Subscriber access provided by CORNELL UNIVERSITY LIBRARY
Article
Diffuse Phase Transitions and Giant Electrostrictive Coefficients in Lead-Free Fe3+-doped 0.5Ba(Zr0.2Ti0.8)O3-0.5(Ba0.7Ca0.3)TiO3 Ferroelectric Ceramics Li Jin, Renjie Huo, Runping Guo, Fei Li, Da-wei Wang, Ye Tian, Qingyuan Hu, Xiaoyong Wei, Zhanbing He, Yan Yan, and Gang Liu ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.6b08879 • Publication Date (Web): 24 Oct 2016 Downloaded from http://pubs.acs.org on October 25, 2016
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
ACS Applied Materials & Interfaces is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 37
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Applied Materials & Interfaces
Diffuse Phase Transitions and Giant Electrostrictive Coefficients in Lead-Free Fe3+-doped 0.5Ba(Zr0.2Ti0.8)O3-0.5(Ba0.7Ca0.3)TiO3 Ferroelectric Ceramics Li Jin,a,* Renjie Huo,a Runping Guo,a Fei Li,a,b Dawei Wang,a Ye Tian,a Qingyuan Hu,a Xiaoyong Wei,a Zhanbing He,c Yan Yan,d and Gang Liud,*
a
Electronic Materials Research Laboratory, Key Laboratory of the Ministry of Education & International
Center for Dielectric Research, Xi’an Jiaotong University, Xi’an 710049, China b
Department of Materials Science and Engineering, Pennsylvania State University, University Park,
Pennsylvania 16802, USA c
State Key Laboratory for Advanced Metals and Materials, University of Science and Technology
Beijing, Beijing 100083, China d
Faculty of Materials and Energy, Southwest University, Chongqing 400715, China
1 ACS Paragon Plus Environment
ACS Applied Materials & Interfaces
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 2 of 37
ABSTRACT: Electrostrictive effect has some advantages over the piezoelectric effect, including temperature-stability and hysteresis-free character. In the present work, we report the diffuse phase
transitions
and
electrostrictive
properties
in
lead-free
Fe3+-doped
0.5Ba(Zr0.2Ti0.8)O3-0.5(Ba0.7Ca0.3)TiO3 (BZT-0.5BCT) ferroelectric ceramics. The doping concentration was set from 0.25 mole % to 2 mole %. It is found that by introducing Fe3+ ion into BZT-0.5BCT, the temperature corresponding to permittivity maximum Tm was shifted toward lower temperature monotonically by 37 oC per mole % Fe3+ ion. Simultaneously, the phase transitions gradually changed from classical ferroelectric-to-paraelectric phase transitions into diffuse phase transitions with a weak relaxor characteristic. Purely electrostrictive responses with giant electrostrictive coefficient Q33 between 0.04 m4/C2 and 0.05 m4/C2 are observed from 25 oC to 100 oC for the compositions doped with 1 mole %~2 mole % Fe3+ ion. The Q33 of Fe3+-doped BZT-0.5BCT ceramics is almost twice higher than the Q33 of other ferroelectric ceramics. These observations suggest that the present system can be considered as a potential lead-free material for the applications in electrostrictive area and BT-based ferroelectric ceramics would have giant electrostrictive coefficient over other ferroelectric systems.
KEYWORDS: diffuse phase transition, electrostriction, doping, lead-free, ferroelectric
2 ACS Paragon Plus Environment
Page 3 of 37
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Applied Materials & Interfaces
INTRODUCTION
Piezoelectric materials are widely employed in many precisely controlled devices and systems due to their actuating resolution as high as nanometer scale.1,2 In principle, the definition of the converse piezoelectric effect renders the relationship between applied electric field and induced mechanical strain linear. However, for most of conventional piezoelectric ceramics, such as Pb(Zr,Ti)O3 (PZT), a strong hysteresis occurs not only in polarization-electric field (P-E) loops but also in strain-electric field (S-E) curves because of the extrinsic contribution by ferroelectric domain wall motion.1-4 Such an inherent hysteresis significantly breaks the actuating performance of piezoelectric materials. On the other hand, it is noticed that all the piezoelectric ceramics should be pre-poled before their use, so that the cost and rate of piezoelectric ceramics are always high.5-10 In the framework of thermodynamics, besides the linear converse piezoelectric effect between electric field and induced strain, the quadratic electrostrictive effect is a more general effect for all dielectrics or insulators, regardless of the symmetry of materials.4,11 Electrostrictive effect has no symmetry limitation so that it can be observed in all crystal symmetries. This universal effect is generally expressed in the form of:4 Sij=QijklPkPl,
(1)
Sij=MijklEkEl,
(2)
where Sij, Pi and Ei are mechanical strain, polarization, and electric field, respectively. Qijkl are the electrostrictive coefficients with respect to polarization, while Mijkl are the electrostrictive coefficients with respect to electric field. For nonlinear dielectrics, Q coefficients are more frequently used than M coefficients, since Eq. (1) is more fundamental while Eq. (2) only holds true when the electric field is quite small. Coupled by electrostrictive effect, mechanical strain can be generated in ferroelectric materials without pro-poling procedure and most importantly, eliminating the influence from extrinsic domain wall motion and associated hysteresis phenomenon. Unfortunately for most of dielectric materials, the electrostrictive effect is trivial and the 3 ACS Paragon Plus Environment
ACS Applied Materials & Interfaces
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 4 of 37
strain level is far below 0.1%. In 1980s, the pioneering works by Uchino et al. triggered the development of electrostrictive materials.12-14 Relaxor ferroelectrics with perovskite ABO3 structures,
including
Pb(Mg1/3Nb2/3)O3
(PMN)
and
Pb(Zn1/3Nb2/3)O3
(PZN)
single
crystals/ceramics were found to possess an temperature-independent electrostrictive strain with its maximum value as high as 0.1 %, while corresponding longitudinal electrostrictive coefficient Q33 values are around 0.015~0.025 m4/C2. These relaxor ferroelectrics were regarded as alternatives to piezoelectric materials for actuating application.15 However, due to the limitation of the lead usage, developing of lead-free ferroelectric materials becomes urgent.16-20 The electrostrictive properties have been reported for several lead-free ferroelectric ceramics.21-27. As suggested by Ang and Yu, the Q33 value of (Sr0.35Na0.25Bi0.35)TiO3 (SNBT) ceramic approaches 0.020 m4/C2, but its driving electric field is much higher than that used in PMN.21 Moreover, Q33 values
from
0.021–0.027
m4/C2
were
reported
in
antiferroelectric
Bi0.5Na0.5TiO3-BaTiO3-K0.5N0.5NbO3 (BNT-BT-KNN) ceramics.22 However, relatively large hysteresis also accompanied inherently due to the nature of antiferroelectricity. Recently, a ultrahigh
Q33
value
of
0.04
m4/C2
was
identified
in
lead-free
(1–
x)Ba(Zr0.2Ti0.8)O3-x(Ba0.7Ca0.3)TiO3 (BZT-xBCT, 0.4≤x≤0.6) ferroelectric ceramics.28 However, due to its high Curie temperature TC (96 oC for BZT-0.5BCT), purely electrostrictive response, i.e. hysteresis-free characteristic, can be only obtained above 100 oC, restricting their application as electrostrictive materials at room temperature. From the point view of application, hysteresis in S-E curves should be avoided as much as possible. A purely hysteresis-free electrostrictive strain can be obtained from relaxor ferroelectrics when the temperature is above their Tm (temperature correspond to dielectric permittivity maximum εm), due to the lack of macroscopic domains.29,30 This characteristic is very important to the precisely controlled devices. Thus searching for lead-free relaxor ferroelectrics with high electrostrictive coefficients is the first step to obtain hysteresis-free electrostrictive materials. Being different from conventional lead-based piezoelectric materials, 4 ACS Paragon Plus Environment
Page 5 of 37
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Applied Materials & Interfaces
such as PZT and PbTiO3 (PT), the ferroelectricity of BZT-xBCT system is rather weak because of its low TC.28 The low TC of BZT-xBCT ceramics limits their applications as piezoelectric materials, especially at high-temperature region. On the other hand, this feature gives us a chance to shift the TC of BZT-xBCT to or below room temperature through chemical substitution, i.e., doping. Therefore, the question is how to lower the TC most effectively? In literatures, it is reported that the Fe3+ ion lower the TC of BaTiO3 (BT) ceramics and crystals by 40~75 oC per mole %.31-33 This modulating effect on TC is the largest compared with the doping effect by other elements. Therefore, a similar effect for the shift of TC is highly expected in Fe3+ ion doped BZT-0.5BCT. Furthermore, in PZT ceramics, Fe3+ ions are introduced in order to form the defect-oxygen dipoles and then pin the domain wall motion, resulting in a decrease of the hysteresis. This effect is also needed to verify in the BZT-xBCT system.3 In this work, we prepared the Fe3+-doped BZT-0.5BCT ceramics with different doping concentration, in order to clarify the doping effect by Fe3+ ions on TC and related ferroelectric and electrostrictive properties. It is found that both Tm and εm are decreased almost linearly by increasing the Fe3+ ions content from 0 to 2 mole %, with rates of 37 oC and 3932 per mole %, respectively. Typical ferroelectric-to-paraelectric phase transitions gradually become diffuse phase transitions, while the diffuseness temperature increases linearly as well. However, compositions with diffuse phase transition features display rather weak relaxor characteristics. A mechanism related to the breaking of the long-range order is proposed to explain the above phenomenon. It is suggested the hysteresis in P-E loops and S-E curves is reduced drastically by increasing the Fe3+ content. Giant Q33 values between 0.04 to 0.05 m4/C2 and a hysteresis-free characteristic in S-E curves are observed over a wide temperature region and compositions. It is believed that the giant Q33 and their good temperature-stability would attract potential interest in search of novel lead-free electrostrictive materials.
EXPERIMENTAL SECTION 5 ACS Paragon Plus Environment
ACS Applied Materials & Interfaces
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Preparation of Fe3+-doped BZT-0.5BCT Ceramics. Undoped and Fe3+-doped BZT-0.5BCT [(Ba0.85Ca0.15)(Zr0.1Ti0.9)O3] ceramic samples with a nominal formula as Ba0.85Ca0.15(Zr0.1Ti0.9)1−xFexO3 with x=0, 0.25 %, 0.5 %, 1 % and 2 % were prepared by a conventional solid state process using the standard mixed oxide route. High purity oxide powders of BaCO3 (99.8 %), CaCO3 (99.5 %), ZrO2 (99.96 %), TiO2 (99.985 %), Fe2O3 (99.99 %) (Aladdin Reagent Co. Ltd, Shanghai, China) were used as starting materials. Raw materials in stoichiometric amounts were mixed for 6 h using ball mixing in alcohol with yttria-stablized zirconia balls. The mixed powders were calcined at 1250 oC and milled again for 24 h. The milled powders were mixed with 5 wt. % polyvinyl acetate binder and uniaxially pressed at 150 MPa to form cylinders with 12 mm in diameter and 10 mm in height. The green bodies were covered by the same composition powders and sintered at temperature of 1450 oC for 4 h in sealed crucibles with a heating rate of 3 oC/min. Ceramic samples were cut into different dimensions according to the requirements for following properties characterizations. Properties Characterization. Grinded powders of the sintering ceramics were used to identify the crystal structures by means of X-ray diffraction (XRD, D/Max-IIIC, Rigaku, Tokyo, Japan) analysis using CuKα radiation in a θ-2θ scanning pattern from 20o to 70o with a step of 0.02o and 1 s per step. Surface morphologies were revealed by means of scanning electron microscopy (SEM, Quanta FEG 250, FEI, Hillsboro, USA). The surfaces of studied sample were polished by silica paper, and then thermal etched at 1350 oC for 15 min. Thin-disk samples with a diameter of 11 mm and thickness of 1 mm were used for electromechanical properties measurement. Silver pastes were coated on the main surfaces of samples and burnt out at 600 oC for 1 h to fabricate electrodes for electric properties characterizations. Dielectric permittivity and loss tangent as a function of temperature were measured by a self-built system at 1 kHz, 10 kHz, and 100 kHz with a heating rate of 2 oC/min combine with a multi-frequency LCR meter (E4980A, Agilent, Palo Alto, USA). P-E loops and S-E curves were measured simultaneously based on a Sawyer-Tower circuit (TF analyzer 2000, aixACCT, Aachen, Germany) combine 6 ACS Paragon Plus Environment
Page 6 of 37
Page 7 of 37
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Applied Materials & Interfaces
with a photonic displacement sensor (MTI-2000, MTI Instruments, Washington, U.S.) at a frequency of 0.1 Hz. Based on P-E loops and S-E curves, longitudinal electrostrictive coefficient Q33 was calculated through the combined S-P curves using Eq. (1).
RESULTS AND DISCUSSION Structural and Microstructural Analyses. Room temperature XRD patterns of
BZT-0.5BCT ceramics doped with different Fe3+ ion content are shown in Figure 1. Single phase perovskite structures are obtained for all studied ceramics samples. Only the diffraction peaks for a pseudocubic symmetry are indexed within the scanning range from 20o to 70o. The peak splitting of the (200) group is absent, suggesting no non-cubic distortions existing in doped compositions. In other acceptor-doped BCZT ceramics, a similar pseudocubic symmetry was reported.34 Figure 2 displays the density as a function of Fe3+ ion content. All ceramic samples were sintered at 1450 oC. With introducing of the Fe3+ ion into BZT-0.5BCT system, the density shows an abrupt drop at first (composition with x=0.25 %). Then the density gradually increases as the doping content is increased. However, all Fe3+-doped samples show lower density compared to undoped composition. The reason for the sintering mechanism is not clear. Figure 3 displays the SEM surface micrographs of the sintered samples with x=0, 0.25 %, 0.5 %, 1 % and 2 %. It can be seen that the grain size is very sensitive to the Fe3+ ion content. For undoped BZT-0.5BCT sample, the mean grain size is about 35 µm, which is estimated by the linear intercept method and in agreement with previous report.34 With increasing the Fe3+ content from x=0.25 % to x=2 %, the grain size drops dramatically from 20 µm to 10 µm. The inhabitation of grain growth is attributed to the doping of Fe3+ ions. It is believed that introducing Fe3+ ions results in oxygen vaccines, which reduce the mobility of grain boundaries, and as a consequence, inhibit the grain growth. This trend is consistent with the studies concerning the acceptor-doped effect on grain size in lead-based and lead-free ferroelectric 7 ACS Paragon Plus Environment
ACS Applied Materials & Interfaces
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ceramics.35-37 The distribution of each chemical element for the composition with x=2 % was characterized by the element mapping of the selected cross section of the sample using energy-dispersion X-ray (EDX) analysis, as shown in Figure 4. One can see that all the elements are homogenously distributed in the ceramic sample. No two phase composite structures could be detected from Figure 4. Dielectric Properties and Diffuse Phase Transitions. The permittivity εr and loss tangent tanδ as a function of temperature of the ceramic samples with different Fe3+ content measured at 1 kHz, 10 kHz, and 100 kHz are illustrated in Figure 5a-e. The dielectric properties of undoped BZT-0.5BCT are shown in Figure 5a. A very sharp peak of εr denoting a typical ferroelectric-to-paraelectric phase transition is observed at 96 oC, which is in agreement with previous observations.34,38 For Fe3+-doped BZT-0.5BCT ceramic samples, the sharp phase transition peaks gradually transfer into broad peaks with respect to the increase of the Fe3+ content from 0.25 mole % to 2 mole %, as shown in Figure 5b-e. In addition, except the composition with x=2 %, the tanδ of other compositions show peaks, and the corresponding temperatures of these peaks are few degrees below the Tm determined from εr. In contrast, the tanδ for x=2 % decreases with increasing the temperature, and no obvious phase transition feature could be detected from it. Figure 5f plots Tm as a function of Fe3+ doping content x. A linear relation between Tm and x is suggested. It is indicated that 1 mole % Fe3+ decreases the Tm by 37 oC. Furthermore, the εm (εr measured at Tm) for each composition decreases as x increases as well. In Figure 5f, a roughly linear relationship between εm and x is extracted, suggesting that 1 mole % Fe3+ lowers the εm by 3932. In conventional lead-based ferroelectrics, such as PZT, Fe3+ dopant shifts the Tm to lower temperature by several degrees per mole.39 In contrast, the shift of Tm in BT is much larger at the same doping level.31-33 This phenomenon could be explained in terms of the origin of ferroelectricity. It is suggested that the orbital hybridization between the Ti 3d states and O 2p 8 ACS Paragon Plus Environment
Page 8 of 37
Page 9 of 37
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Applied Materials & Interfaces
states is essential to the ferroelectric instability in both PT and BT, while the hybridization between Pb 6s and O 2p ions is crucial for larger ferroelectricity in PT.40,41 In this work, we assume that Fe3+ ion would substitute (Zr4+,Ti4+) ion and enter the B-site, since its ion radii is very close to the B-site ions, such as Zr4+ and Ti4+ ions. This conjecture is also verified by means of electronic paramagnetic resonance measurent.42,43 Therefore, to lead-based ferroelectrics, incorporating Fe3+ ion into the perovskite structure would not disrupt the orbital hybridization between the A-site ions and O ions remarkably. In contrast, to lead-free ferroelectrics, it would disrupt orbital hybridization between the B-site ions and O ions seriously, resulting in a drastic decrease of Tm in BZT-0.5BCT system. To further understand the mechanism of Fe3+ that strongly changes properties of doped BCZT system, we performed effective Hamiltonian based Monte-Carlo simulations on Fe3+ doped BaTiO3.44 Our preliminary results have shown that setting the Ti-O dipoles to null at the site of Fe3+ only is able to account for most of the effects caused by doping Fe3+ ion, including the shift of transition temperatures and the widening of dielectric constant peaks. Such results lead us to believe that dipoles caused by doping play a major role in preventing the development of long-range ferroelectric ordering, and therefore inducing relaxor-like properties. For compositions with x=0, 0.25 %, and 0.5 %, weak frequency dispersion are observed in the vicinity of their respective Tm, and their Tm measured at 1 kHz, 10 kHz, and 100 kHz are almost the same. In contrast, for the compositions with x=1 % and 2 %, their εr shows weak frequency dispersions below the Tm (i.e., the εr is insensitive to frequency), while frequency dispersions disappear when the temperature is few degrees above their Tm. These features suggest that Fe3+ ion doping induces a diffuse phase transition (DPT) with a rather weak relaxor characteristic.30,45-48 In order to evaluate the degree of DPT diffuseness (or the width) quantitatively, two methods were adopted. One method is based on the definitions of W2/3M-L and W2/3M-H.49 As shown in Figure 6a, they are defined as the difference between the Tm and the temperatures where εr reaches 2/3 of the εm from low-temperature and high-temperature sides 9 ACS Paragon Plus Environment
ACS Applied Materials & Interfaces
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 10 of 37
of the peak, respectively. Here, we chose W2/3M-H instead of W2/3M-L to evaluate the degree of diffuseness of the DPT peaks, since the Curie-Weiss law only holds for the description of the dielectric response above TC, and the second method also extracts the diffuseness parameters by means of fitting the εr above Tm. The diffuseness temperature as a function of x derived from W2/3M-H measured at 1 kHz is shown in Figure 6a and Table 1. The diffuseness temperature for x=0 is only 15.4 oC. As the x is increased to 2 % ceramic sample, the corresponding temperature is 47.4 oC, which is nearly three times over the magnitude for x=0. By incorporating Fe3+ ion into the system, the diffuseness temperature is increased almost linearly as indicated in Figure 6c. Another method used to calculate the diffuseness temperature (δg) was developed by Smolenskii and Rolve based on a Gaussian distribution of TC.50,51 They considered the diffuseness roughly analogues to the standard deviation of the Gaussian distribution. According to such an assumption, the permittivity εr as a function of temperature is expressed as,
(T − Tm ) 2 ε m = ε r exp , 2 2δ g
(3)
where, δg is the Gaussian diffuseness temperature. By expanding Eq. (3) as a power series and simplify its higher orders with limitation of 1< εm/εr
