- Email: [email protected]
Strong self-biased magnetoelectric charge coupling in a homogenous laminate stack for magnetic sensor
Journal of Alloys and Compounds 686 (2016) 723e726
Contents lists available at ScienceDirect
Journal of Alloys and Compounds journal homepage: http://www.elsevier.com/locate/jalcom
Strong self-biased magnetoelectric charge coupling in a homogenous laminate stack for magnetic sensor Chengpei Tang a, Caijiang Lu b, * a b
School of Engineering, Sun Yat-sen University, Guangzhou 510006, China Guizhou Electric Power Research Institute, China Southern Power Grid, Guiyang 550002, China
a r t i c l e i n f o
a b s t r a c t
Article history: Received 19 November 2015 Received in revised form 1 June 2016 Accepted 8 June 2016 Available online 11 June 2016
In this study, we report a strong self-biased magnetoelectric (ME) charge coupling in homogenous twophase magnetostrictive/piezoelectric laminate stack. The proposed ME stack Ni/PZT-stack is made up of hard-processing Nickel foils (Ni) and a tape-casting multilayer Pb(Zr, Ti)O3 (PZT) plate with high capacitance. Resonant ME couplings of Ni/PZT-stack with different thickness ratio (n) of Ni foil are investigated in detail. The experimental results show that the Ni/PZT-stack with n ¼ 0.4 has maximum zero-biased resonant ME charge coefficient (aQ,r) of 47.52 nC/Oe, which is far larger higher than that previously reported for other ME laminates. The proposed self-biased miniature ME structure may be useful for multifunctional devices such as electromagnetic energy harvesting, magnetic-to-electric generators or magnet field sensing based on a charge-detection method. © 2016 Elsevier B.V. All rights reserved.
Keywords: Self-biased Magnetoelectric Charge-coupling Laminate Sensor
1. Introduction Magnetoelectric (ME) composites consisting of magnetostrictive and piezoelectric components have larger ME effect than that of any natural signal-phase ME material by several orders of magnitude [1]. The extrinsic ME effect has been widely investigated both by theory and through experiment in various magnetostrictive and piezoelectric ME composites operated in several different modes [2e4]. It has been shown that the ME response of the laminated composites is determined by several major aspects: (a) the materials characteristics and mechanical coefficients of the constituents [2e13]; (b) the composition ratio of the piezoelectric and magnetostrictive layers [8e10]; (c) the type of boundary constituents [11]; and (d) the composition structure [4,12e15]. From previous reports, large ME charge couplings are important for developing applications, such as electromagnetic energy harvesters [16], magnetic-to-electric generators [17,18] or magnet field sensor based on a charge-detection method [19,20]. However, most of the reported ME composites in the literature has small charge induced from the ME effect due to the low dielectric capacitance [2e18]. For the magnetostrictive phase, the most severe limitation
* Corresponding author. E-mail address: [email protected] (C. Lu). http://dx.doi.org/10.1016/j.jallcom.2016.06.073 0925-8388/© 2016 Elsevier B.V. All rights reserved.
results from the dependence of the piezomagnetic coefficient dm on dc bias field (Hdc) [1e20]. To overcome these limitations arising from an external bias magnetic field, more and more researchers have focused on self-biased ME effect in the composites [21e25]. For the piezoelectric phase, the PZT have been the first choice in the design of ME laminate composite [2e4]. Unfortunately, the charge induced from the effect is small due to the low dielectric capacitance. And high cost and low Curie temperature of piezoelectric single crystal fibers and complex fabrication process present challenges in implementation at commercial scale [26,27]. As a result, for the physically interesting, technologically important and environmental perspective, it is indeed to design a novel laminate composite consisting of self-biased magnetostrictive component with simple processing and new PZT candidates with high capacitance, which should have large self-biased ME charge coupling. In this paper, we report a large self-biased ME charge coupling in a homogenous laminate stack. The laminate stack is consisting of a PZT multilayer stack as the piezoelectric component and hardprocessing Nickel (Ni) as the magnetostrictive component. By using single-phase homogenous magnetostrictive hard-processing Ni without any complicated synthesis processes, a strong self-biased effect can be obtained. And with use of the PZT multilayer stack with high capacitance, a dramatically giant ME charge coupling is observed. Such brand-new configurations yield the giant ME charge coupling at zero-biased magnetic field. And the preparation
724
C. Tang, C. Lu / Journal of Alloys and Compounds 686 (2016) 723e726
method is simple and feasible. 2. Experimental Fig. 1(a) shows the schematic illustration of tape-casting multilayer PZT plate with high capacitance. The piezoelectric multilayer stack was prepared with PZT-8 powders by a tape casting method. Then it was polarized in the thickness direction under an electric field of 3.5 kV/mm. The multilayer had a dimension of 20 3 0.6 mm3 with 15 layers and 40 mm per-layer. From Fig. 1(a), the positive and negative electrodes for each 40 mm layer of the PZT stack are in parallel, yielding a large capacitance of 236 nF for the multilayer PZT stack. The Ni foils (provided by Baoji metal Materials and Equipment Manufacturing Co., Ltd., China) was cut with the dimension of 12 6 0.1 mm3. As shown in Fig. 1(b), the multilayer ME composites Ni/PZT-stack was obtained by bonding the piezoelectric plate and Ni foils together with epoxy adhesive and pressed using a hydraulic press to make the epoxy layers as thin and perfect as possible. In experiments, a pair of neodymium permanent magnets (NdFeB) was used to provide the dc bias magnetic field (Hdc). A signal generator (Tektronix AFG3021B) provided a controllable input current to a long straight solenoid coil which was used to generate a small ac magnetic field (Hac). We measured the induced charge Q with a charge amplifier connected to an oscilloscope. Then the ME charge coefficient can be obtained, aQ ¼ vQME/vHac. 3. Results and discussion The ME coupling effect is a product property of the piezomagnetism, the piezoelectricity of the corresponding phases and their coupling, which can be characterized by aQ ¼ vQME/vH ¼ kdmdp. where k is a coupling factor between the two phases, dm and dp is the ME coefficient of the composite. Therefore, first we measured Hdc dependence of dynamic magnetostrictive coefficient (d33,m) for the hard-processing Ni foil under free conditions, as shown in Fig. 2. The vibration velocity v of the sample was measured by Doppler vibrometer (Polytec OFV-5000). Then the piezomagnetic coefficient was calculated by d33,m ¼ dl/dH ¼ v/(pGlHac). Here G is the vibration frequency, l is the length, and Hac is the external AC magnetic field. The d33,m of hard-processing Ni foil shows a hysteretic behavior during Hdc sweep (clockwise direction). Importantly, large zerobiased d33,m is observed. As Hdc increases from 600 Oe to 0 Oe, d33,m increases to a peak value of 15.1 nm/A for Hdc ¼ 51Oe. The
Fig. 2. The piezomagnetic coefficient d33,m as a function of Hdc for hard-processing Ni foil.
d33,m at Hdc ¼ 0 Oe is 13.3 nm/A. As Hdc increases from 0 Oe to 100 Oe, the d33,m decreases down to zero for Hdc ¼ 45Oe. Beyond this zero-crossing, a further decrease in Hdc results in an increase in d33,m for 45 Oe < Hdc < 140 Oe. A peak value of 13.1 nm/A is observed at Hdc ¼ 140 Oe and then decreases down for Hdc > 140 Oe. The results demonstrate that the hard-processing Ni foil with hysteretic behavior is effective to obtain self-biased effect. Fig. 3(a) show the zero-biased ME charge coefficient aQ ¼ vQME/ vH as a function of frequency around resonance. In this figure, the thickness of Ni foil is 0.1 mm, or the thickness ration n [n ¼ tm/ (tm þ tp), with tm or tp being the respective thickness of the Ni and the PZT layer, respectively] of Ni foil in ME stack is 0.143. It is clearly that the aQ reaches 41.5 nC/Oe at resonance frequency fr ¼ 91.78 kHz. For the ME composite consisting of mechanically coupled magnetostrictive and piezoelectric layers, the resonance frequency of ME composite is [28].
fr ¼ V 2l
(1)
where V ¼ 1=ðrs11 Þ1=2 (r and s11 are the average density and equivalent elastic compliance, respectively) is the average acoustic velocity. Fig. 3(b) shows the estimated and measured values of resonance frequency (fr) for the ME stack as a function of the volume fraction n of the Ni phase. The variation of fr with respect to n
Fig. 1. (a) Schematic illustration of the tape-casting multilayer Pb(Zr, Ti)O3 (PZT) plate with high capacitance, (b) schematic illustration of the ME stack Ni/PZT-stack of multilayer Ni and PZT multilayer stack. The arrows M and P designate the magnetization and polarization directions, respectively.
C. Tang, C. Lu / Journal of Alloys and Compounds 686 (2016) 723e726
725
Fig. 3. (a) ME charge coefficient aQ as a function of frequency f, (b) variation of the resonance frequency fr with respect to the thickness ratio n of Ni foil for Ni/PZT laminate stack. In (a), the Hdc ¼ 0Oe and n ¼ 0.143. In (b), the dots are the experimental data, and line is the calculation using Eq. (1).
can be calculated using the material parameters shown in Table 1. The calculated and experimental data fit quite well though slight discrepancy occurs. With the thickness ratio increases from 0.143 to 0.454, the measured fr shifts from 91.78 to 110.9 kHz. According to the magneto-elasto-electric equivalent circuit method, the resonance ME voltage coefficient aME,r for L-T mode can be obtained [29]. Then the ME charge coefficient at resonance aQ,r is given by Ref. [19,29]
aQ ;r ¼ Ct aME;r ¼ εSaME;r ¼
8Qmech
p2
Snd33;m d31;p n 1 k231 sE11 þ ð1 nÞsH 33 (2)
m where n ¼ tmtþt is the thickness ratio of magnetostrictive layers in p the composite, ε is dielectric constant of PZT and S the effective electrode area. Importantly, this simple relationship provides that the aQ,r can be enhanced through an increased dielectric capacitance layers that are thinner and larger, which can be easily realized by using the multilayer PZT stack in Fig. 1(a). According to the above formula Eq. (2), the resonance ME field coefficient aQ,r of the laminate composite is not only dependent on various material characteristics of magnetostrictive and piezoelectric materials, but is also influenced by the ratio n. Therefore, we next investigated the aQ,r as a function of Hdc for Ni/PZT-stack with one to five layers of Ni foil (or n ¼ 0.143e0.454), as shown in Fig. 4. It is clearly that (i) large aQ,r at zero-biased is observed for Ni/PZT-stack. (ii) The aQ,r shows a hysteretic behavior during Hdc sweep. According to the relationship for aQ,r, one can derive aQ,r ∝ d33,m ¼ dl/dH. As shown in Fig. 2, the d33,m of hardprocessing Ni foil shows a hysteretic behavior during Hdc sweep (clockwise direction). Thus, the hysteretic behavior of the prepared composite Ni/PZT-stack is related to the nature of magnetostrictive hard-processing Ni. (iii) The loops move along Hdc axis with increasing of n, which can be attributed to the demagnetizing field effect. Since the demagnetizing field in samples increases with the layers of Ni foil, a larger optimum Hdc is needed. (iv) The largest aQ,r
Fig. 4. Resonance ME charge coefficient aQ,r as a function of Hdc for Ni/PZT-stack with one to five layers of Ni foil (or n ¼ 0.142e0.454). The inset shows estimated values of aQ,r a function of the ratio n of the Ni phase.
is achieved when n ¼ 0.4. One observes that the largest zero-biased aQ,r of Ni/PZT-stack with n ¼ 0.4 is 47.52 nC/Oe, which is a large improvement with a simple piezoelectric multilayer vibrator instead of piezoelectric single crystal materials. The inset in Fig. 4 shows the calculated estimated values of aQ,r for the composite as a function of the ratio n of the Ni phase. The material parameters are shown in Table 1. The d33,m of Ni foil is 15 nm/A for the calculation. Comparatively, the calculated aQ,r reaches the maximum values when n ¼ 0.38, which is in agreement with the experimental data (n ¼ 0.4). The discrepancy is due to the fact that the viscoelasticity of the epoxy layers is neglected in the calculation. Since the Ni/PZT-stack has large zero-biased ME charge coupling, it’s sensing characteristics is necessary to investigate.
Table 1 Material parameters for Ni, and PZT used for theoretical estimates.
Ni PZT
Density r (kg/m3)
Elastic constants (1012 m2/N)
Pizeolectic/piezomagnetic coefficient
Dielectric constant ε (ε33/ε0)
8900 6300
4.9 (s11) 13.3 (s33)
15 (nm/A) 175 (1012 m/V)
1750
726
C. Tang, C. Lu / Journal of Alloys and Compounds 686 (2016) 723e726
Acknowledgments This work was supported by the National Key Technology R&D Program of China (2013BAA02B02), Science and Technology Project of Guangdong Province under Grant No. 2013B010401012, Science and Technology Project of Shenzhen under Grant No. JCYJ20140424173418463, Special Funds for the Development of Strategic Emerging Industries in Guangdong Province under Grant No. 2012556036. References
Fig. 5. Magnetoelectric charge Q as a function of the driving magnetic field at low frequency of 1 kHz and resonance frequency of 108 kHz, respectively.
Fig. 5 plots comparatively the zero-biased ME charge as a function of ac magnetic field Hac at low frequency of 1 kHz and resonance frequency of 108 kHz, respectively. It is obvious that the induced ME charge has a good near linear relation with Hac even at a small Hac of 103 Oe under zero-biased magnetic field. From the slope of the plots, the zero-biased ME charge of the sensor at 108 kHz and 1 kHz is determined to be 47.52 nC/Oe and 0.55 nC/Oe, respectively. Clearly, the Ni/PZT-stack sensor seems quite promising for the detection of minute magnetic field variations at ambient conditions. Nevertheless, the obtained ME charge is higher than most of the conventional magnetostrictive/piezoelectric ME composites operating at zero-biased magnetic field [19,20,23,30], which makes it particularly attractive for technological applications.
4. Conclusion In summary, we proposed a miniature homogenous ME stack consisting of hard-processing Ni foils and a tape-casting multilayer PZT plate with high capacitance. The large ME charge effects of Ni/ PZT-stack are observed and investigated in detail. As the thickness ratio n increases, the value of aQ,r first increases and reaches a maximum at approximately n ¼ 0.4, and then decreases afterward. The induced ME charge has a good near linear relation with Hac even at a small Hac of 103 Oe under zero-biased magnetic field. The zero-biased ME charge sensitivities are determined to be 47.52 nC/ Oe and 0.55 nC/Oe at 108 kHz and 1 kHz, respectively, which are far larger higher than that previously reported for other ME laminates. It provides a new route for achieving large zero-bias ME charge response.
[1] M. Fiebig, J. Phys. D 38 (2005) R123eR125. [2] G. Lawes, G. Srinivasan, J. Phys. D Appl. Phys. 44 (2011) 243001. [3] C.A.F. Vaz, Jason Hoffman, Charles H. Ahn, Ramamoorthy Ramesh, Adv. Mater. 22 (2010) 2900e2918. [4] Junyi Zhai, Zengping Xing, Shuxiang Dong, Jiefang Li, Dwight Viehland, J. Am. Ceram. Soc. 91 (2) (2008) 351e358. [5] D.V. Chashin, Y.K. Fetisov, K.E. Kamentsev, G. Srinivasan, Appl. Phys. Lett. 92 (2008) 102511. [6] M.I. Bichurin, R.V. Petrov, V.M. Petrov, Appl. Phys. Lett. 103 (2013) 092902. €chel, [7] A. Kulkarni, K. Meurisch, I. Teliban, R. Jahns, T. Strunskus, A. Piorra, R. Kno F. Faupel, Appl. Phys. Lett. 104 (2014) 022904. [8] J. Das, J. Gao, Z. Xing, J.F. Li, D. Viehland, Appl. Phys. Lett. 95 (2009) 092501. [9] Y.H. Lin, N. Cai, J.Y. Zhai, et al., Phys. Rev. B 72 (2005) 012405. [10] C.H. Sim, A.Z.Z. Pan, J. Wang, J. Appl. Phys. 103 (2008) 124109. [11] J.H. Cheng, Y.G. Wang, D. Xie, Appl. Phys. Lett. 104 (2014) 252411. [12] Shuxiang Dong, Jie-Fang Li, Dwight Viehland, IEEE Trans. Ultrason. Ferroelectr. Freq. Control 51 (7) (2004) 793e798. [13] Henry Greve, Eric Woltermann, Robert Jahns, Stephan Marauska, €chel, Manfred Wuttig, Eckhard Quandt, Appl. Bernhard Wagner, Reinhard Kno Phys. Lett. 97 (2010) 152503. €chel, [14] S. Marauskaa, R. Jahns, C. Kirchhof, M. Claus, E. Quandtc, R. Kno B. Wagner, Sens. Actuators A 189 (2013) 321e327. [15] C.J. Lu, C.B. Xu, L. Wang, J.P. Gao, J.G. Gui, C.H. Lin, Rev. Sci. Instrum. 85 (2014) 115003. [16] Scott D. Moss, Joshua E. McLeod, Ian G. Powlesland, Steve C. Galea, Sens. Actuators A 175 (2013) 165e168. [17] Nicolas Tiercelin, Yannick Dusch, Vladimir Preobrazhensky, Philippe Pernod, J. Appl. Phys. 109 (2011) 07D726. [18] Jing Ma, Zheng Li, Yuanhua Lin, C.W. Nan, J. Magn. Magn. Mater. 323 (2011) 101e103. [19] Shuxiang Dong, Junyi Zhai, Jie-Fang Li, D. Viehland, Eric Summers, J. Appl. Phys. 101 (2007) 124102. [20] Lingying Li, Jie Jiao, Jianwei Chen, Yuting Liu, Chundong Xu, Bo Ren, Wenning Di, Xiangyong Zhao, Haosu Luo, Weiping Jing, Xiumei Shao, Jiaxiong Fang, J. Appl. Phys. 114 (2013) 027011. [21] D.Y. Huang, C.J. Lu, B. Han, X. Wang, C.X. Li, C.B. Xu, J.G. Gui, C.H. Lin, Appl. Phys. Lett. 105 (2014) 63502. € che, E. Quandt, [22] E. Lage, C. Kirchhof, V. Hrkac, L. Kienle, R. Jahns, R. Kno D. Meyners, Nat. Mater. 11 (2012) 523e529. [23] M.H. Li, Z.G. Wang, Y.J. Wang, J.F.G. Li, D. Viehland, J. Appl. Phys. 111 (2012) 104504. [24] Y.X. Zhao, C.J. Lu, Rev. Sci. Instrum. 86 (2015) 036101. [25] Yongke Yan, Yuan Zhou, Shashank Priya, Appl. Phys. Lett. 102 (2013) 052907. [26] D. Zhou, F.F. Wang, L.H. Luo, J. Chen, W.W. Ge, X.Y. Zhao, H.S. Luo, J. Phys. D. Appl. Phys. 41 (18) (2008) 185402. [27] D.R. Patil, Y. Chai, R.C. Kambale, B.G. Jeon, K. Yoo, J. Ryu, W.H. Yoon, D.S. Park, D.Y. Jeong, S.G. Lee, J. Lee, J.H. Nam, J.H. Cho, B.I. Kim, K.H. Kim, Appl. Phys. Lett. 102 (2013) 062909. [28] J.G. Wan, Z.Y. Li, Y. Wang, M. Zeng, G.H. Wang, J.-M. Liu, Appl. Phys. Lett. 86 (2005) 202504. [29] S.X. Dong, J.F. Li, D. Viehland, IEEE Trans. Ultrason. Ferroelectr. Freq. Control 50 (2003) 1253. [30] S.X. Dong, J.Y. Zhai, Z.P. Xing, J.F. Li, D. Viehland, Appl. Phys. Lett. 91 (2007) 022915.
Contents lists available at ScienceDirect
Journal of Alloys and Compounds journal homepage: http://www.elsevier.com/locate/jalcom
Strong self-biased magnetoelectric charge coupling in a homogenous laminate stack for magnetic sensor Chengpei Tang a, Caijiang Lu b, * a b
School of Engineering, Sun Yat-sen University, Guangzhou 510006, China Guizhou Electric Power Research Institute, China Southern Power Grid, Guiyang 550002, China
a r t i c l e i n f o
a b s t r a c t
Article history: Received 19 November 2015 Received in revised form 1 June 2016 Accepted 8 June 2016 Available online 11 June 2016
In this study, we report a strong self-biased magnetoelectric (ME) charge coupling in homogenous twophase magnetostrictive/piezoelectric laminate stack. The proposed ME stack Ni/PZT-stack is made up of hard-processing Nickel foils (Ni) and a tape-casting multilayer Pb(Zr, Ti)O3 (PZT) plate with high capacitance. Resonant ME couplings of Ni/PZT-stack with different thickness ratio (n) of Ni foil are investigated in detail. The experimental results show that the Ni/PZT-stack with n ¼ 0.4 has maximum zero-biased resonant ME charge coefficient (aQ,r) of 47.52 nC/Oe, which is far larger higher than that previously reported for other ME laminates. The proposed self-biased miniature ME structure may be useful for multifunctional devices such as electromagnetic energy harvesting, magnetic-to-electric generators or magnet field sensing based on a charge-detection method. © 2016 Elsevier B.V. All rights reserved.
Keywords: Self-biased Magnetoelectric Charge-coupling Laminate Sensor
1. Introduction Magnetoelectric (ME) composites consisting of magnetostrictive and piezoelectric components have larger ME effect than that of any natural signal-phase ME material by several orders of magnitude [1]. The extrinsic ME effect has been widely investigated both by theory and through experiment in various magnetostrictive and piezoelectric ME composites operated in several different modes [2e4]. It has been shown that the ME response of the laminated composites is determined by several major aspects: (a) the materials characteristics and mechanical coefficients of the constituents [2e13]; (b) the composition ratio of the piezoelectric and magnetostrictive layers [8e10]; (c) the type of boundary constituents [11]; and (d) the composition structure [4,12e15]. From previous reports, large ME charge couplings are important for developing applications, such as electromagnetic energy harvesters [16], magnetic-to-electric generators [17,18] or magnet field sensor based on a charge-detection method [19,20]. However, most of the reported ME composites in the literature has small charge induced from the ME effect due to the low dielectric capacitance [2e18]. For the magnetostrictive phase, the most severe limitation
* Corresponding author. E-mail address: [email protected] (C. Lu). http://dx.doi.org/10.1016/j.jallcom.2016.06.073 0925-8388/© 2016 Elsevier B.V. All rights reserved.
results from the dependence of the piezomagnetic coefficient dm on dc bias field (Hdc) [1e20]. To overcome these limitations arising from an external bias magnetic field, more and more researchers have focused on self-biased ME effect in the composites [21e25]. For the piezoelectric phase, the PZT have been the first choice in the design of ME laminate composite [2e4]. Unfortunately, the charge induced from the effect is small due to the low dielectric capacitance. And high cost and low Curie temperature of piezoelectric single crystal fibers and complex fabrication process present challenges in implementation at commercial scale [26,27]. As a result, for the physically interesting, technologically important and environmental perspective, it is indeed to design a novel laminate composite consisting of self-biased magnetostrictive component with simple processing and new PZT candidates with high capacitance, which should have large self-biased ME charge coupling. In this paper, we report a large self-biased ME charge coupling in a homogenous laminate stack. The laminate stack is consisting of a PZT multilayer stack as the piezoelectric component and hardprocessing Nickel (Ni) as the magnetostrictive component. By using single-phase homogenous magnetostrictive hard-processing Ni without any complicated synthesis processes, a strong self-biased effect can be obtained. And with use of the PZT multilayer stack with high capacitance, a dramatically giant ME charge coupling is observed. Such brand-new configurations yield the giant ME charge coupling at zero-biased magnetic field. And the preparation
724
C. Tang, C. Lu / Journal of Alloys and Compounds 686 (2016) 723e726
method is simple and feasible. 2. Experimental Fig. 1(a) shows the schematic illustration of tape-casting multilayer PZT plate with high capacitance. The piezoelectric multilayer stack was prepared with PZT-8 powders by a tape casting method. Then it was polarized in the thickness direction under an electric field of 3.5 kV/mm. The multilayer had a dimension of 20 3 0.6 mm3 with 15 layers and 40 mm per-layer. From Fig. 1(a), the positive and negative electrodes for each 40 mm layer of the PZT stack are in parallel, yielding a large capacitance of 236 nF for the multilayer PZT stack. The Ni foils (provided by Baoji metal Materials and Equipment Manufacturing Co., Ltd., China) was cut with the dimension of 12 6 0.1 mm3. As shown in Fig. 1(b), the multilayer ME composites Ni/PZT-stack was obtained by bonding the piezoelectric plate and Ni foils together with epoxy adhesive and pressed using a hydraulic press to make the epoxy layers as thin and perfect as possible. In experiments, a pair of neodymium permanent magnets (NdFeB) was used to provide the dc bias magnetic field (Hdc). A signal generator (Tektronix AFG3021B) provided a controllable input current to a long straight solenoid coil which was used to generate a small ac magnetic field (Hac). We measured the induced charge Q with a charge amplifier connected to an oscilloscope. Then the ME charge coefficient can be obtained, aQ ¼ vQME/vHac. 3. Results and discussion The ME coupling effect is a product property of the piezomagnetism, the piezoelectricity of the corresponding phases and their coupling, which can be characterized by aQ ¼ vQME/vH ¼ kdmdp. where k is a coupling factor between the two phases, dm and dp is the ME coefficient of the composite. Therefore, first we measured Hdc dependence of dynamic magnetostrictive coefficient (d33,m) for the hard-processing Ni foil under free conditions, as shown in Fig. 2. The vibration velocity v of the sample was measured by Doppler vibrometer (Polytec OFV-5000). Then the piezomagnetic coefficient was calculated by d33,m ¼ dl/dH ¼ v/(pGlHac). Here G is the vibration frequency, l is the length, and Hac is the external AC magnetic field. The d33,m of hard-processing Ni foil shows a hysteretic behavior during Hdc sweep (clockwise direction). Importantly, large zerobiased d33,m is observed. As Hdc increases from 600 Oe to 0 Oe, d33,m increases to a peak value of 15.1 nm/A for Hdc ¼ 51Oe. The
Fig. 2. The piezomagnetic coefficient d33,m as a function of Hdc for hard-processing Ni foil.
d33,m at Hdc ¼ 0 Oe is 13.3 nm/A. As Hdc increases from 0 Oe to 100 Oe, the d33,m decreases down to zero for Hdc ¼ 45Oe. Beyond this zero-crossing, a further decrease in Hdc results in an increase in d33,m for 45 Oe < Hdc < 140 Oe. A peak value of 13.1 nm/A is observed at Hdc ¼ 140 Oe and then decreases down for Hdc > 140 Oe. The results demonstrate that the hard-processing Ni foil with hysteretic behavior is effective to obtain self-biased effect. Fig. 3(a) show the zero-biased ME charge coefficient aQ ¼ vQME/ vH as a function of frequency around resonance. In this figure, the thickness of Ni foil is 0.1 mm, or the thickness ration n [n ¼ tm/ (tm þ tp), with tm or tp being the respective thickness of the Ni and the PZT layer, respectively] of Ni foil in ME stack is 0.143. It is clearly that the aQ reaches 41.5 nC/Oe at resonance frequency fr ¼ 91.78 kHz. For the ME composite consisting of mechanically coupled magnetostrictive and piezoelectric layers, the resonance frequency of ME composite is [28].
fr ¼ V 2l
(1)
where V ¼ 1=ðrs11 Þ1=2 (r and s11 are the average density and equivalent elastic compliance, respectively) is the average acoustic velocity. Fig. 3(b) shows the estimated and measured values of resonance frequency (fr) for the ME stack as a function of the volume fraction n of the Ni phase. The variation of fr with respect to n
Fig. 1. (a) Schematic illustration of the tape-casting multilayer Pb(Zr, Ti)O3 (PZT) plate with high capacitance, (b) schematic illustration of the ME stack Ni/PZT-stack of multilayer Ni and PZT multilayer stack. The arrows M and P designate the magnetization and polarization directions, respectively.
C. Tang, C. Lu / Journal of Alloys and Compounds 686 (2016) 723e726
725
Fig. 3. (a) ME charge coefficient aQ as a function of frequency f, (b) variation of the resonance frequency fr with respect to the thickness ratio n of Ni foil for Ni/PZT laminate stack. In (a), the Hdc ¼ 0Oe and n ¼ 0.143. In (b), the dots are the experimental data, and line is the calculation using Eq. (1).
can be calculated using the material parameters shown in Table 1. The calculated and experimental data fit quite well though slight discrepancy occurs. With the thickness ratio increases from 0.143 to 0.454, the measured fr shifts from 91.78 to 110.9 kHz. According to the magneto-elasto-electric equivalent circuit method, the resonance ME voltage coefficient aME,r for L-T mode can be obtained [29]. Then the ME charge coefficient at resonance aQ,r is given by Ref. [19,29]
aQ ;r ¼ Ct aME;r ¼ εSaME;r ¼
8Qmech
p2
Snd33;m d31;p n 1 k231 sE11 þ ð1 nÞsH 33 (2)
m where n ¼ tmtþt is the thickness ratio of magnetostrictive layers in p the composite, ε is dielectric constant of PZT and S the effective electrode area. Importantly, this simple relationship provides that the aQ,r can be enhanced through an increased dielectric capacitance layers that are thinner and larger, which can be easily realized by using the multilayer PZT stack in Fig. 1(a). According to the above formula Eq. (2), the resonance ME field coefficient aQ,r of the laminate composite is not only dependent on various material characteristics of magnetostrictive and piezoelectric materials, but is also influenced by the ratio n. Therefore, we next investigated the aQ,r as a function of Hdc for Ni/PZT-stack with one to five layers of Ni foil (or n ¼ 0.143e0.454), as shown in Fig. 4. It is clearly that (i) large aQ,r at zero-biased is observed for Ni/PZT-stack. (ii) The aQ,r shows a hysteretic behavior during Hdc sweep. According to the relationship for aQ,r, one can derive aQ,r ∝ d33,m ¼ dl/dH. As shown in Fig. 2, the d33,m of hardprocessing Ni foil shows a hysteretic behavior during Hdc sweep (clockwise direction). Thus, the hysteretic behavior of the prepared composite Ni/PZT-stack is related to the nature of magnetostrictive hard-processing Ni. (iii) The loops move along Hdc axis with increasing of n, which can be attributed to the demagnetizing field effect. Since the demagnetizing field in samples increases with the layers of Ni foil, a larger optimum Hdc is needed. (iv) The largest aQ,r
Fig. 4. Resonance ME charge coefficient aQ,r as a function of Hdc for Ni/PZT-stack with one to five layers of Ni foil (or n ¼ 0.142e0.454). The inset shows estimated values of aQ,r a function of the ratio n of the Ni phase.
is achieved when n ¼ 0.4. One observes that the largest zero-biased aQ,r of Ni/PZT-stack with n ¼ 0.4 is 47.52 nC/Oe, which is a large improvement with a simple piezoelectric multilayer vibrator instead of piezoelectric single crystal materials. The inset in Fig. 4 shows the calculated estimated values of aQ,r for the composite as a function of the ratio n of the Ni phase. The material parameters are shown in Table 1. The d33,m of Ni foil is 15 nm/A for the calculation. Comparatively, the calculated aQ,r reaches the maximum values when n ¼ 0.38, which is in agreement with the experimental data (n ¼ 0.4). The discrepancy is due to the fact that the viscoelasticity of the epoxy layers is neglected in the calculation. Since the Ni/PZT-stack has large zero-biased ME charge coupling, it’s sensing characteristics is necessary to investigate.
Table 1 Material parameters for Ni, and PZT used for theoretical estimates.
Ni PZT
Density r (kg/m3)
Elastic constants (1012 m2/N)
Pizeolectic/piezomagnetic coefficient
Dielectric constant ε (ε33/ε0)
8900 6300
4.9 (s11) 13.3 (s33)
15 (nm/A) 175 (1012 m/V)
1750
726
C. Tang, C. Lu / Journal of Alloys and Compounds 686 (2016) 723e726
Acknowledgments This work was supported by the National Key Technology R&D Program of China (2013BAA02B02), Science and Technology Project of Guangdong Province under Grant No. 2013B010401012, Science and Technology Project of Shenzhen under Grant No. JCYJ20140424173418463, Special Funds for the Development of Strategic Emerging Industries in Guangdong Province under Grant No. 2012556036. References
Fig. 5. Magnetoelectric charge Q as a function of the driving magnetic field at low frequency of 1 kHz and resonance frequency of 108 kHz, respectively.
Fig. 5 plots comparatively the zero-biased ME charge as a function of ac magnetic field Hac at low frequency of 1 kHz and resonance frequency of 108 kHz, respectively. It is obvious that the induced ME charge has a good near linear relation with Hac even at a small Hac of 103 Oe under zero-biased magnetic field. From the slope of the plots, the zero-biased ME charge of the sensor at 108 kHz and 1 kHz is determined to be 47.52 nC/Oe and 0.55 nC/Oe, respectively. Clearly, the Ni/PZT-stack sensor seems quite promising for the detection of minute magnetic field variations at ambient conditions. Nevertheless, the obtained ME charge is higher than most of the conventional magnetostrictive/piezoelectric ME composites operating at zero-biased magnetic field [19,20,23,30], which makes it particularly attractive for technological applications.
4. Conclusion In summary, we proposed a miniature homogenous ME stack consisting of hard-processing Ni foils and a tape-casting multilayer PZT plate with high capacitance. The large ME charge effects of Ni/ PZT-stack are observed and investigated in detail. As the thickness ratio n increases, the value of aQ,r first increases and reaches a maximum at approximately n ¼ 0.4, and then decreases afterward. The induced ME charge has a good near linear relation with Hac even at a small Hac of 103 Oe under zero-biased magnetic field. The zero-biased ME charge sensitivities are determined to be 47.52 nC/ Oe and 0.55 nC/Oe at 108 kHz and 1 kHz, respectively, which are far larger higher than that previously reported for other ME laminates. It provides a new route for achieving large zero-bias ME charge response.
[1] M. Fiebig, J. Phys. D 38 (2005) R123eR125. [2] G. Lawes, G. Srinivasan, J. Phys. D Appl. Phys. 44 (2011) 243001. [3] C.A.F. Vaz, Jason Hoffman, Charles H. Ahn, Ramamoorthy Ramesh, Adv. Mater. 22 (2010) 2900e2918. [4] Junyi Zhai, Zengping Xing, Shuxiang Dong, Jiefang Li, Dwight Viehland, J. Am. Ceram. Soc. 91 (2) (2008) 351e358. [5] D.V. Chashin, Y.K. Fetisov, K.E. Kamentsev, G. Srinivasan, Appl. Phys. Lett. 92 (2008) 102511. [6] M.I. Bichurin, R.V. Petrov, V.M. Petrov, Appl. Phys. Lett. 103 (2013) 092902. €chel, [7] A. Kulkarni, K. Meurisch, I. Teliban, R. Jahns, T. Strunskus, A. Piorra, R. Kno F. Faupel, Appl. Phys. Lett. 104 (2014) 022904. [8] J. Das, J. Gao, Z. Xing, J.F. Li, D. Viehland, Appl. Phys. Lett. 95 (2009) 092501. [9] Y.H. Lin, N. Cai, J.Y. Zhai, et al., Phys. Rev. B 72 (2005) 012405. [10] C.H. Sim, A.Z.Z. Pan, J. Wang, J. Appl. Phys. 103 (2008) 124109. [11] J.H. Cheng, Y.G. Wang, D. Xie, Appl. Phys. Lett. 104 (2014) 252411. [12] Shuxiang Dong, Jie-Fang Li, Dwight Viehland, IEEE Trans. Ultrason. Ferroelectr. Freq. Control 51 (7) (2004) 793e798. [13] Henry Greve, Eric Woltermann, Robert Jahns, Stephan Marauska, €chel, Manfred Wuttig, Eckhard Quandt, Appl. Bernhard Wagner, Reinhard Kno Phys. Lett. 97 (2010) 152503. €chel, [14] S. Marauskaa, R. Jahns, C. Kirchhof, M. Claus, E. Quandtc, R. Kno B. Wagner, Sens. Actuators A 189 (2013) 321e327. [15] C.J. Lu, C.B. Xu, L. Wang, J.P. Gao, J.G. Gui, C.H. Lin, Rev. Sci. Instrum. 85 (2014) 115003. [16] Scott D. Moss, Joshua E. McLeod, Ian G. Powlesland, Steve C. Galea, Sens. Actuators A 175 (2013) 165e168. [17] Nicolas Tiercelin, Yannick Dusch, Vladimir Preobrazhensky, Philippe Pernod, J. Appl. Phys. 109 (2011) 07D726. [18] Jing Ma, Zheng Li, Yuanhua Lin, C.W. Nan, J. Magn. Magn. Mater. 323 (2011) 101e103. [19] Shuxiang Dong, Junyi Zhai, Jie-Fang Li, D. Viehland, Eric Summers, J. Appl. Phys. 101 (2007) 124102. [20] Lingying Li, Jie Jiao, Jianwei Chen, Yuting Liu, Chundong Xu, Bo Ren, Wenning Di, Xiangyong Zhao, Haosu Luo, Weiping Jing, Xiumei Shao, Jiaxiong Fang, J. Appl. Phys. 114 (2013) 027011. [21] D.Y. Huang, C.J. Lu, B. Han, X. Wang, C.X. Li, C.B. Xu, J.G. Gui, C.H. Lin, Appl. Phys. Lett. 105 (2014) 63502. € che, E. Quandt, [22] E. Lage, C. Kirchhof, V. Hrkac, L. Kienle, R. Jahns, R. Kno D. Meyners, Nat. Mater. 11 (2012) 523e529. [23] M.H. Li, Z.G. Wang, Y.J. Wang, J.F.G. Li, D. Viehland, J. Appl. Phys. 111 (2012) 104504. [24] Y.X. Zhao, C.J. Lu, Rev. Sci. Instrum. 86 (2015) 036101. [25] Yongke Yan, Yuan Zhou, Shashank Priya, Appl. Phys. Lett. 102 (2013) 052907. [26] D. Zhou, F.F. Wang, L.H. Luo, J. Chen, W.W. Ge, X.Y. Zhao, H.S. Luo, J. Phys. D. Appl. Phys. 41 (18) (2008) 185402. [27] D.R. Patil, Y. Chai, R.C. Kambale, B.G. Jeon, K. Yoo, J. Ryu, W.H. Yoon, D.S. Park, D.Y. Jeong, S.G. Lee, J. Lee, J.H. Nam, J.H. Cho, B.I. Kim, K.H. Kim, Appl. Phys. Lett. 102 (2013) 062909. [28] J.G. Wan, Z.Y. Li, Y. Wang, M. Zeng, G.H. Wang, J.-M. Liu, Appl. Phys. Lett. 86 (2005) 202504. [29] S.X. Dong, J.F. Li, D. Viehland, IEEE Trans. Ultrason. Ferroelectr. Freq. Control 50 (2003) 1253. [30] S.X. Dong, J.Y. Zhai, Z.P. Xing, J.F. Li, D. Viehland, Appl. Phys. Lett. 91 (2007) 022915.