PCCP View Article Online
Published on 31 March 2014. Downloaded by University of Newcastle on 12/09/2014 15:35:38.
PAPER
Cite this: Phys. Chem. Chem. Phys., 2014, 16, 13078
View Journal | View Issue
Enhanced piezoelectric and mechanical properties of AlN-modified BaTiO3 composite ceramics Dan Xu,ab Lidong Wang,a Weili Li,ac Wei Wang,a Yafei Hou,a Wenping Cao,a Yu Fenga and Weidong Fei*a BaTiO3–xAlN (BT–xAlN) composite ceramics were prepared by conventional solid state reaction sintering. The effects of the AlN content on the crystalline structures, densities, and electrical and mechanical properties of the BT ceramics were investigated. The BT–1.5%AlN ceramic exhibits a good piezoelectric constant of 305 pC N1 and an improved Vickers hardness of 5.9 GPa. The enhanced piezoelectricity originates from interactions between defect dipoles and spontaneous polarization inside the domains due to the occurrence of local symmetry, caused by the preferential distribution of the Al3+–N3 pairs vertical
Received 24th February 2014, Accepted 30th March 2014
to the c axis. The hardening of the material is attributed to the improved density, and particle and grain
DOI: 10.1039/c4cp00796d
boundary strengthening. Our work indicates that if a suitable doping ion pair is designed, lead-free
www.rsc.org/pccp
ceramic systems prepared from ordinary raw materials by a conventional sintering method have a high probability of exhibiting good piezoelectric and mechanical properties simultaneously.
1. Introduction In recent years, considering the effects of lead on human health and environmental protection, lead-free piezoelectric ceramics have been extensively investigated in order to find a replacement for the toxic lead zirconate titanate (PZT).1–5 Among the candidates is barium titanate (BaTiO3, BT), with an ABO3-type perovskite structure, it was one of the most widely used ferroelectric materials before the discovery of high-performance PZT, which has been used in a variety of applications such as PTC thermistors, multi-layer ceramic capacitors (MLCC), ferroelectric random access memories (FRAM), and optical devices, because of its outstanding dielectric and ferroelectric properties.6–8 Nowadays, however, the practical applications of BT ceramics are still subject to their relatively poor piezoelectric properties in comparison with those of PZT. BT ceramics with high piezoelectric properties are hard to acquire just by using ordinary raw materials and a conventional sintering method. BT ceramics are therefore used mainly as dielectric materials rather than as piezoelectric materials, unless employing special fine powders8 and unusual sintering techniques9–12 to enhance their piezoelectric properties. For example, especially high d33 values have been reported for those BT ceramics prepared by
a
School of Materials Science and Engineering, Harbin Institute of Technology, Harbin 150001, People’s Republic of China. E-mail: [email protected]; Fax: +86-451-86413908; Tel: +86-451-86413908 b Jilin Institute of Chemical Technology, Jilin 132022, People’s Republic of China c National Key Laboratory of Science and Technology on Precision Heat Processing of Metals, Harbin Institute of Technology, Harbin 150001, People’s Republic of China
13078 | Phys. Chem. Chem. Phys., 2014, 16, 13078--13085
special fabrication methods such as spark plasma sintering (416 pC N1),9 microwave sintering (370 pC N1)10 and two-step sintering (460 pC N1).11 The origin of the high piezoelectric properties for these special BT ceramics is still controversial.9,13–16 Two crucial factors, fine grains (several micrometers) and nano-domain structures (tens of nanometers), are believed to be responsible for the high d33 values observed in these BT ceramics. In this case, the existing problem is how to obtain high piezoelectric properties for BT and other ABO3-type ferroelectric ceramics using a traditional preparation method. Up to now, the most basic approach available for achieving high piezoelectricities of piezoelectric ceramics, excluding the fine-grain effect, is to induce a phase transition between two ferroelectric phases either by the morphotropic phase boundary (MPB) effect or by the polymorphic phase transition (PPT) effect. As an example, Ren et al. reported that the Ba(Ti0.8Zr0.2)O3– x(Ba0.7Ca0.3)TiO3 (BZT–xBCT) system exhibited a very high piezoelectric coefficient (d33) of E620 pC N1 at the MPB starting from a tricritical triple point (TCP) of a cubic paraelectric phase (C), a ferroelectric rhombohedral (R) phase, and a tetragonal (T) phase.17 Unfortunately, the usefulness of the BZT–xBCT system is still restricted due to its relatively low Curie temperatures (TC). Another high piezoelectric mechanism is the PPT mainly found in the (K, Na)NbO3 (KNN) system.18,19 The high levels of piezoelectricity reported in the KNN system are the result of the enhanced polarizability associated with the orthorhombic (O)–tetragonal (T) polymorphic phase transition being compositionally shifted downward to near room temperature. Such a temperature-driven ferroelectric–ferroelectric phase transition may facilitate polarization rotations. Although a high d33 of
This journal is © the Owner Societies 2014
View Article Online
Published on 31 March 2014. Downloaded by University of Newcastle on 12/09/2014 15:35:38.
Paper
E300 pC N1 is achieved for non-textured KNN-based ceramics, the inherently large temperature dependence of their piezoelectric properties becomes the biggest problem for the stability and reliability of the devices operating at high temperatures. On the other hand, although relatively high piezoelectric constants (d33 = 200–300 pC N1) have been realized, the ordinary BT-based piezoelectric ceramics still suffer from low mechanical strength and poor reliability under loading conditions, which limits their applications as transducers, sensors and actuators where high fracture toughness and fracture strength are required. Many studies on BT-based composites incorporated nano-particulates, including Al2O3,20 ZrO2,21 MoO3,22 NiO,22 SiC,23 Ag24,25 and Ni,26 in order to improve their mechanical properties. It was found that both oxide and metal additives could improve the mechanical properties of the BT-based composite ceramics. However, there are only a few studies on their electrical properties and in particular their piezoelectric properties.20–26 T. Kojima et al.25 reported that the Vickers hardness (Hv) of the BT–Ag composite prepared by pressureless sintering was 6.2 GPa, much higher than the monolithic BT ceramic with Hv = 2.5 GPa. BaZr0.07Ti0.93O3 ceramics, synthesized using a modified two step sintering process, exhibited an enhanced Knoop hardness (Hk) of 5.5 GPa, and the dwell time was up to 4 h.21 S. Jiansirisomboon et al.20 found that the Knoop hardness of BT/xAl2O3 nanocomposite ceramics increased with increasing Al2O3 content, and that the ceramic with optimum properties (Hk = 4.4 GPa, d33 = 178 pC N1) was obtained at x = 0.5 vol%. Obviously, it is not easy to obtain a BT ceramic with simultaneously good piezoelectric and mechanical properties. BT composite ceramics will sacrifice their piezoelectric properties for the enhancement of mechanical properties in most cases. Thus there is an urgent need to develop BT ceramics with enhanced piezoelectric and mechanical properties. As is well known, the microstructures and physical properties of BT ceramics can be modified by appropriate substitutions at A- and/or B-sites.27–31 However, previous works have seldom involved A- and O-site replacements. Therefore, in this paper, the non-oxide compound, AlN, is added to the BT matrix. The electrical and mechanical properties of BT–xAlN composite ceramics, prepared from ordinary raw materials by a conventional solid state reaction sintering method, were investigated and analyzed. It is expected that the addition of AlN to the BT would be a good choice to enhance its mechanical properties without deteriorating its electrical properties. The reinforcement mechanisms of AlN on the piezoelectric and mechanical properties of BT composite ceramics are discussed as well.
2. Experimental BaTiO3–xAlN composite ceramics (abbreviated as BT–xAlN, x = 0–14 mol%) were prepared by conventional solid state reaction sintering. Commercially available BT (99.9%, Aladdin Chemistry Co. Ltd, Shanghai, China) and AlN (99.9%, Eno High-Tech Material Development Co., Ltd, Qinhuangdao,
This journal is © the Owner Societies 2014
PCCP
China) powders were used as the starting materials. The BT and AlN powders were weighed in the desired compositions and ball-milled for 8 h by planetary milling with zirconia balls in alcohol. After ball-milling, the slurries were dried at 80 1C for 12 h, ground and sieved. The mixed powders were subsequently pressed into pellets of 10 mm in diameter and 1 mm in thickness using a few drops of 5 wt% polyvinyl alcohol (PVA) as a binder. After burning off the PVA, the BT–xAlN pellets embedded in the BT powders were sintered in covered alumina crucibles at 1325 1C for 2 h with a heating rate of 5 1C min1. The densities were measured by the Archimedes method. The crystalline structure of each sample was analyzed by X-ray diffraction (XRD) on a Philips X’Pert diffractometer with Cu Ka radiation, operating at 40 kV and 40 mA. The y–2y scan was performed at room temperature with a step size of 0.021 and time per step of 2 s. The microstructures of the sintered samples were examined by scanning electron microscopy (SEM) using a Quanta 200F. The temperature dependence of the dielectric constants was measured at 1 kHz in the temperature range of 25–200 1C on a LCR meter (Agilent, E4980A). Polarization versus electric field (P–E) hysteresis loops were determined by a Radiant Technologies Precision workstation. The samples were poled at 80 1C in a silicone oil bath under a DC field of 3 kV mm1 for 30 min. Piezoelectric constants (d33) were measured on a quasi-static piezoelectric constant testing meter (ZJ-4AN, Institute of Acoustics, Chinese Academy of Science). The well-polished ceramics were subjected to a microhardness tester (HV-1000, Cany Precision Instrument Co., Shanghai, China) for Vickers hardness (Hv) determination using a 300 g load with a holding time of 15 s for hardness measurements.
3. Results and discussion 3.1.
Phase and microstructure analysis
The room temperature XRD patterns of the BT–xAlN (0 r x r 14 mol%) ceramic powders in the 2y range of 20–851 are shown in Fig. 1a. A pure perovskite phase is formed up to x r 1.5%, while weak second phases around 2y = 28.2–28.81 are observed when x Z 3%, which are identified as BaAl2O4 (PDF No. 17-0306), and are marked with a shaded rectangle in the XRD patterns (Fig. 1a). The solubility limit of Al in the BT lattice is between x = 1.5–3%. The fine scan XRD patterns for the BT–xAlN ceramic powders in the 2y ranges of 64.5–67.01 and 83.0–84.01 are shown in Fig. 1b and c. All of the ceramic powders appear to adopt a tetragonal crystal symmetry (PDF No. 83-1880), as evidenced by the splitting of the 202/220 peaks at 2y E 661 (Fig. 1b) and the existence of a single 222 peak at 2y E 83.51 (Fig. 1c). The slightly reduced 222 interplanar spacing (d222) with increasing AlN content, as revealed in Fig. 1d, further affirms that the BT has been doped with AlN. Considering the very low solubility limit of Al in BT and the smaller ionic radius of Al3+ compared with those of Ba2+ and Ti4+,32 it is reasonable that the 222 peak position shifts slightly towards right with increasing AlN content (Fig. 1c). However, it is
Phys. Chem. Chem. Phys., 2014, 16, 13078--13085 | 13079
View Article Online
Published on 31 March 2014. Downloaded by University of Newcastle on 12/09/2014 15:35:38.
PCCP
Paper
Fig. 1 (a) XRD patterns of BT–xAlN ceramic powders measured at room temperature. Fine scan XRD patterns in the 2y ranges of (b) 64.5–67.01 and (c) 83.0–84.01. (d) 222 interplanar spacing (d222) with increasing AlN content. (e) Variation of the relative intensities of the 002/200 peaks.
still unclear how the Al3+ and N3 ions arrange their locations in the BT lattice. In order to study the distribution of the Al3+ and N3 ions in the BT lattice, the variation of the relative intensities of the 002/200 peaks (I002/I200) was plotted, as shown in Fig. 1e. It was found that the I002/I200 ratio first increased, achieved a maximum value at x = 1.5%, and subsequently decreased with increasing AlN content. It must be noted that the variation of the relative intensities of the 002/200 peaks with AlN content cannot be explained just by a random distribution of Al3+ and N3 ions. If such a random distribution was true, then the XRD results of Fig. 1e would not be observed as a consequence of the same influence on the structure factor (F) for the 002 and 200 diffractions in this case. According to the ionic radii and valences, we believe that the Al3+ and N3 ions should form Al3+–N3 pairs in the same cell in order to minimize the electrostatic energy. Therefore, the most energy-favored way of replacement is to exchange Al3+ for Ba2+ sites, and N3 for O2 sites, accompanied by the maintenance of the local charge neutrality. Thus, there exist two possible distributions for the Al3+–N3 pairs. The first is that the distribution direction of the Al3+–N3 pairs is perpendicular to the c axis, and the second is that the Li+–Al3+ pairs arrange their locations in other directions. Due to the different ionic radii among the Al3+–N3 pairs, and the Ba2+ and O2 ions, the occurrence of Al3+–N3 pairs will induce lattice distortions in the BT, leading to a weakened XRD intensity.33 In the case of the (002), (020) and (200) planes, the
13080 | Phys. Chem. Chem. Phys., 2014, 16, 13078--13085
eff eff effective structure factors, F eff 002, F 020, and F 200, adjusted by lattice distortions, can be expressed as follows 8p2 sin2 y 2 eff F200 ¼ F200 expðM1 Þ ¼ F200 exp Dx (1) l2
8p2 sin2 y 2 eff ¼ F020 expðM2 Þ ¼ F020 exp Dy F020 l2
(2)
8p2 sin2 y 2 eff ¼ F002 expðM3 Þ ¼ F002 exp Dz F002 l2
(3)
where F% 200, F% 020 and F% 002 are the average structure factors of the distortion-free crystal lattices, and Dx2, Dy2 and Dz2, which represent deviations from the ideal BT lattice sites, are the mean square displacements in the [100], [010] and [001] directions, respectively. Since the diffraction intensity (I) is proportional to the structure factor’s square modulus (|F|2), the lattice distortion caused by the replacement of Al3+–N3 pairs, which primarily occurs along the Al3+–N3 lines, is an important factor influencing the XRD intensity. Based on the above analysis, the observed variations in the I002/I200 intensity ratio shown in Fig. 1e can be explained as follows; when x r 1.5%, the Al3+–N3 pairs tend to adopt a preferential distribution vertical to the c axis and hence, the lattice distortion induced by the Al3+–N3 pairs will generate little influence on the M3 value of eqn (3), because the lattice distortion primarily occurs along the Al3+–N3 lines. However, as the AlN content further
This journal is © the Owner Societies 2014
View Article Online
Published on 31 March 2014. Downloaded by University of Newcastle on 12/09/2014 15:35:38.
Paper
Fig. 2
PCCP
Surface SEM micrographs of BT–xAlN ceramics, (a) x = 0%, (b) x = 0.75%, (c) x = 1.5%, (d) x = 3%, (e) x = 9% and (f) x = 14%, sintered at 1325 1C for 2 h.
increases (x 4 1.5%), the I002/I200 intensity ratio becomes weakened, indicating that the number of Al3+–N3 pairs vertical to the c axis begins to decrease due to the presence of the second phase, BaAl2O4, and the elastic energy limitation. The surface SEM micrographs of the BT–xAlN ceramics with different AlN contents are presented in Fig. 2. All of the ceramic samples are very dense without clear pores on the ceramic surfaces. As shown in Fig. 2, the abnormal growth of grains can be found in the pure BT ceramic, and this case was modulated by incorporating a small amount of AlN (Fig. 2b, c) to the BT matrix. Fig. 3 shows the variations of the bulk densities and the average grain sizes of the BT–xAlN ceramics with different AlN content. It is evident that the density is improved to some extent by introducing AlN. Furthermore, the average grain size decreases with increasing AlN content, indicating that the addition of AlN can effectively limit the grain growth of BT. 3.2. Dielectric ‘‘relaxorlike’’ behavior and ferroelectric properties The temperature-dependent dielectric constants of BT–xAlN composite ceramics with compositions of x = 0, 0.75, 1.5, 3, 9 and 14 mol% measured at 1 kHz are displayed in Fig. 4.
Fig. 3
Density (r) as a function of average grain size (d).
This journal is © the Owner Societies 2014
As revealed in Fig. 4, there is only one dielectric anomaly around 120 1C–130 1C, characterized by a phase transition from the ferroelectric phase (tetragonal phase, T) to the paraelectric phase (cubic phase, C) within the measured temperature range. The e peak associated with the T–C phase transformation is much sharper for the coarse-grained samples and becomes more diffuse with reducing grain size (as x Z 9%). This result indicates that the BT–9%AlN and BT–14%AlN ceramics show a relaxorlike characteristic due to the occurrence of the diffuse phase transition (DPT). It is well known that for a normal ferroelectric, the dielectric permittivity above TC follows the Curie–Weiss law, 1 T TC ¼ ; e C1
T 4 TC
(4)
where C1 is the Curie–Weiss constant. The inverse dielectric permittivity measured at 1 kHz was plotted as a function of temperature, as shown in Fig. 5. For a relaxor ferroelectric with a DPT, the degree of diffuseness of the paraelectric–ferroelectric phase transition can be more effectively described by a modified Curie–Weiss law,34,35 1 1 ðT Tm Þg ¼ ; C2 e em
1g2
(5)
where e is the dielectric permittivity at temperature T, em is the maximum value of the dielectric permittivity, Tm is the temperature at the peak of the dielectric permittivity, C2 is the Curie constant, and g is the diffuseness degree, ranging from 1 for a normal ferroelectric to 2 for an ideal relaxor ferroelectric. To confirm the DPT behavior of the BT–9%AlN and BT–14%AlN ceramics, the plots of ln(1/e 1/emax) vs. ln(T Tmax) at 1 kHz were also plotted, as shown in the insets of Fig. 5. The g values were determined from the slopes of the fitting lines. It was found that the g values increased from approximately 1 (x o 3%) to 1.6 (x 4 3%) with increasing AlN content. This result confirms that the BT–xAlN composite ceramics exhibit relaxorlike behavior when x 4 3%. In the present work, however, the ceramics with compositions of x Z 9% are still normal ferroelectrics
Phys. Chem. Chem. Phys., 2014, 16, 13078--13085 | 13081
View Article Online
Published on 31 March 2014. Downloaded by University of Newcastle on 12/09/2014 15:35:38.
PCCP
Paper
Fig. 4 Temperature dependence of the dielectric permittivities at 1 kHz for the BT–xAlN composite ceramics.
Fig. 5 Inverse dielectric permittivities at 1 kHz as a function of temperature for BT–xAlN composite ceramics: (a) x = 0%, (b) x = 0.75%, (c) x = 1.5%, (d) x = 3%, (e) x = 9% and (f) x = 14%. The insets are the corresponding plots of ln(1/e 1/emax) vs. ln(T Tmax) at 1 kHz.
rather than relaxors because many nonrelaxor ferroelectrics also have a DPT.36 The broadening of the permittivity peaks can be related to a non-homogeneous dopant distribution and the higher internal stresses formed from the non-ferroelectric BaAl2O4 grain boundaries in the fine-grained BT–xAlN composite ceramics.37 Fig. 6a shows the room temperature P–E hysteresis loops of BT–xAlN composite ceramics measured at different electric fields. A continuous shape change of the loops is observed with increasing AlN content. The ceramics with the compositions of x = 0%, x = 0.75% and x = 1.5% exhibit the wellsaturated and square-like P–E hysteresis loops at an electric field
13082 | Phys. Chem. Chem. Phys., 2014, 16, 13078--13085
of 40 kV cm1, which are subsequently replaced by the slanted and flat P–E loops with suppressed remnant polarizations (Pr) with increasing AlN content. Fig. 6b shows the 2Pr values of BT–xAlN composite ceramics with different AlN contents. As the AlN content increases, the 2Pr reaches a maximum value of 28.9 mC cm2 for the BT–0.75%AlN ceramic, but shows an obvious decrease with further increases in the AlN content. 3.3.
Piezoelectric and mechanical properties
Fig. 7a shows the variations of the piezoelectric constants (d33) and Vickers hardnesses (Hv) of the BT–xAlN composite ceramics as a function of AlN content. The d33 first increases, reaches an
This journal is © the Owner Societies 2014
View Article Online
Published on 31 March 2014. Downloaded by University of Newcastle on 12/09/2014 15:35:38.
Paper
Fig. 6 (a) Room temperature P–E hysteresis loops for the BT–xAlN composite ceramics, and (b) 2Pr values with different AlN contents.
optimum value of 305 pC N1 at x = 1.5%, and then begins to decrease with the further addition of AlN. According to previous reports,38–40 ceramics with finer grains tend to have smaller domain sizes and higher d33 values, due to the easier motion of the ferroelectric domain walls with small areas when exposed to external electrical or stress signals. The biggest problem in our case, though, is that the ceramics with smaller grain sizes (E6.2 mm for BT–9%AlN ceramic and E5.6 mm for BT–14%AlN ceramic) are unable to obtain the highest d33 values. In that way, the high piezoelectric theory, which originates from the domain migration for fine grains, cannot explain the fact that the BT–1.5%AlN ceramic has the highest d33 value instead of its counterparts with smaller grain sizes and nearly the same densities (Fig. 3). Therefore, there must be something else contributing to the enhancement of their piezoelectric properties. As pointed out in the XRD analysis, the Al3+–N3 pairs will align along the direction vertical to the c axis when x r 1.5%, thus leading to the occurrence of a coordinated lattice distortion. According to the symmetry-conforming property of defect dipoles (PD),41,42 the BT crystal symmetry tends to follow the local symmetry of the short-range order distribution of the Al3+–N3 pairs when in equilibrium, with respect to the early formation of the Al3+–N3 pairs during sintering. The local symmetry of the Al3+–N3 pairs will therefore prompt an unstable spontaneous polarization (PS) reorientation inside the tetragonal 901 domains by shortrange interactions under the effect of the poling electric field.
This journal is © the Owner Societies 2014
PCCP
Fig. 7 (a) Piezoelectric constants (d33) and Vickers hardnesses (Hv) of the BT–xAlN composite ceramics as a function of AlN content, and (b) Vickers hardnesses (Hv) as a function of the inverse square root of the average grain sizes (d).
In other words, the ordered PD (Al3+–N3 pairs) facilitates the switching of 901 domains and causes the rotation of PS from the [001] direction to the [100] direction during the poling process. This is clearly the case for the BT–1.5%AlN ceramic. Although this material does not have exactly the same MPB (as in the PZT and BZT–BCT systems) and PPT (as in the KNN system), the high d33 of 305 pC N1 is obtained due to the response of PS to the local symmetry of the ordered PD during the poling process. Whereas the lowering-local symmetry of the Al3+–N3 pairs, which is caused by the increase in the number of differently oriented neighbouring Al3+–N3 pairs, as well as the presence of the second phase, BaAl2O4, results in a clear downward trend of d33 when x 4 1.5%, as shown in Fig. 7a. On the other hand, the Vickers hardnesses (Hv) increase dramatically with AlN content (Fig. 7a). It is worth noting that the Hv value is about 5.90 GPa for the BT–1.5%AlN ceramic with a maximum piezoelectric constant of d33 = 305 pC N1, obviously higher than the pure BT ceramic with Hv E 5.25 GPa. This result is also comparable to the hardness values of BT-based composites prepared by pressureless sintering25 and two step sintering.21 Considering the existing strengthening models, the following explanations are presented. Firstly, the improved density (Fig. 3) is helpful for enhancing the hardness of the BT–xAlN composite ceramics. Secondly, when x 4 1.5%, the presence of the second phase, BaAl2O4 (Fig. 1a), namely,
Phys. Chem. Chem. Phys., 2014, 16, 13078--13085 | 13083
View Article Online
Published on 31 March 2014. Downloaded by University of Newcastle on 12/09/2014 15:35:38.
PCCP
Paper
content. Without lowering the Curie temperature, the BT–1.5%AlN ceramic sintered at 1325 1C exhibited a high piezoelectric coefficient of 305 pC N1 and an improved Vickers hardness of 5.9 GPa. The enhanced piezoelectricity is from the interactions between defect dipoles and spontaneous polarization inside the domains. The hardened material is attributed to the improved densities and particle and grain boundary strengthening of the BaTiO3–xAlN composite ceramics. Our work indicates that piezoelectric and mechanical properties can be simultaneously enhanced by adding appropriate AlN content.
Acknowledgements Fig. 8 Temperature dependence of the d33 values of the poled BT–xAlN composite ceramics measured ex situ.
This work was financially supported by the National Nature Science Foundation of China (Grant No. 11272102).
non-primary particles precipitated from a supersaturated solid solution, may be partly associated with a strengthening contribution as well. More importantly, as revealed in Fig. 7b, the Hall–Petch relationship (Hv = Hv0 + kd1/2; where Hv0 and k are materials constants) is confirmed over the entire grain size range investigated. Classically, high values of Hv are considered to be related to the effect of increased grain boundaries,43–46 which are found in samples with smaller grain sizes, providing additional obstacles for the movement of lattice dislocations in the adjacent grains, thus leading to harder materials. In summary, the grain size reduction with increasing AlN content (Fig. 3), observed in our BT–xAlN composite ceramics, is believed to be the decisive factor influencing the hardening of the BT–xAlN composite ceramics. Our work illustrates that, when an appropriate AlN content is added to BT ceramic, the piezoelectric constant (d33) and Vickers hardness (Hv) can be improved simultaneously.
References
3.4.
Temperature dependence of the piezoelectric constant
Fig. 8 shows the temperature dependence of the d33 values of poled BT–xAlN composite ceramics measured ex situ. The temperature dependence of d33 was measured at room temperature after annealing the poled samples at various temperatures for 45 min. The depolarization temperature (Td) is defined as the point where d33 decreases sharply. As the temperature increases, the d33 value of the pure BT ceramic decreases remarkably above 60 1C due to thermal depoling, but in contrast the d33 values of the AlN-doped BT ceramics remain almost unchanged and then drop dramatically above the critical temperature (Td) of approximately 115 1C. The improved piezoelectric temperature stability of the AlN-doped BT ceramics can be attributed to the formation of Al3+–N3 pairs, which are insensitive to the temperature variations and phase transitions.
4. Conclusions The effects of the AlN content on the phase, density, grain size, and electrical and mechanical properties of BT ceramics were investigated and analyzed. The densities of the BT–xAlN composite ceramics were improved by adding appropriate AlN
13084 | Phys. Chem. Chem. Phys., 2014, 16, 13078--13085
¨del, S. Kalinin and N. Balke, Adv. 1 R. Dittmer, W. Jo, J. Ro Funct. Mater., 2012, 22, 4208–4215. 2 C. R. Zhou, A. Feteira, X. Shan, H. B. Yang, Q. Zhou, J. Cheng, W. Z. Li and H. Wang, Appl. Phys. Lett., 2012, 101, 032901. 3 S. Gupta and S. Priya, Appl. Phys. Lett., 2013, 102, 012906. 4 P. Zheng, J. L. Zhang, Y. Q. Tan and C. L. Wang, Acta Mater., 2012, 60, 5022–5030. 5 S. T. Zhang, B. Yang and W. W. Cao, Acta Mater., 2012, 60, 469–475. 6 Z. B. Tian, X. H. Wang, Y. C. Zhang, J. Fang, T. H. Song, K. H. Hur, S. Lee and L. T. Li, J. Am. Ceram. Soc., 2010, 93, 171–175. 7 H. R. Xu and L. Gao, J. Am. Ceram. Soc., 2003, 86, 203–205. 8 T. Karaki, K. Yan, T. Miyamoto and M. Adachi, Jpn. J. Appl. Phys., 2007, 46, L97. 9 Z. Y. Shen and J. F. Li, J. Ceram. Soc. Jpn., 2010, 118, 940–943. 10 H. Takahashi, Y. Numamoto, J. Tani and S. Tsurekawa, Jpn. J. Appl. Phys., 2008, 47, 8468. 11 S. M. Moon, X. Wang and N. H. Cho, J. Ceram. Soc. Jpn., 2009, 117, 729. 12 S. Wada, K. Takeda, T. Muraishi, H. Kakemoto, T. Tsurumi and T. Kimura, Jpn. J. Appl. Phys., 2007, 46, 8039. 13 H. Takahashi, Y. Numamoto, J. Tani and S. Tsurekawa, Jpn. J. Appl. Phys., 2007, 46, 7044. 14 H. Takahashi, Y. Numamoto, J. Tani and S. Tsurekawa, Jpn. J. Appl. Phys., 2006, 45, 7405. 15 W. W. Cao and C. A. Randall, J. Phys. Chem. Solids, 1996, 57, 1499–1505. 16 S. F. Shao, J. L. Zhang, Z. Zhang, P. Zheng, M. L. Zhao, J. C. Li and C. L. Wang, J. Phys. D: Appl. Phys., 2008, 41, 125408. 17 W. F. Liu and X. B. Ren, Phys. Rev. Lett., 2009, 103, 257602. 18 R. Z. Zuo, J. Fu, D. Y. Lv and Y. Liu, J. Am. Ceram. Soc., 2010, 93, 2783–2787. 19 Y. Saito, H. Takao, T. Tani, T. Nonoyama, K. Tajatiri, T. Homma, T. Nagaya and M. Nakamura, Nature, 2004, 432, 84–87.
This journal is © the Owner Societies 2014
View Article Online
Published on 31 March 2014. Downloaded by University of Newcastle on 12/09/2014 15:35:38.
Paper
20 S. Jiansirisomboon and A. Watcharapasorn, Curr. Appl. Phys., 2008, 8, 48–52. 21 C. Kruea-in, S. Eitssayeam, K. Pengpat, T. Tunkasiri and G. Rujijanagul, Ferroelectrics, 2011, 415, 127–134. 22 H. J. Hwang, M. Toriyama, T. Sekino and K. Niihara, J. Eur. Ceram. Soc., 1998, 18, 2193. 23 H. J. Hwang, T. Sekino, K. Ota and K. Niihara, J. Mater. Sci., 1996, 31, 4617–4624. 24 S. Panteny, C. R. Bowen and R. Stevens, J. Mater. Sci., 2006, 41, 3837–3843. 25 T. Kojima, M. Sugihara, Y. Hosoi, N. Uekawa and K. Kakegawa, J. Ceram. Soc. Jpn., 2009, 117, 1328–1332. 26 H. Hyuga, Y. Hayashi, T. Sekino and K. Niihara, Nanostruct. Mater., 1997, 9, 547–550. 27 C. Ang, Z. Yu, Z. Jing, R. Y. Guo, A. S. Bhalla and L. E. Cross, Appl. Phys. Lett., 2002, 80, 3424. 28 S. Sharma, P. Kumar and P. Palei, Ceram. Int., 2012, 38, 5597. 29 W. G. Yang, B. P. Zhang, N. Ma and L. Zhao, J. Eur. Ceram. Soc., 2012, 32, 899. 30 R. L. Brutchey, G. S. Cheng, Q. Gu and D. E. Morse, Adv. Mater., 2008, 20, 1029–1033. 31 S. Tangjuank and T. Tunkasiri, Appl. Phys. Lett., 2007, 90, 072908. 32 R. D. Shannon, Acta Crystallogr., 1976, A32, 751. 33 Y. M. Wang, X-ray diffraction of noncrystal and crystal with defects, Science Press, Beijing, 1st edn, 1988, in Chinese.
This journal is © the Owner Societies 2014
PCCP
34 H. L. Cheng, H. L. Du, W. C. Zhou, D. M. Zhu, F. Luo and B. X. Xu, J. Am. Ceram. Soc., 2013, 96, 833–837. 35 V. V. Shvartsman and D. C. Lupascu, J. Am. Ceram. Soc., 2012, 95, 1–26. 36 R. Z. Zuo, J. Fu, S. B. Lu and Z. K. Xu, J. Am. Ceram. Soc., 2011, 94, 4352–4357. 37 Z. H. Yao, H. X. Liu, Y. Liu, Z. H. Wu, Z. Y. Shen, Y. Liu and M. H. Cao, Mater. Chem. Phys., 2008, 109, 475–481. 38 G. Arlt, D. Hennings and G. D. With, J. Appl. Phys., 1985, 58, 1619–1625. 39 Z. Zhao, V. Buscaglia, M. Viviani, M. T. Buscaglia, L. Mitoseriu, A. Testino, M. Nygren, M. Johnsson and P. Nanni, Phys. Rev. B: Condens. Matter Mater. Phys., 2004, 70, 024107. 40 H. T. Zhang, X. Y. Deng, T. Li, W. Zhang, R. Chen, W. W. Tian, J. B. Li, X. H. Wang and L. T. Li, Appl. Phys. Lett., 2010, 97, 162913. 41 K. Wang and J. F. Li, Adv. Funct. Mater., 2010, 20, 1924–1929. 42 X. B. Ren, Nat. Mater., 2004, 3, 91–94. 43 H. Hahn and K. A. Padmanabhan, Nanostruct. Mater., 1995, 6, 191–200. 44 K. L. Kendig and D. B. Miracle, Acta Mater., 2002, 50, 4165–4175. 45 M. A. Meyers, A. Mishra and D. J. Benson, Prog. Mater. Sci., 2006, 51, 427–556. 46 C. S. Pande and K. P. Cooper, Prog. Mater. Sci., 2009, 54, 689–706.
Phys. Chem. Chem. Phys., 2014, 16, 13078--13085 | 13085
Published on 31 March 2014. Downloaded by University of Newcastle on 12/09/2014 15:35:38.
PAPER
Cite this: Phys. Chem. Chem. Phys., 2014, 16, 13078
View Journal | View Issue
Enhanced piezoelectric and mechanical properties of AlN-modified BaTiO3 composite ceramics Dan Xu,ab Lidong Wang,a Weili Li,ac Wei Wang,a Yafei Hou,a Wenping Cao,a Yu Fenga and Weidong Fei*a BaTiO3–xAlN (BT–xAlN) composite ceramics were prepared by conventional solid state reaction sintering. The effects of the AlN content on the crystalline structures, densities, and electrical and mechanical properties of the BT ceramics were investigated. The BT–1.5%AlN ceramic exhibits a good piezoelectric constant of 305 pC N1 and an improved Vickers hardness of 5.9 GPa. The enhanced piezoelectricity originates from interactions between defect dipoles and spontaneous polarization inside the domains due to the occurrence of local symmetry, caused by the preferential distribution of the Al3+–N3 pairs vertical
Received 24th February 2014, Accepted 30th March 2014
to the c axis. The hardening of the material is attributed to the improved density, and particle and grain
DOI: 10.1039/c4cp00796d
boundary strengthening. Our work indicates that if a suitable doping ion pair is designed, lead-free
www.rsc.org/pccp
ceramic systems prepared from ordinary raw materials by a conventional sintering method have a high probability of exhibiting good piezoelectric and mechanical properties simultaneously.
1. Introduction In recent years, considering the effects of lead on human health and environmental protection, lead-free piezoelectric ceramics have been extensively investigated in order to find a replacement for the toxic lead zirconate titanate (PZT).1–5 Among the candidates is barium titanate (BaTiO3, BT), with an ABO3-type perovskite structure, it was one of the most widely used ferroelectric materials before the discovery of high-performance PZT, which has been used in a variety of applications such as PTC thermistors, multi-layer ceramic capacitors (MLCC), ferroelectric random access memories (FRAM), and optical devices, because of its outstanding dielectric and ferroelectric properties.6–8 Nowadays, however, the practical applications of BT ceramics are still subject to their relatively poor piezoelectric properties in comparison with those of PZT. BT ceramics with high piezoelectric properties are hard to acquire just by using ordinary raw materials and a conventional sintering method. BT ceramics are therefore used mainly as dielectric materials rather than as piezoelectric materials, unless employing special fine powders8 and unusual sintering techniques9–12 to enhance their piezoelectric properties. For example, especially high d33 values have been reported for those BT ceramics prepared by
a
School of Materials Science and Engineering, Harbin Institute of Technology, Harbin 150001, People’s Republic of China. E-mail: [email protected]; Fax: +86-451-86413908; Tel: +86-451-86413908 b Jilin Institute of Chemical Technology, Jilin 132022, People’s Republic of China c National Key Laboratory of Science and Technology on Precision Heat Processing of Metals, Harbin Institute of Technology, Harbin 150001, People’s Republic of China
13078 | Phys. Chem. Chem. Phys., 2014, 16, 13078--13085
special fabrication methods such as spark plasma sintering (416 pC N1),9 microwave sintering (370 pC N1)10 and two-step sintering (460 pC N1).11 The origin of the high piezoelectric properties for these special BT ceramics is still controversial.9,13–16 Two crucial factors, fine grains (several micrometers) and nano-domain structures (tens of nanometers), are believed to be responsible for the high d33 values observed in these BT ceramics. In this case, the existing problem is how to obtain high piezoelectric properties for BT and other ABO3-type ferroelectric ceramics using a traditional preparation method. Up to now, the most basic approach available for achieving high piezoelectricities of piezoelectric ceramics, excluding the fine-grain effect, is to induce a phase transition between two ferroelectric phases either by the morphotropic phase boundary (MPB) effect or by the polymorphic phase transition (PPT) effect. As an example, Ren et al. reported that the Ba(Ti0.8Zr0.2)O3– x(Ba0.7Ca0.3)TiO3 (BZT–xBCT) system exhibited a very high piezoelectric coefficient (d33) of E620 pC N1 at the MPB starting from a tricritical triple point (TCP) of a cubic paraelectric phase (C), a ferroelectric rhombohedral (R) phase, and a tetragonal (T) phase.17 Unfortunately, the usefulness of the BZT–xBCT system is still restricted due to its relatively low Curie temperatures (TC). Another high piezoelectric mechanism is the PPT mainly found in the (K, Na)NbO3 (KNN) system.18,19 The high levels of piezoelectricity reported in the KNN system are the result of the enhanced polarizability associated with the orthorhombic (O)–tetragonal (T) polymorphic phase transition being compositionally shifted downward to near room temperature. Such a temperature-driven ferroelectric–ferroelectric phase transition may facilitate polarization rotations. Although a high d33 of
This journal is © the Owner Societies 2014
View Article Online
Published on 31 March 2014. Downloaded by University of Newcastle on 12/09/2014 15:35:38.
Paper
E300 pC N1 is achieved for non-textured KNN-based ceramics, the inherently large temperature dependence of their piezoelectric properties becomes the biggest problem for the stability and reliability of the devices operating at high temperatures. On the other hand, although relatively high piezoelectric constants (d33 = 200–300 pC N1) have been realized, the ordinary BT-based piezoelectric ceramics still suffer from low mechanical strength and poor reliability under loading conditions, which limits their applications as transducers, sensors and actuators where high fracture toughness and fracture strength are required. Many studies on BT-based composites incorporated nano-particulates, including Al2O3,20 ZrO2,21 MoO3,22 NiO,22 SiC,23 Ag24,25 and Ni,26 in order to improve their mechanical properties. It was found that both oxide and metal additives could improve the mechanical properties of the BT-based composite ceramics. However, there are only a few studies on their electrical properties and in particular their piezoelectric properties.20–26 T. Kojima et al.25 reported that the Vickers hardness (Hv) of the BT–Ag composite prepared by pressureless sintering was 6.2 GPa, much higher than the monolithic BT ceramic with Hv = 2.5 GPa. BaZr0.07Ti0.93O3 ceramics, synthesized using a modified two step sintering process, exhibited an enhanced Knoop hardness (Hk) of 5.5 GPa, and the dwell time was up to 4 h.21 S. Jiansirisomboon et al.20 found that the Knoop hardness of BT/xAl2O3 nanocomposite ceramics increased with increasing Al2O3 content, and that the ceramic with optimum properties (Hk = 4.4 GPa, d33 = 178 pC N1) was obtained at x = 0.5 vol%. Obviously, it is not easy to obtain a BT ceramic with simultaneously good piezoelectric and mechanical properties. BT composite ceramics will sacrifice their piezoelectric properties for the enhancement of mechanical properties in most cases. Thus there is an urgent need to develop BT ceramics with enhanced piezoelectric and mechanical properties. As is well known, the microstructures and physical properties of BT ceramics can be modified by appropriate substitutions at A- and/or B-sites.27–31 However, previous works have seldom involved A- and O-site replacements. Therefore, in this paper, the non-oxide compound, AlN, is added to the BT matrix. The electrical and mechanical properties of BT–xAlN composite ceramics, prepared from ordinary raw materials by a conventional solid state reaction sintering method, were investigated and analyzed. It is expected that the addition of AlN to the BT would be a good choice to enhance its mechanical properties without deteriorating its electrical properties. The reinforcement mechanisms of AlN on the piezoelectric and mechanical properties of BT composite ceramics are discussed as well.
2. Experimental BaTiO3–xAlN composite ceramics (abbreviated as BT–xAlN, x = 0–14 mol%) were prepared by conventional solid state reaction sintering. Commercially available BT (99.9%, Aladdin Chemistry Co. Ltd, Shanghai, China) and AlN (99.9%, Eno High-Tech Material Development Co., Ltd, Qinhuangdao,
This journal is © the Owner Societies 2014
PCCP
China) powders were used as the starting materials. The BT and AlN powders were weighed in the desired compositions and ball-milled for 8 h by planetary milling with zirconia balls in alcohol. After ball-milling, the slurries were dried at 80 1C for 12 h, ground and sieved. The mixed powders were subsequently pressed into pellets of 10 mm in diameter and 1 mm in thickness using a few drops of 5 wt% polyvinyl alcohol (PVA) as a binder. After burning off the PVA, the BT–xAlN pellets embedded in the BT powders were sintered in covered alumina crucibles at 1325 1C for 2 h with a heating rate of 5 1C min1. The densities were measured by the Archimedes method. The crystalline structure of each sample was analyzed by X-ray diffraction (XRD) on a Philips X’Pert diffractometer with Cu Ka radiation, operating at 40 kV and 40 mA. The y–2y scan was performed at room temperature with a step size of 0.021 and time per step of 2 s. The microstructures of the sintered samples were examined by scanning electron microscopy (SEM) using a Quanta 200F. The temperature dependence of the dielectric constants was measured at 1 kHz in the temperature range of 25–200 1C on a LCR meter (Agilent, E4980A). Polarization versus electric field (P–E) hysteresis loops were determined by a Radiant Technologies Precision workstation. The samples were poled at 80 1C in a silicone oil bath under a DC field of 3 kV mm1 for 30 min. Piezoelectric constants (d33) were measured on a quasi-static piezoelectric constant testing meter (ZJ-4AN, Institute of Acoustics, Chinese Academy of Science). The well-polished ceramics were subjected to a microhardness tester (HV-1000, Cany Precision Instrument Co., Shanghai, China) for Vickers hardness (Hv) determination using a 300 g load with a holding time of 15 s for hardness measurements.
3. Results and discussion 3.1.
Phase and microstructure analysis
The room temperature XRD patterns of the BT–xAlN (0 r x r 14 mol%) ceramic powders in the 2y range of 20–851 are shown in Fig. 1a. A pure perovskite phase is formed up to x r 1.5%, while weak second phases around 2y = 28.2–28.81 are observed when x Z 3%, which are identified as BaAl2O4 (PDF No. 17-0306), and are marked with a shaded rectangle in the XRD patterns (Fig. 1a). The solubility limit of Al in the BT lattice is between x = 1.5–3%. The fine scan XRD patterns for the BT–xAlN ceramic powders in the 2y ranges of 64.5–67.01 and 83.0–84.01 are shown in Fig. 1b and c. All of the ceramic powders appear to adopt a tetragonal crystal symmetry (PDF No. 83-1880), as evidenced by the splitting of the 202/220 peaks at 2y E 661 (Fig. 1b) and the existence of a single 222 peak at 2y E 83.51 (Fig. 1c). The slightly reduced 222 interplanar spacing (d222) with increasing AlN content, as revealed in Fig. 1d, further affirms that the BT has been doped with AlN. Considering the very low solubility limit of Al in BT and the smaller ionic radius of Al3+ compared with those of Ba2+ and Ti4+,32 it is reasonable that the 222 peak position shifts slightly towards right with increasing AlN content (Fig. 1c). However, it is
Phys. Chem. Chem. Phys., 2014, 16, 13078--13085 | 13079
View Article Online
Published on 31 March 2014. Downloaded by University of Newcastle on 12/09/2014 15:35:38.
PCCP
Paper
Fig. 1 (a) XRD patterns of BT–xAlN ceramic powders measured at room temperature. Fine scan XRD patterns in the 2y ranges of (b) 64.5–67.01 and (c) 83.0–84.01. (d) 222 interplanar spacing (d222) with increasing AlN content. (e) Variation of the relative intensities of the 002/200 peaks.
still unclear how the Al3+ and N3 ions arrange their locations in the BT lattice. In order to study the distribution of the Al3+ and N3 ions in the BT lattice, the variation of the relative intensities of the 002/200 peaks (I002/I200) was plotted, as shown in Fig. 1e. It was found that the I002/I200 ratio first increased, achieved a maximum value at x = 1.5%, and subsequently decreased with increasing AlN content. It must be noted that the variation of the relative intensities of the 002/200 peaks with AlN content cannot be explained just by a random distribution of Al3+ and N3 ions. If such a random distribution was true, then the XRD results of Fig. 1e would not be observed as a consequence of the same influence on the structure factor (F) for the 002 and 200 diffractions in this case. According to the ionic radii and valences, we believe that the Al3+ and N3 ions should form Al3+–N3 pairs in the same cell in order to minimize the electrostatic energy. Therefore, the most energy-favored way of replacement is to exchange Al3+ for Ba2+ sites, and N3 for O2 sites, accompanied by the maintenance of the local charge neutrality. Thus, there exist two possible distributions for the Al3+–N3 pairs. The first is that the distribution direction of the Al3+–N3 pairs is perpendicular to the c axis, and the second is that the Li+–Al3+ pairs arrange their locations in other directions. Due to the different ionic radii among the Al3+–N3 pairs, and the Ba2+ and O2 ions, the occurrence of Al3+–N3 pairs will induce lattice distortions in the BT, leading to a weakened XRD intensity.33 In the case of the (002), (020) and (200) planes, the
13080 | Phys. Chem. Chem. Phys., 2014, 16, 13078--13085
eff eff effective structure factors, F eff 002, F 020, and F 200, adjusted by lattice distortions, can be expressed as follows 8p2 sin2 y 2 eff F200 ¼ F200 expðM1 Þ ¼ F200 exp Dx (1) l2
8p2 sin2 y 2 eff ¼ F020 expðM2 Þ ¼ F020 exp Dy F020 l2
(2)
8p2 sin2 y 2 eff ¼ F002 expðM3 Þ ¼ F002 exp Dz F002 l2
(3)
where F% 200, F% 020 and F% 002 are the average structure factors of the distortion-free crystal lattices, and Dx2, Dy2 and Dz2, which represent deviations from the ideal BT lattice sites, are the mean square displacements in the [100], [010] and [001] directions, respectively. Since the diffraction intensity (I) is proportional to the structure factor’s square modulus (|F|2), the lattice distortion caused by the replacement of Al3+–N3 pairs, which primarily occurs along the Al3+–N3 lines, is an important factor influencing the XRD intensity. Based on the above analysis, the observed variations in the I002/I200 intensity ratio shown in Fig. 1e can be explained as follows; when x r 1.5%, the Al3+–N3 pairs tend to adopt a preferential distribution vertical to the c axis and hence, the lattice distortion induced by the Al3+–N3 pairs will generate little influence on the M3 value of eqn (3), because the lattice distortion primarily occurs along the Al3+–N3 lines. However, as the AlN content further
This journal is © the Owner Societies 2014
View Article Online
Published on 31 March 2014. Downloaded by University of Newcastle on 12/09/2014 15:35:38.
Paper
Fig. 2
PCCP
Surface SEM micrographs of BT–xAlN ceramics, (a) x = 0%, (b) x = 0.75%, (c) x = 1.5%, (d) x = 3%, (e) x = 9% and (f) x = 14%, sintered at 1325 1C for 2 h.
increases (x 4 1.5%), the I002/I200 intensity ratio becomes weakened, indicating that the number of Al3+–N3 pairs vertical to the c axis begins to decrease due to the presence of the second phase, BaAl2O4, and the elastic energy limitation. The surface SEM micrographs of the BT–xAlN ceramics with different AlN contents are presented in Fig. 2. All of the ceramic samples are very dense without clear pores on the ceramic surfaces. As shown in Fig. 2, the abnormal growth of grains can be found in the pure BT ceramic, and this case was modulated by incorporating a small amount of AlN (Fig. 2b, c) to the BT matrix. Fig. 3 shows the variations of the bulk densities and the average grain sizes of the BT–xAlN ceramics with different AlN content. It is evident that the density is improved to some extent by introducing AlN. Furthermore, the average grain size decreases with increasing AlN content, indicating that the addition of AlN can effectively limit the grain growth of BT. 3.2. Dielectric ‘‘relaxorlike’’ behavior and ferroelectric properties The temperature-dependent dielectric constants of BT–xAlN composite ceramics with compositions of x = 0, 0.75, 1.5, 3, 9 and 14 mol% measured at 1 kHz are displayed in Fig. 4.
Fig. 3
Density (r) as a function of average grain size (d).
This journal is © the Owner Societies 2014
As revealed in Fig. 4, there is only one dielectric anomaly around 120 1C–130 1C, characterized by a phase transition from the ferroelectric phase (tetragonal phase, T) to the paraelectric phase (cubic phase, C) within the measured temperature range. The e peak associated with the T–C phase transformation is much sharper for the coarse-grained samples and becomes more diffuse with reducing grain size (as x Z 9%). This result indicates that the BT–9%AlN and BT–14%AlN ceramics show a relaxorlike characteristic due to the occurrence of the diffuse phase transition (DPT). It is well known that for a normal ferroelectric, the dielectric permittivity above TC follows the Curie–Weiss law, 1 T TC ¼ ; e C1
T 4 TC
(4)
where C1 is the Curie–Weiss constant. The inverse dielectric permittivity measured at 1 kHz was plotted as a function of temperature, as shown in Fig. 5. For a relaxor ferroelectric with a DPT, the degree of diffuseness of the paraelectric–ferroelectric phase transition can be more effectively described by a modified Curie–Weiss law,34,35 1 1 ðT Tm Þg ¼ ; C2 e em
1g2
(5)
where e is the dielectric permittivity at temperature T, em is the maximum value of the dielectric permittivity, Tm is the temperature at the peak of the dielectric permittivity, C2 is the Curie constant, and g is the diffuseness degree, ranging from 1 for a normal ferroelectric to 2 for an ideal relaxor ferroelectric. To confirm the DPT behavior of the BT–9%AlN and BT–14%AlN ceramics, the plots of ln(1/e 1/emax) vs. ln(T Tmax) at 1 kHz were also plotted, as shown in the insets of Fig. 5. The g values were determined from the slopes of the fitting lines. It was found that the g values increased from approximately 1 (x o 3%) to 1.6 (x 4 3%) with increasing AlN content. This result confirms that the BT–xAlN composite ceramics exhibit relaxorlike behavior when x 4 3%. In the present work, however, the ceramics with compositions of x Z 9% are still normal ferroelectrics
Phys. Chem. Chem. Phys., 2014, 16, 13078--13085 | 13081
View Article Online
Published on 31 March 2014. Downloaded by University of Newcastle on 12/09/2014 15:35:38.
PCCP
Paper
Fig. 4 Temperature dependence of the dielectric permittivities at 1 kHz for the BT–xAlN composite ceramics.
Fig. 5 Inverse dielectric permittivities at 1 kHz as a function of temperature for BT–xAlN composite ceramics: (a) x = 0%, (b) x = 0.75%, (c) x = 1.5%, (d) x = 3%, (e) x = 9% and (f) x = 14%. The insets are the corresponding plots of ln(1/e 1/emax) vs. ln(T Tmax) at 1 kHz.
rather than relaxors because many nonrelaxor ferroelectrics also have a DPT.36 The broadening of the permittivity peaks can be related to a non-homogeneous dopant distribution and the higher internal stresses formed from the non-ferroelectric BaAl2O4 grain boundaries in the fine-grained BT–xAlN composite ceramics.37 Fig. 6a shows the room temperature P–E hysteresis loops of BT–xAlN composite ceramics measured at different electric fields. A continuous shape change of the loops is observed with increasing AlN content. The ceramics with the compositions of x = 0%, x = 0.75% and x = 1.5% exhibit the wellsaturated and square-like P–E hysteresis loops at an electric field
13082 | Phys. Chem. Chem. Phys., 2014, 16, 13078--13085
of 40 kV cm1, which are subsequently replaced by the slanted and flat P–E loops with suppressed remnant polarizations (Pr) with increasing AlN content. Fig. 6b shows the 2Pr values of BT–xAlN composite ceramics with different AlN contents. As the AlN content increases, the 2Pr reaches a maximum value of 28.9 mC cm2 for the BT–0.75%AlN ceramic, but shows an obvious decrease with further increases in the AlN content. 3.3.
Piezoelectric and mechanical properties
Fig. 7a shows the variations of the piezoelectric constants (d33) and Vickers hardnesses (Hv) of the BT–xAlN composite ceramics as a function of AlN content. The d33 first increases, reaches an
This journal is © the Owner Societies 2014
View Article Online
Published on 31 March 2014. Downloaded by University of Newcastle on 12/09/2014 15:35:38.
Paper
Fig. 6 (a) Room temperature P–E hysteresis loops for the BT–xAlN composite ceramics, and (b) 2Pr values with different AlN contents.
optimum value of 305 pC N1 at x = 1.5%, and then begins to decrease with the further addition of AlN. According to previous reports,38–40 ceramics with finer grains tend to have smaller domain sizes and higher d33 values, due to the easier motion of the ferroelectric domain walls with small areas when exposed to external electrical or stress signals. The biggest problem in our case, though, is that the ceramics with smaller grain sizes (E6.2 mm for BT–9%AlN ceramic and E5.6 mm for BT–14%AlN ceramic) are unable to obtain the highest d33 values. In that way, the high piezoelectric theory, which originates from the domain migration for fine grains, cannot explain the fact that the BT–1.5%AlN ceramic has the highest d33 value instead of its counterparts with smaller grain sizes and nearly the same densities (Fig. 3). Therefore, there must be something else contributing to the enhancement of their piezoelectric properties. As pointed out in the XRD analysis, the Al3+–N3 pairs will align along the direction vertical to the c axis when x r 1.5%, thus leading to the occurrence of a coordinated lattice distortion. According to the symmetry-conforming property of defect dipoles (PD),41,42 the BT crystal symmetry tends to follow the local symmetry of the short-range order distribution of the Al3+–N3 pairs when in equilibrium, with respect to the early formation of the Al3+–N3 pairs during sintering. The local symmetry of the Al3+–N3 pairs will therefore prompt an unstable spontaneous polarization (PS) reorientation inside the tetragonal 901 domains by shortrange interactions under the effect of the poling electric field.
This journal is © the Owner Societies 2014
PCCP
Fig. 7 (a) Piezoelectric constants (d33) and Vickers hardnesses (Hv) of the BT–xAlN composite ceramics as a function of AlN content, and (b) Vickers hardnesses (Hv) as a function of the inverse square root of the average grain sizes (d).
In other words, the ordered PD (Al3+–N3 pairs) facilitates the switching of 901 domains and causes the rotation of PS from the [001] direction to the [100] direction during the poling process. This is clearly the case for the BT–1.5%AlN ceramic. Although this material does not have exactly the same MPB (as in the PZT and BZT–BCT systems) and PPT (as in the KNN system), the high d33 of 305 pC N1 is obtained due to the response of PS to the local symmetry of the ordered PD during the poling process. Whereas the lowering-local symmetry of the Al3+–N3 pairs, which is caused by the increase in the number of differently oriented neighbouring Al3+–N3 pairs, as well as the presence of the second phase, BaAl2O4, results in a clear downward trend of d33 when x 4 1.5%, as shown in Fig. 7a. On the other hand, the Vickers hardnesses (Hv) increase dramatically with AlN content (Fig. 7a). It is worth noting that the Hv value is about 5.90 GPa for the BT–1.5%AlN ceramic with a maximum piezoelectric constant of d33 = 305 pC N1, obviously higher than the pure BT ceramic with Hv E 5.25 GPa. This result is also comparable to the hardness values of BT-based composites prepared by pressureless sintering25 and two step sintering.21 Considering the existing strengthening models, the following explanations are presented. Firstly, the improved density (Fig. 3) is helpful for enhancing the hardness of the BT–xAlN composite ceramics. Secondly, when x 4 1.5%, the presence of the second phase, BaAl2O4 (Fig. 1a), namely,
Phys. Chem. Chem. Phys., 2014, 16, 13078--13085 | 13083
View Article Online
Published on 31 March 2014. Downloaded by University of Newcastle on 12/09/2014 15:35:38.
PCCP
Paper
content. Without lowering the Curie temperature, the BT–1.5%AlN ceramic sintered at 1325 1C exhibited a high piezoelectric coefficient of 305 pC N1 and an improved Vickers hardness of 5.9 GPa. The enhanced piezoelectricity is from the interactions between defect dipoles and spontaneous polarization inside the domains. The hardened material is attributed to the improved densities and particle and grain boundary strengthening of the BaTiO3–xAlN composite ceramics. Our work indicates that piezoelectric and mechanical properties can be simultaneously enhanced by adding appropriate AlN content.
Acknowledgements Fig. 8 Temperature dependence of the d33 values of the poled BT–xAlN composite ceramics measured ex situ.
This work was financially supported by the National Nature Science Foundation of China (Grant No. 11272102).
non-primary particles precipitated from a supersaturated solid solution, may be partly associated with a strengthening contribution as well. More importantly, as revealed in Fig. 7b, the Hall–Petch relationship (Hv = Hv0 + kd1/2; where Hv0 and k are materials constants) is confirmed over the entire grain size range investigated. Classically, high values of Hv are considered to be related to the effect of increased grain boundaries,43–46 which are found in samples with smaller grain sizes, providing additional obstacles for the movement of lattice dislocations in the adjacent grains, thus leading to harder materials. In summary, the grain size reduction with increasing AlN content (Fig. 3), observed in our BT–xAlN composite ceramics, is believed to be the decisive factor influencing the hardening of the BT–xAlN composite ceramics. Our work illustrates that, when an appropriate AlN content is added to BT ceramic, the piezoelectric constant (d33) and Vickers hardness (Hv) can be improved simultaneously.
References
3.4.
Temperature dependence of the piezoelectric constant
Fig. 8 shows the temperature dependence of the d33 values of poled BT–xAlN composite ceramics measured ex situ. The temperature dependence of d33 was measured at room temperature after annealing the poled samples at various temperatures for 45 min. The depolarization temperature (Td) is defined as the point where d33 decreases sharply. As the temperature increases, the d33 value of the pure BT ceramic decreases remarkably above 60 1C due to thermal depoling, but in contrast the d33 values of the AlN-doped BT ceramics remain almost unchanged and then drop dramatically above the critical temperature (Td) of approximately 115 1C. The improved piezoelectric temperature stability of the AlN-doped BT ceramics can be attributed to the formation of Al3+–N3 pairs, which are insensitive to the temperature variations and phase transitions.
4. Conclusions The effects of the AlN content on the phase, density, grain size, and electrical and mechanical properties of BT ceramics were investigated and analyzed. The densities of the BT–xAlN composite ceramics were improved by adding appropriate AlN
13084 | Phys. Chem. Chem. Phys., 2014, 16, 13078--13085
¨del, S. Kalinin and N. Balke, Adv. 1 R. Dittmer, W. Jo, J. Ro Funct. Mater., 2012, 22, 4208–4215. 2 C. R. Zhou, A. Feteira, X. Shan, H. B. Yang, Q. Zhou, J. Cheng, W. Z. Li and H. Wang, Appl. Phys. Lett., 2012, 101, 032901. 3 S. Gupta and S. Priya, Appl. Phys. Lett., 2013, 102, 012906. 4 P. Zheng, J. L. Zhang, Y. Q. Tan and C. L. Wang, Acta Mater., 2012, 60, 5022–5030. 5 S. T. Zhang, B. Yang and W. W. Cao, Acta Mater., 2012, 60, 469–475. 6 Z. B. Tian, X. H. Wang, Y. C. Zhang, J. Fang, T. H. Song, K. H. Hur, S. Lee and L. T. Li, J. Am. Ceram. Soc., 2010, 93, 171–175. 7 H. R. Xu and L. Gao, J. Am. Ceram. Soc., 2003, 86, 203–205. 8 T. Karaki, K. Yan, T. Miyamoto and M. Adachi, Jpn. J. Appl. Phys., 2007, 46, L97. 9 Z. Y. Shen and J. F. Li, J. Ceram. Soc. Jpn., 2010, 118, 940–943. 10 H. Takahashi, Y. Numamoto, J. Tani and S. Tsurekawa, Jpn. J. Appl. Phys., 2008, 47, 8468. 11 S. M. Moon, X. Wang and N. H. Cho, J. Ceram. Soc. Jpn., 2009, 117, 729. 12 S. Wada, K. Takeda, T. Muraishi, H. Kakemoto, T. Tsurumi and T. Kimura, Jpn. J. Appl. Phys., 2007, 46, 8039. 13 H. Takahashi, Y. Numamoto, J. Tani and S. Tsurekawa, Jpn. J. Appl. Phys., 2007, 46, 7044. 14 H. Takahashi, Y. Numamoto, J. Tani and S. Tsurekawa, Jpn. J. Appl. Phys., 2006, 45, 7405. 15 W. W. Cao and C. A. Randall, J. Phys. Chem. Solids, 1996, 57, 1499–1505. 16 S. F. Shao, J. L. Zhang, Z. Zhang, P. Zheng, M. L. Zhao, J. C. Li and C. L. Wang, J. Phys. D: Appl. Phys., 2008, 41, 125408. 17 W. F. Liu and X. B. Ren, Phys. Rev. Lett., 2009, 103, 257602. 18 R. Z. Zuo, J. Fu, D. Y. Lv and Y. Liu, J. Am. Ceram. Soc., 2010, 93, 2783–2787. 19 Y. Saito, H. Takao, T. Tani, T. Nonoyama, K. Tajatiri, T. Homma, T. Nagaya and M. Nakamura, Nature, 2004, 432, 84–87.
This journal is © the Owner Societies 2014
View Article Online
Published on 31 March 2014. Downloaded by University of Newcastle on 12/09/2014 15:35:38.
Paper
20 S. Jiansirisomboon and A. Watcharapasorn, Curr. Appl. Phys., 2008, 8, 48–52. 21 C. Kruea-in, S. Eitssayeam, K. Pengpat, T. Tunkasiri and G. Rujijanagul, Ferroelectrics, 2011, 415, 127–134. 22 H. J. Hwang, M. Toriyama, T. Sekino and K. Niihara, J. Eur. Ceram. Soc., 1998, 18, 2193. 23 H. J. Hwang, T. Sekino, K. Ota and K. Niihara, J. Mater. Sci., 1996, 31, 4617–4624. 24 S. Panteny, C. R. Bowen and R. Stevens, J. Mater. Sci., 2006, 41, 3837–3843. 25 T. Kojima, M. Sugihara, Y. Hosoi, N. Uekawa and K. Kakegawa, J. Ceram. Soc. Jpn., 2009, 117, 1328–1332. 26 H. Hyuga, Y. Hayashi, T. Sekino and K. Niihara, Nanostruct. Mater., 1997, 9, 547–550. 27 C. Ang, Z. Yu, Z. Jing, R. Y. Guo, A. S. Bhalla and L. E. Cross, Appl. Phys. Lett., 2002, 80, 3424. 28 S. Sharma, P. Kumar and P. Palei, Ceram. Int., 2012, 38, 5597. 29 W. G. Yang, B. P. Zhang, N. Ma and L. Zhao, J. Eur. Ceram. Soc., 2012, 32, 899. 30 R. L. Brutchey, G. S. Cheng, Q. Gu and D. E. Morse, Adv. Mater., 2008, 20, 1029–1033. 31 S. Tangjuank and T. Tunkasiri, Appl. Phys. Lett., 2007, 90, 072908. 32 R. D. Shannon, Acta Crystallogr., 1976, A32, 751. 33 Y. M. Wang, X-ray diffraction of noncrystal and crystal with defects, Science Press, Beijing, 1st edn, 1988, in Chinese.
This journal is © the Owner Societies 2014
PCCP
34 H. L. Cheng, H. L. Du, W. C. Zhou, D. M. Zhu, F. Luo and B. X. Xu, J. Am. Ceram. Soc., 2013, 96, 833–837. 35 V. V. Shvartsman and D. C. Lupascu, J. Am. Ceram. Soc., 2012, 95, 1–26. 36 R. Z. Zuo, J. Fu, S. B. Lu and Z. K. Xu, J. Am. Ceram. Soc., 2011, 94, 4352–4357. 37 Z. H. Yao, H. X. Liu, Y. Liu, Z. H. Wu, Z. Y. Shen, Y. Liu and M. H. Cao, Mater. Chem. Phys., 2008, 109, 475–481. 38 G. Arlt, D. Hennings and G. D. With, J. Appl. Phys., 1985, 58, 1619–1625. 39 Z. Zhao, V. Buscaglia, M. Viviani, M. T. Buscaglia, L. Mitoseriu, A. Testino, M. Nygren, M. Johnsson and P. Nanni, Phys. Rev. B: Condens. Matter Mater. Phys., 2004, 70, 024107. 40 H. T. Zhang, X. Y. Deng, T. Li, W. Zhang, R. Chen, W. W. Tian, J. B. Li, X. H. Wang and L. T. Li, Appl. Phys. Lett., 2010, 97, 162913. 41 K. Wang and J. F. Li, Adv. Funct. Mater., 2010, 20, 1924–1929. 42 X. B. Ren, Nat. Mater., 2004, 3, 91–94. 43 H. Hahn and K. A. Padmanabhan, Nanostruct. Mater., 1995, 6, 191–200. 44 K. L. Kendig and D. B. Miracle, Acta Mater., 2002, 50, 4165–4175. 45 M. A. Meyers, A. Mishra and D. J. Benson, Prog. Mater. Sci., 2006, 51, 427–556. 46 C. S. Pande and K. P. Cooper, Prog. Mater. Sci., 2009, 54, 689–706.
Phys. Chem. Chem. Phys., 2014, 16, 13078--13085 | 13085
