Abstract. We present the results of X-ray diffraction (XRD), Raman scattering, absorption, impedance spectroscopy and magnetization studies of Ni2+-doped BaTi1−xNixO3. Some Ba atoms were replaced by Ni with x atoms in the range of 0.0 - 0.4. XRD analysis of these samples indicates a change in the cell parameters with doping content and there appears an abnormal point at x = 0.10. Our main focus is the Raman peak vs. Ni content, which have separated peaks in the range of 450 cm−1 - 600 cm−1. The shift in Raman and change in absorption spectra are evidence that would relate to the structure change or disorder and impurity in the samples. From the above results, a critical substituent content at which the structural phase change takes place is shown. We also present the impedance spectroscopy and magnetization vs. magnetic field to elucidate the role of dopant in the BaTi1−xNixO3 samples.
7 trang |
Chia sẻ: thanhle95 | Lượt xem: 206 | Lượt tải: 0
Bạn đang xem nội dung tài liệu A study of optical and magnetoelectric properties of BaTi1−xNixO3, để tải tài liệu về máy bạn click vào nút DOWNLOAD ở trên
JOURNAL OF SCIENCE OF HNUE
Mathematical and Physical Sci., 2012, Vol. 57, No. 7, pp. 44-50
This paper is available online at
A STUDY OF OPTICAL ANDMAGNETOELECTRIC PROPERTIES OF BaTi1−xNixO3
Nguyen Van Khanh
Faculty of Physics, Hanoi National University of Education
Abstract. We present the results of X-ray diffraction (XRD), Raman scattering,
absorption, impedance spectroscopy and magnetization studies of Ni2+-doped
BaTi1−xNixO3. Some Ba atoms were replaced by Ni with x atoms in the range of
0.0 - 0.4. XRD analysis of these samples indicates a change in the cell parameters
with doping content and there appears an abnormal point at x = 0.10. Our main
focus is the Raman peak vs. Ni content, which have separated peaks in the range
of 450 cm−1 - 600 cm−1. The shift in Raman and change in absorption spectra
are evidence that would relate to the structure change or disorder and impurity
in the samples. From the above results, a critical substituent content at which
the structural phase change takes place is shown. We also present the impedance
spectroscopy and magnetization vs. magnetic field to elucidate the role of dopant
in the BaTi1−xNixO3 samples.
Keywords: BaTi1−xNixO3, Raman scattering, absorption, impedance
spectroscopy, magnetization.
1. Introduction
Multiferroic materials are of great scientific and technological interest due to
their magnetoelectric properties, originating from the coupling between ferroelectric
and ferromagnetic order parameters [10]. The interplay between ferroelectricity and
magnetism allows a magnetic control of ferroelectric properties and an electric control
of magnetic properties and could yield new device concepts, such as ferroelectric and
multiferroic tunnel junctions [13].
Barium titanate (BaTiO3) is a ferroelectric material widely studied because of
its many potential useful properties. The phase of BaTiO3 at room temperature is a
Received October 15, 2011. Accepted February 23, 2012.
Physics Subject Classification: 62 44 07 01.
Contact Nguyen Van Khanh, e-mail address: khanhnvsp@gmail.com
44
A study of optical and magnetoelectric properties of BaTi1−xNixO3
tetragonal phase, which then transforms into cubic phase at 1300C. There are numerous
proposals regarding its application in electronic and electro-optic devices apart from its
high permittivity phenomena, associated with the basic ferroelectric behavior.
In particular, the transition metal (TM) ion doped BaTiO3 system has been of
great interest [7]. There are some reports on the effect of the doping concentration
on ferromagnetism of the BaTiO3 system. Rajamani et al. [9] prepared ferromagnetic
Ba(Ti1−xFex)O3 thin films and found that the saturation magnetization (MS) gradually
increased with a change in Fe content. On the other hand,Maier and Cohn [3] reported that
the change in magnetization with the doping concentration showed two different trends
for Ba(Ti1−xFex)O3 thin films grown on MgO and SrTiO3 substrates. W. Weidong et al.,
[12] investigated the structure of Co-BaTiO3 nano-composite films and found that the
interaction between the BaTiO3 matrix and Co nanoparticles results in a Raman shift and
an enhanced Raman scattering intensity but they were not concerned with ferromagnetic
property.
Although the dependence of magnetization on the doping concentration was
measured, the mechanism responsible has not yet been put forward because the practical
TM valence in ferromagnetic transition metal doped BaTiO3 has not been investigated
experimentally or theoretically up to now, and this is essential for discussion of optical
and magnetic behavior.
In the present study, Ni2+-doped BaTiO3 ceramics using various concentrations
were fabricated by solid-state reaction using the sol gel method. We mainly discuss
the influence of the doping concentration on the structure, optical properties and
ferromagnetism of Ni2+-doped BaTiO3 samples.
2. Content
2.1. Experiment
Nanopowders of BaTi1−xNixO3 were prepared by the polymeric precursor method
(PPM) which is based on the chelation of the metal cations by citric acid in a solution
of water and ethylene glycol. The precursor solution was prepared from a titanium
citrate formed by dissolving titanium isopropoxide in an aqueous solution of citric
acid heated to about 700C. A stoichiometric amount of Ba(NO3)2 was added to the
titanium citrate solution, which was stirred slowly until the reactional mixture became
clear. Ni(NO3)2.6H2O was then added slowly. To completely dissolve the Ba(NO3)2 and
Ni(NO3)2.6H2O, ammonium hydroxide was added a drop at a time until the pH reached
7 - 8. The complete dissolution of the salts resulted in a transparent solution. After the
solution containing Ba and Ni cations was homogenized, ethylene glycol was added to
45
Nguyen Van Khanh
promote a polyesterification reaction. The solution became more viscous as the heat
rose to 900C, albeit without any visible phase separation. The mass ratio of the citric
acid/ethylene glycol was set at 60:40. This resin was then placed in a furnace and heated
to 4500C for 4 hours, causing it to form a powder. This powder was heated at 3000C for
10 hours to obtain the disordered phase of the BaTi1−xNixO3. The crystalline phase was
obtained by heating the powder at 12000C for 8 hours.
Structural characterization was performed by means of X-ray diffraction using a
D5005 diffractometer with Cu Kα radiation. The absorption spectra were recorded using
a Jasco 670 UV-vis spectrometer and the Raman measurements were performed using
the 514.5 nm line of an Ar ion laser in a back scattering geometry using Jobin Yvon T
64000 triple spectrometer equipped with a cryogenic charge-coupled device (CCD) array
detector. Impedance spectroscopy data was obtained using Le Croy equipment.
2.2. Results and discussion
Samples of BaTiO3 doped with Ni
2+ ions of different concentrations and sintered
at 12000C were characterized by X-ray diffraction. The XRD patterns of BaTiO3 powders
doped with Ni2+ ions of various concentrations are presented in Figure 1. Peaks shown in
the patterns are sharp and well defined, indicating that all samples are well crystallized.
The patterns were compared to the reference standard cards for tetragonal and cubic
BaTiO3 [6].
An analysis of the XRD patterns showed that all samples were in the tetragonal
phase. It is well known that the dopant ionic radius is a main parameter determining the
substitutions in the BaTiO3 crystal lattice. Thus, Ni
2+ ions (0.69 A˚) are mainly placed at
Ti4+ sites (0.68 A˚) and not at Ba2+ sites (1.35 A˚) due to size incompatibility. Increasing
dopant concentration causes an overall shrinkage of the BaTiO3 unit cell, changing lattice
parameters and resulting in a change of the crystal structure (see Figure 2).
We assume that the 10% content is the limit of solubility of Ni in the BaTiO3. Using
the scherrer formula:
D =
0.89λ
β cos θ
(2.1)
and the particle size is about 130 nm.
Only the tetragonal phase is found in the BaTiO3 specimens. After the addition of
Ni, some new peaks are raised. However, the tetragonal phase is the major phase that
is found in the Ni2+-doped BaTiO3 specimens. The values of c/a ratio are shown as a
function of Ni content. The c/a ratio values drop from 1.011 to a lower value of 1.005
as the Ni content is raised above 0.10. This indicates that the Ni2+ ion is capable of
46
A study of optical and magnetoelectric properties of BaTi1−xNixO3
dissolving into BaTiO3 crystals and the solubility of Ni
2+ ion in BaTiO3 is 0.10 [5]. This
is in agreement with XRD patterns in Figure 1 and data of cell parameter in Figure 2.
Figure 1. XRD patterns Figure 2. Cell parameter vs.
of BaTi1−xNixO3 ceramics Ni content
It was known that the solubility of a dopant in BaTiO3 depends not only on the
charge but also on the radius of the dopant [8]. Though the charge of the Ni2+ ion is the
same as that of Ba2+, the radius of the Ni2+ ion (0.069 nm) is close to that of the Ti4+ ion
(0.0605 nm). Therefore, it is possible for the Ni2+ ion to substitute for the Ti2+ ion, as in
the reaction shown below:
Ni
BaTiO3−−−−−−−→ Ni′′Ti + V••O (2.2)
V ••O is the oxygen deficiency. Since the charge of the Ni
2+ ion is lower than that of
the Ti4+ ion, the Ni ion acts as the acceptor to BaTiO3.
Furthermore, an oxygen vacancy is needed to compensate for the substitution of
the Ti4+ ion with an Ni2+ ion. As the Ni content approaches the solid solution limit, the
concentration of oxygen vacancy also reaches its highest value.
The solubility of Ni in BaTiO3 is as high as 0.10, from Eq. 2.2, and the
concentration of oxygen vacancy is the same as that of Ni solute. This suggests that an
amount of oxygen vacancy is formed due to the solution of Ni into BaTiO3. The addition
of the acceptor is thus an effective method to produce a relatively large amount of oxygen
vacancy.
The Raman spectra for BaTiO3 ceramic powder samples are presented in Figure
3. BaTiO3 has five atoms and fifteen degrees of freedom per unit cell. In cubic phase
it has Oh symmetry and 15 degrees of freedom divided into the optical representations
3F1u + F2u, while another F1u symmetry mode corresponds to the acoustical branch. At
room temperature BaTiO3 is tetragonal and has C4v symmetry. The frequency covered
47
Nguyen Van Khanh
ranged from 100 cm−1 to 800 cm−1. Based on the crystallography, Raman-active modes
for tetragonal BaTiO3 (P4mm) are 4E (TO + LO) + 3A1 (TO + LO) + B1 (TO + LO) while
no Raman-active mode is predicted for the cubic phase (Pm3m). The three E (TO) modes
with frequencies of approximately 190, 280 and 516 cm−1 are labeled in Figure 3. As
discussed above, the 190 cm−1 and the 516 cm−1 modes come from the F1u cubic phase
modes and the 303 cm−1 mode comes from the splitting of the cubic silent F2u mode. The
140, 303, 640 cm−1, and the somewhat broader 720 cm−1 modes constitute the E(LO)
modes. The TO [LO splitting is fairly small and cannot be identified]. The A1(TO) mode
at 280 cm−1 is also shown in Figure 3. The intensity of the peak at around 303 cm−1 was
assigned to the overlap of E (3TO) + E (2LO) + B1. Many researchers have found that
the Raman mode around 303 cm−1 is characteristic of the tetragonal BaTiO3. The Raman
peak near 303 cm−1 shows a large decrease in intensity for increasing the Ni. Specifally,
the band near 516 cm−1 was separated into two peaks in the case of Ni2+-doped BaTiO3.
This evidence supports the comment in the XRD analysis above.
Figure 3. Raman spectra Figure 4. Absorption spectra
of BaTi1−xNixO3 ceramics of BaTi1−xNixO3 ceramics
Table 1. Optical phonon frequencies (ω) and their mode symmetry assignments
in tetragonal BaTiO3 [11]
ω (cm−1) Symmetry ω (cm−1) Symmetry
170 A1(TO) 475 A1(LO)
185 E(TO + LO) 486 E(TO)
270 A1(LO) 518 E(TO)
305 A1(LO) 520 A1(TO)
305 E(TO + LO) 715 E(TO)
463 B1 720 A1(LO)
48
A study of optical and magnetoelectric properties of BaTi1−xNixO3
Figure 5 shows the absorption spectra of BaTi1−xNixO3 ceramics with various Ni
content. Besides the absorption edge of around 450 nm in the spectra of Ni2+-doped
BaTiO3 samples, there appear some others at around 900 nm and 1350 nm. These
absorption peaks are until now not explained and an investigation is in progress. It has
already been shown that intensity dependent absorption is consistent with the presence
of secondary centers [2]. These are intermediate-level charge trapping impurity sites that
are usually highly ionized at room temperature, but can be populated by photoinduced
charge carriers from the deep centres. The photoinduced transfer of charge from a deep
to a secondary centre leads to a change in the absorption of the crystal if different
photoionisation cross sections exist for these centres. Direct intra band charge transfer
is also possible in principle but it is not usually included in band transport models.
Figure. 5 Magnetization vs. magnetic field of BaTi1−xNixO3 ceramics
Ni2+-doped BaTiO3 also produced signatures of ferromagnetism, as is shown in
Figure 2. The coercivities were again in the order a few hundred Gauss at 300K for all Ni
concentrations. XRD data showed an evidence of secondary phase formation related to
the presence of Ni clusters that would influence the magnetic properties of these samples.
The mechanism for the observed ferromagnetism is still not clear and may be
due to bound magnetic polarons [1] or carrier-induced magnetism inherent in the Zener
mechanism [4].
3. Conclusion
Barium titanate powder was obtained via the sol-gel route. The structure of BaTiO3
nanocrystallites was determined. The influence of dopant concentration on the crystal
structure of BaTiO3 was established. An addition of 0.10 Ni is essential to the formation
of a foreign phase. The critical content of 0.10 corresponds to the solid solution limit of
Ni in BaTiO3. This was also confirmed by Raman spectroscopy. The Ni
2+ ion acts as
49
Nguyen Van Khanh
the acceptor and induces the increase of oxygen vacancy concentration. The large amount
of oxygen vacancy helps the formation of h-BaTiO3. The ferromagnetic property was
detected at room temperature.
Acknowledgment. This work was supported by National Foundation for Science and
Technology Development (NAFOSTED) of Vietnam.
REFERENCES
[1] Angelescu DE, Bhatt RN, 2002. Phys. Rev. B 65 075211.
[2] Brost GA, Motes RA and Rotge JR, 1998. J. Opt. Soc. Am. B 5 1879.
[3] Cohn JL, 2002. J. Appl. Phys. 92 5429.
[4] Dietl T, Ohno H, Matsukura F, Cibert J, Ferrand D, 2000. Science 287.
[5] Huang YC, Tuan WH, 2007. Materials Chemistry and Physics 105 320.
[6] JCPDS Powder Diffraction file Card No. 31-0174, 1967. International Center for
Diffraction Data, Newtome Square, PA.
[7] Lin F, Jiang D, Ma X, Shi W, 2008. Journal of Magnetism and Magnetic Materials
320 691.
[8] Makovec D, Samardzija Z, Drofenik M, 2004. J. Am. Ceram. Soc. 87 1324.
[9] Rajamani A, Dionne GF, Bono D, Ross CA, 2005. J. Appl. Phys. 98 063907.
[10] Spaldin NA and Fiebig M, 2005. Science 309 391.
[11] Venkateswaran UD, Naik VM and Naik R, 1998. Phys. Rev. B 58 14256.
[12] Weidong W, Yingjie H, Feng W, Zhenghao C, Yongjian T, Weiguo S, 2006. Journal
of Crystal Growth 289 408.
[13] Zhuravlev MY, Jaswal SS, Tsymbal EY and Sabirianov RF, 2005. Appl. Phys. Lett.
87 22114.
50