
[5] Lee, S.B., Mitchell, D.T., Rofin, L.T., Ne vanen, T.K.,
Sçderlund, H. and Martin, C.R. (2002) Antibody-based
bio-nanotube membranes for enantiomeric drug separa-
tions. Scie nc e , 296(5576), 2198-2200.
[6] Sha, J., Niu, J., Ma, X., Xu, J., Zhang, X., Yang, Q. and
Yang, D. (2002) Silicon nanotubes. Advanced Materials,
14(17), 1219-1224.
[7] Junquera, J. and Ghosez, P. (2003) Critical thickness for
ferroelectricity in perovskite ultrathin films. Nature,
422(6931), 534-539.
[8] Wang, Y. and Santiago-Aviles, J.J. (2004) Synthesis of
lead zirconate titanate nanofibres and the Fourier-tran-
sform infrared characterization of their metallo-organic
decomposition process. Nanotechnology, 15, 32.
[9] Luo, Y., Szafraniak, I., Zakharo, N.D., Nagarajan, V.,
Steinhart, M., Ehrspohn, R.B.W., Endorff, J.H.W.,
Ramesh, R. and Alexe, M. (2003) Nanoshell tubes of
ferroelectric lead zirconate titanate and barium titanate.
Applied Physics Letters, 83(5377), 440-442.
[10] Chu, M.W., Szafraniak, I., Scholz, R., Harnagea, C.,
Hesse, D., Alexe, M. and Gosele, U. (2004) Impact of
misfit dislocations on the polarization instability of
epitaxial nanostructured ferroelectric perovskites. Na-
ture Materials, 3(2), 87-90.
[11] Roelofs, A., Schneller, I., Szot, K. and Waser, R. (2002)
Piezoresponse force microscopy of lead titanate nano-
grains possibly reaching the limit of ferroelectricity. Ap-
plied Physics Letters, 81(27), 5231-5233.
[12] Hernandez, B.A., Chang, K.S., Fisher, E.R. and Dorhout,
P.K. (2002) Sol-Gel template synthesis and characteriza-
tion of batio3 and pbtio3 nanotubes. Chemistry of Mate-
rials, 4, 480.
[13] Zhang, X.Y., Zhao, X., Lai, C.W., Wang, J., Tang, X.G.
and Dai, J.Y. (2004) Synthesis and piezoresponse of
highly ordered Pb(Zr0.53Ti0.47)O3 nanowire arrays. Ap-
plied Physics Letters, 85(18), 4190-4192.
[14] Singh, S. and Krupanidhi, S.B. (2007) Synthesis and
structural characterization of Ba0.6Sr0.4TiO3 nanotubes.
Physics Letter A, 367(4-5), 356-359.
[15] Subramanian, M.A., Li, D., Duan, N., Reisner, B.A. and
Sleight, A.W. (2000) High dielectric constant in ACu3-
Ti 4O12 and ACu3Ti 3FeO12 phase. Journal of Solid State
Chemistry, 151(2), 323-325.
[16] Liu, J. J., Sui, Y.C., Duan, C.-G., Mei, W.-N., Smith,
R.W., and Hardy, J.R. (2006) CaCu3Ti 4O12: Low-tem-
perature synthesis by pyrolysis of an organic solution.
Chemistry Materials, 18(16), 3878-3882.
[17] Sinclair, D.C., Adams, T.B., Morrison, F.D. and West,
A.R. (2002) CaCu3Ti 4O12: One-step internal barrier layer
capacitor. Applied Physics Letters, 80(12), 2153-2155.
[18] Liu, J., Duan, C., Yin, W., Mei, W.N., Smith, R.W. and
Hardy, J.R. (2004) Large dielectric constant and max-
well-wagner relaxation in Bi2⁄3Cu3Ti4O12. Physics Review
B, 70, 144106.
[19] Lunkenheimer, P., Bobnar, V., Pronin, A.V., Ritus, A.I.,
Volkov, A.A. and Loidl, A. (2002) Origin of apparent
colossal dielectric constants. Physics Review B, 66,
052105.
[20] Thomas, P., Dwarakanath, K., Varma, K.B.R. and Kutty,
T.R.N. (2008) Nanoparticles of the giant dielectric mate-
rial, CaCu3Ti4O12 from a precursor route. Journal of
Physics and Chemistry of Solids, 69(10), 2594-2604.
[21] Zhang, X.Y., Lai, C.W., Zhao, X., Wang, D.Y. and Dai,
J.Y. (2005) Synthesis and ferroelectric properties of mul-
tiferroic BiFeO3 nanotube arrays. Applied Physics Letters,
87, 143102.
[22] Wang, Y. and Cao, G. (2007) Synthesis and electro-
chemical properties of InVO4 nanotube arrays. Journal
of Materials Chemistry, 17(2298), 894-899.