Ceramics International xxx (xxxx) xxx–xxx

Contents lists available at ScienceDirect

Ceramics International
journal homepage: www.elsevier.com/locate/ceramint

Electronic structure, orbital symmetry transformation, charge transfer, and

valence state studies on Fe3+ - substituted CaCu3Ti4O12 quadruple
perovskites using X-ray photoelectron spectroscopy
Urmila M. Meshiyaa, Pooja Y. Ravala, Pooja R. Pansaraa, Monika Nehra2, Narendra Jakharc,
Sandeep Kumar2, Kunal B. Modia,∗, Dong-Kwon Lim4, Rishi Kumar Singhalc
a
Department of Physics, Saurashtra University, Rajkot, 360005, India
2
Department of Bio and Nanotechnology, Guru Jambheshwar University of Science and Technology, Hisar, 125001, India
c
Department of Physics, University of Rajasthan, Jaipur, 302004, India
4
KU-KIST Graduate School of Converging Science and Technology, Korea University, 145, Anam-ro, Seongbuk-gu, Seoul, Republic of Korea

A R T I C LE I N FO A B S T R A C T

Keywords: Here, we report a detailed study on electronic structure of quadruple perovskite system, CaCu3-xTi4-xFe2xO12
Calcium-copper-titanate (x = 0.0, 0.1, 0.3, 0.5 and 0.7), using the powerful X-ray photoelectron spectroscopy (XPS). The Ca2p and Ti2p
X-ray photoelectron spectroscopy XPS spectra indicate that Ca and Ti ions exist in the divalent and tetravalent state, respectively, and no visible
Electronic structure change is observed in their valence states on Fe3+-substitution (x). The asymmetrically shaped peaks of Ti2p3/2
Orbital symmetry transformation
infer the significant orbital hybridization of Ti atoms with other metallic cations like Cu and Fe, i.e. the Ti2p-
Cu3d-Fe3d mixed states, which shows an increment with the Fe3+-content. The spectra also indicate that the Fe-
substitution causes an increase in the strength of CuFe3d-O2p hybridization while weakening the strength of
TiFe3d-O2p orbital hybridization. The Cu2p spectra suggest that Cu ions exist in two valence states i.e. the Cu2+
and Cu3+ out of which Cu2+ is dominant. The XPS results show clear evidence of charge transfer amongst Ti–Cu
ions and evolution of n-type charge carriers. The O1s XPS spectra indicate anti-bonding states of t2g and eg type
molecular orbitals separated by an energy separation of ~1.6 eV. A systematic relative change in the intensity of
t2g and eg peaks with Fe-substitution point towards a clear transformation of symmetry of the system from TiO6
octahedral to the square-planar.

1. Introduction The X-ray photoelectron spectroscopy (XPS), termed as electron

spectroscopy for chemical analysis (ESCA), is a surface-sensitive pow-
The quadruple perovskites having a general formula, AA′3B4O12, are erful spectroscopic technique employed to explore the chemical com-
well investigated since the year 2000. Due to their fascinating di- position. Through the chemical shift, the electronic state of each ele-
electric, electric, magnetic, photocatalytic properties, their use has in- ment present on the surface of the complex material is studied [5,6].
creased in several applications such as supercapacitors, varistors, The XPS provides very definitive details on the electronic charge
memory cells, sensors, etc. [1,2]. Calcium-copper-titanate, Ca- transfer, symmetry transformation, ionicity and defects (cationic/an-
Cu3Ti4O12, belongs to this class of material, in which the Cu2+ ions are ionic vacancies) in the multicomponent system [7,8]. In the manu-
coordinated by four oxygen atoms forming a square-planar (A′-site) facturing industries, XPS is used to understand surface-related problems
array bonded to Ca2+ ions coordinated by 12 oxygen atoms. As a result, such as adhesion, surface defects or strains, and to verify the compo-
an alternating Cu/Ca network is extended in three distinct directions sition of modified surface and thin-film structures. For some high-per-
[100], [010], and [011] [3]. Furthermore, these Cu and Ca atomic formance materials, the composition of the surface is so critical to their
layers are interspersed by somewhat tilted [TiO6] octahedral con- performance that XPS is used to monitor a manufacturing process and
fronting each other. The crystal structures of pure and Fe-substituted ensure the product quality. Both the X-ray absorption near edge
CaCu3Ti4O12, generated using the XCrysDen software [4] are shown in structure (XANES) and XPS are element specific techniques. However,
Fig. 1 for the comparison purpose. while the XANES provides information on the unoccupied electronic

∗
Corresponding author.
E-mail address: kunalbmodi2003@yahoo.com (K.B. Modi).

https://doi.org/10.1016/j.ceramint.2019.09.198
Received 12 July 2019; Received in revised form 13 September 2019; Accepted 20 September 2019
0272-8842/ © 2019 Elsevier Ltd and Techna Group S.r.l. All rights reserved.

Please cite this article as: Urmila M. Meshiya, et al., Ceramics International, https://doi.org/10.1016/j.ceramint.2019.09.198
U.M. Meshiya, et al. Ceramics International xxx (xxxx) xxx–xxx

Fig. 1. Crystal structures of pure and Fe-substituted CaCu3Ti4O12.

states, the XPS provides the information on the density of occupied They possess coarse-grained microstructure, and grain size varies from
states of a system. Therefore, for a detailed interpretation of the results 3.4 μm to 7.5 μm for different compositions. The substituted Fe3+ ions
from one technique, the inputs from the other are very crucial to put the have a strong preference for the A′-site as verified by Rietveld-refine-
findings on a firm footing. So both the techniques are complementary to ment of X-ray diffraction data and 57Fe Mossbauer spectral analysis at
each other. 300 K and 5 K. The study of electronic structure carried out by em-
According to Zhang et al. [9], XPS analysis of Ti2p peak of Ca- ploying the powerful XANES spectroscopy demonstrates the reversal of
Cu3Ti4O12 revealed the simultaneous presence of the Ti3+ and Ti4+ orbital symmetry from the tilted octahedra to square-planar symmetry
ions which are attributed to microscopic origination of polaronic ac- in CaCu3Ti4O12 when Fe3+- ions are substituted for Cu2+and Ti4+ ions.
tions. Further, the presence of Ti3+ is balanced by oxygen vacancies. In The Cu, Fe and Ti edge XANES results confirmed that Cu, Fe, and Ti
the case of Ni2+-substitution on Cu-site and Ti-site of CaCu3Ti4O12, XPS atoms are in stable divalent (+2), trivalent (+3) and tetravalent (+4)
studies confirm the presence of Cu2+ and Ti4+ in both the samples and state respectively for all the compositions.
the presence of Ni2+ on either site is reflected from binding energy and The present work investigates the consequences of Fe3+ substitution
peaks shape [10]. Wang et al. [11] have detected Ti-rich and Cu-poor for Cu2+ and Ti4+ on (i) electronic structure of CaCu3Ti4O12 (ii) change
grain boundary layers in CaCu3Ti4O12 by XPS analysis. Yu et al. [12] in the valance state of the various metallic cations if any (iii) correlation
have shown that the valence of the Cu and Ti elements is influenced by study between previously studied structural, magnetic properties and
oxygen partial pressure in the course of sintering of CaCu3Ti4O12 outcome of X-ray absorption spectroscopy (XAS) on account of present
ceramics, merging with the results of XPS analysis. A polaronic stacking findings of XPS investigation with specific emphasis on orbital sym-
fault defect model for CaCu3Ti4O12 material was suggested by Bueno metry transformation and (iv) charge transfer mechanism. The core-
et al. [3] to elucidate the origination of the huge dielectric constant and level XPS measurements were performed at Ca2p, Cu2p, Ti2p, Fe2p, and
semiconducting coexistence features in agreement with XPS results. O1s states. Such systematic investigation has not been carried out thus
Very recently, XPS was employed to explore the origin of semi- far over the wide range of Fe3+-substitution on simultaneous replace-
conducting grains in Al3+ and Nb5+ co-doped CaCu3Ti4O12 [13]. The ment of magnetic Jahn-Teller Cu2+ ions and non-magnetic Ti4+ ions in
coexistence of Cu1+/Cu2+ and Ti3+/Ti4+ ions was confirmed by XPS CaCu3Ti4O12.
analysis [14] and these elements are referred as the considerable con-
tributors to the electrical conductivity of Ca1-xLaxCu3Ti4O12 2. Experimental details
(0 ≤ x ≤ 0.3) ceramics. In CaCu2.95Co0.05Ti4O12 thin film, monovalent
cation (Cu1+) detected by XPS is found to be responsible for decreased Complete particulars on preparation of five polycrystalline samples
leakage current [15]. Zhu et al. [16] have found that the Cu1+ vacancy of quadruple perovskite series, CaCu3-xTi4-xFe2xO12 with x = 0.0,0.1,
on the surface of CaCu3Ti4O12 is essential to activate the sulfate radicals 0.3, 0.5 and 0.7, and experimental procedures for the characterization
by forming a Cu1+→Cu2+ redox couple that is responsible for the ex- techniques are given in [19–23]. The elemental composition of the
peditious and effective dismissal of ibuprofen. In the system, Ca1- sample was examined by X-ray photoelectron spectroscopy (XPS; PHI X-
xLaxCu3Ti4O12 (x = 0.0, 0.03, 0.06 and 0.09), the binding energy of Cu tool, ULVAC Technologies, Inc., Japan) employing micro-focused
and Ti increases slightly with La3+-substitution which indicates an in- scanning monochromatic Al Kα X-ray source with a 20 μm minimum
crease in the oxidation degree, and enhancement in chemical bonds of beam diameter and an operating voltage of 1486.6 eV. The spectro-
O with cations, i.e., Cu and Ti [17]. The presence of Cu1+ and Ti3+ ions meter has energy resolution: FWHM Ag3d5/2 < 0.5 eV. The X-tools
in the Cu2.95Mg0.05Ti2Ta2O12 ceramic was confirmed by XPS analysis by sample holder is 75 × 75 mm and can support the measurement of
Zeng et al. [18]. Above studies thus strongly suggest versatile applica- multiple analysis points on a single large sample or multiple smaller
tions of XPS based investigation in multi-component systems. samples. The entire surface area of the sample holder is accessible for
So far, we have investigated diverse physical properties of per- making XPS measurements.
ovskite system, CaCu3-xTi4-xFe2xO12 where x = 0.0, 0.1, 0.3, 0.5 and
0.7, by means of various experimental techniques [19–23]. Summar-
3. Results and discussion
ising the outcome to facilitate the discussion on XPS data analysis: all
the samples possess expected chemical stoichiometry and they are
The full scan XPS survey spectra were recorded for all the five
single-phase materials which belong to body-centered cubic perovskite
compositions of the series, CaCu3-xTi4-xFe2xO12 (for x = 0.0, 0.1, 0.3,
crystal structure with space group Im3(No. 204) and point group Th.
0.5 and 0.7). The full scan XPS profiles for two of the samples; the one

2
U.M. Meshiya, et al. Ceramics International xxx (xxxx) xxx–xxx

Fig. 3. The Ca2p XPS spectra for the CaCu3-xTi4-xFe2xO12 system with x = 0.0,
0.1, 0.3, 0.5 and 0.7 compositions.

Table 1
Ca2p XPS fitting parameters for the CaCu3-xTi4-xFe2xO12 quadruple perovskite
system.
Fe3+ - content (x) Peak centers (eV) Peak difference (eV) Area ratio

Ca2p3/2 Ca2p1/2

0.0 346.16 349.63 3.47 1.32

0.1 346.04 349.51 3.47 1.14
0.3 346.07 349.49 3.42 1.23
0.5 346.07 349.53 3.46 1.16
0.7 346.03 349.45 3.42 1.24

Fig. 2. Full scan spectra for the typical compositions with x = 0.0 and 0.5 of the
system, CaCu3-xTi4-xFe2xO12.

without Fe (x = 0.0) and the other with Fe (x = 0.5), are displayed in

Fig. 2 that show the presence of various atoms and their binding energy
positions. The high-resolution element-specific fine scans are recorded
for Ca2p, Cu2p, Fe2p, and O1s that are discussed in detail hereafter one
by one in the preceding discussion. All the XPS spectra were standar-
dized using the C1s peak situated at 284.6 eV.
Fig. 3 shows the Ca2p XPS spectra recorded for x = 0.0, 0.1, 0.3, 0.5
and 0.7 compositions. Two prominent peaks have been observed at
energy positions, ~346.03 eV and ~349.5eV. These peaks correspond
to the spin-orbit doublet Ca states i.e. Ca2p3/2 and Ca2p1/2, respectively,
which can be designated to Ca2+ state by virtue of the accustomed line
positions chart. Here, we emphasize that these energy values might
differ marginally from one machine to other. Accordingly, for correct-
ness in referencing, one must employ the energy values measured with
the selfsame set-up with similar calibration conditions. The separation Fig. 4. The Ti2p XPS spectra for all the samples of CaCu3-xTi4-xFe2xO12 system
among these lines (~3.47 eV) is also compatible with the reported with x = 0.0, 0.1, 0.3, 0.5 and 0.7.
customary binding energy value for CaCO3 (~3.45 eV) [24] which re-
confirms that Ca ions exist in the divalent state in all these samples. Fig. 4 displays the Ti2p XPS spectra for all the studied compositions.
However, the contributions from surface oxides, hydroxides or hy- The two prominent peaks observed at ~457.81 eV and ~463.48eV,
drogen carbonates cannot be ruled out as that can be seen in the form of corresponding to the spin-orbit doublet Ti states, Ti2p3/2 and Ti2p1/2,
broadening of the peaks. Further, there is no measurable change ob- respectively. These peaks are quite accordant with tetravalent Ti ions in
served in the Ca2p XPS spectra upon Fe substitution. Table 1 sum- TiO2 lattice, i.e. the Ti4+ [25,26]. Ti2p3/2 peaks have asymmetric
marises the XPS data in terms of Ca2p3/2 and 2p1/2 peak centers, the structures therefore, the spectra were deconvoluted and two peaks
peak difference and the area ratio. Thus, through the analysis of Ca2p fitted to them. The fitted spectra are displayed in Figs. 5 and 6. The
XPS spectra, one can conclude that Ca ions stay in the divalent state and fitting parameters for the peak fitting of Ti 2p3/2 peak for all the sam-
there is no visible change in it by Fe substitution in the system. ples have been shown in Table 2. The principal peak at ~457.5 eV

3
U.M. Meshiya, et al. Ceramics International xxx (xxxx) xxx–xxx

Table 2
Ti2p3/2 XPS peak fitting parameters for the CaCu3-xTi4-xFe2xO12 system.
Fe3+ - content (x) Peak positions Area P1 Area P2 Area ratio (P2/P1)

P1 P2

0.0 457.52 458.84 32069 7907 0.24

0.1 457.49 458.49 27158 17928 0.66
0.3 457.44 458.39 24240 22266 0.91
0.5 457.39 458.31 24192 20110 0.83
0.7 457.33 458.31 22911 19973 0.87

of Fang et al. [27] and Jumpatam et al. [28], the presence of Ti3+
couldn't be modeled using Gaussian-Lorentzian profile fitting after
several trials. Therefore, one can infer that the presence of Ti3+ ions
can be ruled out in all these samples. The nonexistence of Ti3+ is also
Fig. 5. Ti2p3/2 XPS peak fitted by two peaks for the pristine composition,
consistent with the earlier XANES measurement [20,22]. These spectra
CaCu3Ti4O12
(x = 0.0). also reveal that there is substantial evidence of charge transfer and
orbital hybridization which increases with Fe-substitution.
The O1s XPS spectra for all the five compositions are manifested in
correlates to the tetravalent Ti i.e. the Ti4+ state and the other peak Fig. 7. The O1s spectra for the samples with different Fe-concentration
fitted at ~458.5 eV in the Ti2p3/2 can be ascribed to orbital hy- (x) show a broad peaked structure. All the spectra were therefore fitted
bridization of Ti2p states with 3d orbitals of other metallic cations, Cu with two Gaussian peaks, P1 and P2, at energy positions 529.06 eV, and
and Fe i.e. Ti2p-Cu3d-Fe3d. The ratio of the higher energy peak at 530.79 eV, respectively (Table 3). Two of the fitted spectra are shown in
~458.5 eV, with the main peak at ~457.5 eV, shows a systematic in- Fig. 8 for x = 0.0, and x = 0.5 compositions. These features P1 and P2
crement with the introduction of Fe ions, which points to the increased associated with the anti-bonding t2g-type and eg-type molecular orbi-
amount of hybridization with an increase in Fe3+ content (x) in the tals, corresponding to the Cu–Fe 3d and Ti–Fe 3d bands, constructed
system. One more interesting point that can be noted from Table 2 is the mainly from the O 2p atomic orbitals and the 3d atomic orbitals of the
peak centers of both the fitted peaks of Ti2p3/2 for all the samples are same atoms. The t2g and eg symmetry is separated by the ligand-field
shifted towards lower binding energy side with an increase in Fe-con- splitting by an energy gap of nearly 1.7 eV. However, some role of O2p
centration (x) which can be assigned to a possible charge transfer with states hybridized with 4s and 4p states of Ca and Ca 3d states and with
the neighboring ions present in the lattice. However, unlike the reports

Fig. 6. Ti2p3/2 XPS peak fitted by two peaks for the cubic perovskite system, CaCu3-xTi4-xFe2xO12 with (a) x = 0.1, (b) x = 0.3, (c) x = 0.5, (d) x = 0.7.

4
U.M. Meshiya, et al. Ceramics International xxx (xxxx) xxx–xxx

Fe3d symmetry. These admixed 3d states then hybridize with the O2p
orbitals to give Cu3dFe3d-O2p states. With the increase in Fe-sub-
stitution, one would have more density of CuFe3d-2p hybridized states
over the TiFe3d-O2p states, as seen in the XPS spectra. The intensity
ratio (P1/P2) has been shown in Table 3. The complete reversal in their
relative intensity with Fe substitution in the samples is noteworthy.
One more interesting behavior observed in the O1s XPS spectra is
that the energy difference between peak P1 and P2 shows a systematic
decrease with an increase in Fe concentration (Table 3). To understand
this let us look at Fig. 9 which compares the 3d states for (i) a free 3d
ion, (ii) an ion under the octahedral field, and (iii) an ion under the
square-planar field, respectively. The 3d states get split in the t2g and eg
states due to the crystal field. In the octahedral field, the upper eg states
are made up of dx2-y2 and dz2 orbitals while the lower t2g consists of the
dxy, dyz and dxz orbitals. In the square-planar crystal field, there is a
drastic change in the energy position of these orbitals as compared to
the octahedral field. While the orbitals dxy (in t2g) and dx2-y2 (in eg) re-
main at the same energy positions in the square-planar field, as those in
the octahedral field, the dz2 orbitals are lowered and shifts to energy
Fig. 7. The O1s XPS spectra for the system, CaCu3-xTi4-xFe2xO12 (with x = 0.0,
position even lower than the dxy, dyz and dxz orbitals of the octahedral
0.1, 0.3, 0.5 and 0.7).
field. Also, the dyz and dxz orbitals are lowered remarkably in the
square-planar field. Therefore, one can say if there is a transformation
Table 3 of symmetry from octahedral to the square planar, the density of states
O1s XPS fitting parameters for CaCu3-xTi4-xFe2xO12 system. in eg states is reduced but in the t2g states, that is increased. An average
Fe3+ - content Peak centres (eV) Peak separation Area ratio (P2/ energy position of t2g states in the square-planar field can be crudely
(x) (eV) P1) seen on the extreme right-hand side of Fig. 9, which now sets itself
lower than the t2g states of that in the octahedral field. In light of the
P1 P2
above picture, it is quite clear that intensity of the peak P2 in the O1s
0.0 529.06 530.79 1.73 3.60 XPS spectra, corresponding to the eg states, would be reduced under the
0.1 529.10 530.72 1.62 0.846 transformation of octahedral to square-planar symmetry, while that of
0.3 529.16 530.75 1.59 0.798 the peak P1 (t2g) would increase as is the case seen in our O1s XPS
0.5 529.13 530.77 1.64 0.805
spectra. We have also tabulated the energy positions of the P1 and P2
0.7 529.13 530.72 1.59 0.810
peaks in Table 3 wherefrom it can be seen that the energy difference
between P1 and P2 peaks decreases with Fe – substitution.
4sp states of Cu–Fe and Ti–Fe, in the intensity of peak P1, cannot be From the O1s XPS spectra, it can be said that difference of energy
overlooked. positions, as well as a systematic relative change in the intensity of t2g
Very notable observation in these O1s XPS spectra is the behavior of and eg peaks, point to a transformation of symmetry from TiO6 octa-
the relative intensity of P1 (t2g) and P2 (eg) peaks with an increase in Fe- hedral to the square-planar. This shows a fair agreement with Rietveld
concentration (x) as well as their energy positions. As the concentration refinement of X-ray powder diffraction data [20] and Mössbauer
of Fe increases, the t2g peak shows a strong rise in intensity while the spectral analysis [19,23].
peak eg shows a strong suppression. This could be ascribed to the strong Fig. 10 shows the Cu2p XPS spectra for all the compositions. The
hybridization of Cu–Fe3d with O2p orbitals and weakening in the hy- two peaks observed at ~933.5 eV and ~953.78 eV; corresponding to
bridization of Ti–Fe3d with O2p orbitals. The Fe3d orbitals are likely to the spin-orbit doublet Cu states Cu2p3/2 and Cu2p1/2, respectively, can
mix with the Cu3d orbitals with much more preference/probability be assigned to Cu2+ state and these peaks are consistent with Cu2+ in
than with the Ti3d orbitals, due to the energy considerations. Therefore, the CuO lattice [3,29]. However, the asymmetry and width of the line
when Fe is substituted in the system, preferentially a much stronger suggest more structures; therefore, fitting was performed on these
Cu3d-Fe3d admixed symmetry is expected to be observed than the Ti3d- peaks. Satellites of these peaks were observed at 941.54 eV and

Fig. 8. O1s XPS peak fitted with two Gaussian peaks P1 and P2 for two typical samples of. the series CaCu3-xTi4-xFe2xO12 (a) x = 0.0 and (b) x = 0.5.

5
U.M. Meshiya, et al. Ceramics International xxx (xxxx) xxx–xxx

Fig. 9. Energy level diagram of 3d states under octahedral and square-planar fields.

of charge transfer with other ions. The observed increment in the Cu

valence state points to the strong possibility of the charge transfer be-
tween Ti and Cu as evidenced by Ti XPS spectra. Also, the enhancement
in Cu valence state points to an n-type conduction mechanism in the
CaCu3Ti4O12 system via a charge transfer amongst Ti and Ti and Cu.
The finding is quite consistent with thermoelectric power measure-
ments on pure and Fe3+-substituted CaCu3Ti4O12 systems [30,31]
where electrons found responsible for the condition.
Fig. 12(a) shows the raw Fe2p XPS spectra for the typical compo-
sitions with x = 0.1,0.5, and 0.7. The diverse structures owing to initial
and final state effects, comprising asymmetric peaks, spin-orbit cou-
pling, multiplets splitting as well as shake-up satellites, are shown in the
spectra. First of all, we see that all the spectra show two major peaks at
~711.6 eV and ~724.5 eV and there is no change in their energy po-
sitions with a change in Fe-concentration (x). These peaks correspond
to the spin-orbit Fe states Fe2p3/2 and Fe2p1/2, respectively. The broad
satellites of these peaks were observed at ~720 eV and ~732 eV, re-
spectively. Fig. 12(b) shows the smoothened Fe2p XPS spectra.
It is well known that spectral peak positions and structures show
Fig. 10. The Cu2p XPS spectra of the quadruple perovskite system, CaCu3-xTi4-
chemical shifts for atoms in different chemical states, produced by ionic
xFe2xO12, with x = 0.0, 0.1, 0.3, 0.5 and 0.7 compositions.
and covalent bond differences between atoms. These shifts can very
reliably be used to distinguish between different oxidation states of the
961.52 eV, respectively. Cu 2p3/2 peak was deconvoluted into three metal ions. Therefore the Fe2p XPS spectra from reference compounds
different peaks and the fitted spectra are shown in Fig. 11, which are have been shown in Fig. 13, taken from [32]. Fig. 13 contains (i) Fe 2p
ascribed to dissimilar Cu–O coordination sites. The major peak at XPS spectrum of Fe metal (i.e. the Fe (0), measured from high purity
~933.3 eV is allocated to species comprising Cu2+ (i.e. CuO) and the iron metal foil (ii) Fe 2p XPS spectrum of Fe (II), measured from iron
component at the higher energy, ~934.7 eV, is associated with Cu3+ oxide (FeO) film and (iii) Fe (III) spectrum, measured from high purity
(i.e. Cu2O3) valance state in these samples [3,28]. Fe2O3 powder (grade 99.9995). The peak positions of Fe (0), Fe (II) and
It can be seen in Table 4 that Cu2+/Cu3+ ratio reduces considerably Fe (III) are observed at 706.8, 709.8 and 711 eV, respectively. Since the
for the substituted samples as compared to the un-substituted one energy positions of Fe2p3/2 peaks in all the investigated compositions
showing that the Cu valence state increases from 2+ towards 3+ by Fe- have been observed at ~711.6 eV, these peak positions can be safely
substitution. Now it is very relevant and important to recall our findings assigned to the trivalent state of Fe atoms and it is clear that there is no
of Ti 2p XPS spectra here. The shift of Ti 2p3/2 peaks towards the lower change in Fe3+ with a change in Fe-concentration (x). The observed
energy side with increasing Fe-concentration (x) revealed a possibility trivalence shows a fair agreement with our 57Fe Mossbauer spectral

6
U.M. Meshiya, et al. Ceramics International xxx (xxxx) xxx–xxx

Fig. 11. The Cu2p3/2 XPS peak fitted with three different Gaussian peaks for. CaCu3-xTi4-xFe2xO12 system, with (a) x = 0.0, (b) x = 0.1, (c) x = 0.3, (d) x = 0.5.

Table 4
XPS peak fitted data for Cu2p3/2 peak with two Gaussian components for Cu2+
and. Cu3+ as P1 and P2 respectively.
Fe3+ - concentration Peak positions Area P1 Area P2 Area ratio (P1/
(x) P1 P2 P2)

0.0 933.54 936.04 46958 2023 23.21

0.1 933.49 935.67 46205 7314 6.31
0.3 933.67 935.65 55226 4285 12.88
0.5 933.58 935.88 51914 4070 12.75

analysis [19,23]. The spectra are scattered having poor statistics, owing
to the low concentration of Fe in the samples, no further conclusion
except the valence state of Fe could be drawn.

4. Conclusions

From the XPS studies on quadruple perovskite system, CaCu3-xTi4-

xFe2xO12, it is deduced that Ca, Fe, and Ti metallic cations exist in di-
valent, trivalent and tetravalent stable oxidation states, respectively.
The Cu ions exist in two valence states of Cu i.e. the Cu2+ and Cu3+and
Fig. 12a. The raw Fe2p XPS spectra for the system, CaCu3-xTi4-xFe2xO12, with
enhancement in Cu valence state with an increase in Fe-content are
x = 0.1, 0.5 and 0.7 compositions.
responsible for n-type conduction mechanism. The O1s and Ti2p spec-
tral analysis indicates a strengthening of CuFe3d-O2p hybridization and
weakening of TiFe3d-O2p hybridization on Fe3+ substitution. The O1s Acknowledgments
spectra indicate an antibonding state of t2g and eg type molecular or-
bitals and a systematic relative change in the intensity of t2g and eg Sandeep Kumar thanks HSCST, Government of Haryana, India (re-
peaks points to a clear transformation of symmetry from TiO6 octahe- search grant vide letter No. HSCST/R & D/2018/2103 dated 01-08-
dral to the square-planar on Fe-substitution. 2018) and DST-PURSE sanctioned to GJUS & T, Hisar under PURSE
program No. SR/PURSE Phase 2/40 (G). Authors (PYR and KBM) want

7
U.M. Meshiya, et al. Ceramics International xxx (xxxx) xxx–xxx

coexistent features, J. Phys. D Appl. Phys. 42 (2009) 055404.

[4] A. Kokalj, Computer graphics and graphical user interfaces as tools in simulations of
matter at the atomic scale, Comput. Mater. Sci. 28 (2003) 155–168.
[5] D.C. Koningsberger, R. Prins (Eds.), X-ray Absorption, Wiley, New York, 1987.
[6] J.W. Niemantsverdriet, Spectroscopy in Catalysis, third ed., Wiley-VCH, Weinheim,
2007.
[7] R.K. Singhal, A. Samariya, Y.T. Xing, S. Kumar, S.N. Dolia, U.P. Deshpande,
T. Shripathi, E.B. Saitovitch, Electronic and magnetic properties of Co-doped ZnO
diluted magnetic semiconductor, J. Alloy. Comp. 496 (2010) 324–330.
[8] R.K. Singhal, A. Samariya, S. Kumar, Y.T. Xing, S.N. Dolia, U.P. Deshpande,
T. Shripathi, E.B. Saitovitch, Defect induced reversible ferromagnetism in hydro-
genated ZnO:Co, J. Magn. Magn. Mater. 322 (2010) 2187–2190.
[9] L. Zhang, Z.-J. Tang, Polaron relaxation and variable-range-hopping conductivity in
the giant-dielectric-constant material CaCu3Ti4O12, Phys. Rev. B. 70 (2004)
174306.
[10] A.K. Rai, K.D. Mandal, D. Kumar, Om Parkash, Characterization of nickel-doped
CCTO: CaCu2.9Ni0.1Ti4O12 and CaCu3Ti3.9Ni0.1O12 synthesized by semi-wet route, J.
Alloy. Comp. 491 (2010) 507–512.
[11] C. Wang, H.J. Zhang, P.M. He, G.H. Cao, Ti-rich and Cu-poor grain-boundary layers
of CaCu3Ti4O12 detected by x-ray photoelectron spectroscopy, Appl. Phys. Lett. 91
(2007) 052910.
[12] R. Yu, H. Xue, Z. Cao, L. Chen, Z. Xiong, Effect of oxygen sintering atmosphere on
the electrical behavior of CCTO ceramics, J. Eur. Ceram. Soc. 32 (2012) 1245–1249.
[13] A. Wen, Y. Zhang, J. Zhu, D. Yuan, Study on the semiconducting grain and in-
sulating barrier layer in aluminum/niobium co-doped CCTO, Chin. Mater Conf.
(2018) 195–203.
[14] X. Jia, B. Zhang, A. Chang, New negative temperature coefficient ceramics in La-
doped CaCu3Ti4O12 system, J. Mater. Sci. Mater. Electron. 30 (11) (2019)
Fig. 12b. The smoothened Fe2p XPS system for the system, CaCu3-xTi4- 10217–10223.
xFe2xO12,with x = 0.1, 0.5 and 0.7 compositions. [15] D. Xu, X. Yue, Y. Zhang, J. Song, X. Chem, S. Zhang, J. Ma, L. Ba, L. Zhang, S. Du,
Enhanced dielectric properties and electrical responses of cobalt-doped
CaCu3Ti4O12 thin films, J. Alloy. Comp. 773 (2019) 853–859.
[16] Y. Zhu, T. Wang, W. Wang, S. Chen, E. Lichtfouse, C. Cheng, J. Zhao, CaCu3Ti4O12,
an efficient catalyst for ibuprofen removal by activation of peroxymonosulfate
under visible-light irradiation, Environ. Chem. Lett. 17 (1) (2019) 481–486.
[17] P. Cheng, Z. Cao, M. Zhou, Q. Wang, S. Li, J. Li, Dielectric properties of
CaCu3Ti4O12 ceramics doped by La3+, Ceram. Int. (2019), https://doi.org/10.
1016/j.ceramint.2019.05.023.
[18] J. Zeng, J. Zhuang, T. He, Q. Chen, Y. Liu, Microstructure and dielectric response of
Mg-doped Cu3Ti2Ta2O12 ceramics, J. Mater. Sci. Mater. Electron. 30 (8) (2019)
2652–2658.
[19] P.R. Pansara, P.Y. Raval, N.H. Vasoya, S.N. Dolia, K.B. Modi, Intriguing structural
and magnetic properties correlation study on Fe3+-substituted calcium-copper-ti-
tanate, Phys. Chem. Chem. Phys. 20 (3) (2018) 1914–1922.
[20] P.Y. Raval, N.P. Joshi, P.R. Pansara, N.H. Vasoya, S. Kumar, S.N. Dolia, K.B. Modi,
R.K. Singhal, A Ti L3,2 - and K- edge XANES and EXAFS study on Fe3+ - substituted
CaCu3Ti4O12, Ceram. Int. 44 (17) (2018) 20716–20722.
[21] P.R. Pansara, U.M. Meshiya, A.R. Makadiya, P.Y. Raval, K.B. Modi,
P.M.G. Nambissan, Defect structure transformation during substitution in quad-
ruple perovskite CaCu3-xTi4-xFe2xO12 studied by positron annihilation spectroscopy,
Ceram. Int. 45 (2019) 18599–18603.
[22] P.Y. Raval, U.M. Meshiya, P.R. Pansara, N.H. Vasoya, C.L. Chen, C.L. Dong,
K. Asokan, K.B. Modi, Observation of orbital symmetry reversal in Fe3+-substituted
CaCu3Ti4O12 by X-ray absorption spectroscopy, Mater. Res. Bull. (2019) (commu-
nicated).
[23] P. R Pansara, P.Y. Raval, R. Pandit, J. Nehra, S.N. Dolia, Usha Chandra, K.B. Modi,
First experimental evidence of non-collinear spin structure in CaCu2.3Ti3.3Fe1.4O12
quadruple perovskite through 57Fe low-temperature, in-field Mossbauer spectro-
scopy, Mater. Res. Express (2019) (communicated).
[24] C.S. Gopinath, S.G. Hedge, A.V. Ramaswamy, S. Mahapatra, Photoemission studies
of polymorphic CaCO3 materials, Mater. Res. Bull. 37 (7) (2002) 1323–1332.
[25] R. Sanjinés, H. Tang, H. Berger, F. Gozzo, G. Margaritondo, F. Levy, Electronic
structure of anatase TiO2 oxide, J. Appl. Phys. 75 (1994) 2945–2951.
[26] N. Drnovšek, N. Daneu A. Rečnik, M. Mazaj, J. Kovač, S. Novak, Hydrothermal
Fig. 13. The reference Fe2p XPS spectra for Fe metal, FeO and Fe2O3 from [32].
synthesis of a nanocrystalline anatase layer on Ti6A4V, Implants Surf. and Tech.
203 (2009) 1462–1468.
[27] T.T. Fang, Y.H. Wang, Reassessment of copper and titanium valences and excess
to convey appreciation to Dr. S. D. Gupta, Dr. S. & S.S. Ghandhy gov-
holes in oxygen 2p levels of CaCu3Ti4O12, J. Electrochem. Soc. 158 (9) (2011)
ernment engineering college, Surat, India for his technical help to G207–G210.
provide crystal structures. [28] J. Jumpatam, B. Putasaeng, Narong Chanlek, P. Kidkhunthod, P. Thongbai,
S. Maensiri, P. Chindaprasirt, Improved giant dielectric properties of CaCu3Ti4O12
via simultaneously tuning the electrical properties of grains and grain boundaries
References by F− substitution, RSC Adv. 7 (2017) 4092–4101.
[29] C. Yin, X. Liang, G. Hu, X. Hu, X. Chen, P. Li, W. Xiang, Mater. Char. 126 (2017)
[1] M. Ahmadipur, M.F. Ain, Z.A. Ahmad, A short review on copper calcium titanate 96–103.
electro ceramics: synthesis, dielectric properties, film deposition, and sensing ap- [30] G. Chiodelli, V. Massarotti, D. Capsoni, M. Bini, C.B. Azzoni, M.C. Mozzati,
plication, Nano-Micro Lett. 8 (4) (2016) 291–311. P. Lupotto, Preparation and characterization of RF magnetron sputtered CuO/
[2] R. Schmidt, D.C. Sinclair, CaCu3Ti4O12 (CCTO) ceramics capacitor application, K. CaTi4O9 thin films with enhanced third-order nonlinear optical properties, Solid
N. Muller, in: K.N. Mullar (Ed.), Capacitors: Theory of Operation, Behavior and State Commun. 132 (2004) 241–246.
Safety Regulation, Nova Sci. Pub. Inc., 2013, pp. 1–34. [31] P. Leret, J.F. Fernandez, J. De Frutos, Nonlinear I–V electrical behaviour of doped
[3] P.R. Bueno, R. Tararan, R. Parra, E. Joanni, M.A. Ramirez, W.C. Ribeiro, E. Longo, CaCu3Ti4O12 ceramics, J. Eur. Ceram. Soc. 27 (13–15) (2007) 3901–3905.
J.A. Varela, A polaronic stacking fault defect model for CaCu3Ti4O12 material: an [32] M. Oku, K. Hirokawa, XPS observation of Fe1−xO at temperatures between 25 and
approach for the origin of the huge dielectric constant and semiconducting 800 °C, J. Appl. Phys. 50 (10) (1979) 6303.