Enhanced Self-Biased Magnetoelectric Coupling ... - ACS Publications

Jan 8, 2018 - The tunability of self-biased ME coupling in PZT/Ni composite has been found ... harvesters, etc.1−5 The ME coupling in a laminate com...
0 downloads 0 Views 6MB Size
Download PDF
Research Article Cite This: ACS Appl. Mater. Interfaces 2018, 10, 11018−11025

www.acsami.org

Enhanced Self-Biased Magnetoelectric Coupling in Laser-Annealed Pb(Zr,Ti)O3 Thick Film Deposited on Ni Foil Haribabu Palneedi,† Deepam Maurya,‡ Liwei D. Geng,§ Hyun-Cheol Song,‡,¶ Geon-Tae Hwang,† Mahesh Peddigari,† Venkateswarlu Annapureddy,# Kyung Song,▽ Yoon Seok Oh,∥ Su-Chul Yang,⊥ Yu U. Wang,§ Shashank Priya,*,‡ and Jungho Ryu*,†,▲ †

Functional Ceramics Group, Korea Institute of Materials Science (KIMS), Changwon 51508, Korea Bio-inspired Materials and Devices Laboratory (BMDL), Center for Energy Harvesting Materials and Systems (CEHMS), Virginia Tech, Blacksburg, Virginia 24061, United States § Department of Materials Science and Engineering, Michigan Technological University, Houghton, Michigan 49931, United States ¶ Center for Electronic Materials, Korea Institute of Science and Technology (KIST), Seoul 02792, Korea # CSIR-National Physical Laboratory, Dr. K. S. Krishnan Road, New Delhi 110012, India ▽ Department of Materials Modeling and Characterization, Korea Institute of Materials Science (KIMS), Changwon 51508, Korea ∥ Department of Physics, Ulsan National Institute of Science and Technology (UNIST), Ulsan 44919, Korea ⊥ Department of Chemical Engineering, Dong-A University, Busan 49315, Korea ▲ School of Materials Science and Engineering, Yeungnam University, Gyeongsan 38541, Korea

Downloaded via KAOHSIUNG MEDICAL UNIV on July 6, 2018 at 08:12:30 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

‡

S Supporting Information *

ABSTRACT: Enhanced and self-biased magnetoelectric (ME) coupling is demonstrated in a laminate heterostructure comprising 4 μm-thick Pb(Zr,Ti)O3 (PZT) film deposited on 50 μm-thick flexible nickel (Ni) foil. A unique fabrication approach, combining room temperature deposition of PZT film by granule spray in vacuum (GSV) process and localized thermal treatment of the film by laser radiation, is utilized. This approach addresses the challenges in integrating ceramic films on metal substrates, which is often limited by the interfacial chemical reactions occurring at high processing temperatures. Laserinduced crystallinity improvement in the PZT thick film led to enhanced dielectric, ferroelectric, and magnetoelectric properties of the PZT/Ni composite. A high self-biased ME response on the order of 3.15 V/cm·Oe was obtained from the laser-annealed PZT/Ni film heterostructure. This value corresponds to a ∼2000% increment from the ME response (0.16 V/cm·Oe) measured from the as-deposited PZT/Ni sample. This result is also one of the highest reported values among similar ME composite systems. The tunability of self-biased ME coupling in PZT/Ni composite has been found to be related to the demagnetization field in Ni, strain mismatch between PZT and Ni, and flexural moment of the laminate structure. The phase-field model provides quantitative insight into these factors and illustrates their contributions toward the observed self-biased ME response. The results present a viable pathway toward designing and integrating ME components for a new generation of miniaturized tunable electronic devices. KEYWORDS: Pb(Zr,Ti)O3, nickel, laser annealing, self-biased, magnetoelectric

1. INTRODUCTION

(Hbias), the magnetically induced mechanical deformation (or AC magnetostriction) is negligible in most magnetostrictive materials. Consequently, the piezomagnetic coefficient (qij = dλij/dH) could be nearly zero for Hbias = 0. The variation of αME is similar to that of qij, and thus an optimum DC bias is required, in addition to AC magnetic field, to achieve the maximum in ME coupling of the composites. 7,8 This

Magnetoelectric (ME) composites comprising magnetostrictive and piezoelectric materials are promising candidates for a variety of applications including magnetic sensors, voltage tunable inductors, data storage elements, spintronics, energy harvesters, etc.1−5 The ME coupling in a laminate composite, quantified in terms of ME voltage coefficient (αME), depends primarily on the interfacial strain transfer between the magnetostrictive and piezoelectric layers.6 Magnetostrictive strain (λij) exhibits quadratic variation with the external magnetic field (H) applied. In the absence of DC bias © 2018 American Chemical Society

Received: November 2, 2017 Accepted: January 8, 2018 Published: January 8, 2018 11018

DOI: 10.1021/acsami.7b16706 ACS Appl. Mater. Interfaces 2018, 10, 11018−11025

Research Article

ACS Applied Materials & Interfaces necessitates the use of permanent magnets or another DC magnetic source around ME composites, resulting in bulky devices with additional electromagnetic interference, which hinders their on-chip integration. To circumvent above-mentioned issues, research has focused on the development of self-biased magnetoelectric (SME) composites exhibiting finite αME at the zero magnetic bias.9,10 In composites, the SME coupling can be realized by inducing a built-in magnetic bias in the magnetostrictive layer. Among different approaches proposed for achieving SME, the most elegant technique relies on taking advantage of magnetization hysteresis in magnetostrictive material.10 The remanent magnetization in hysteretic magnetostrictive material can be modulated by applying AC magnetic field (shifting of the hysteresis loop on field axis) to produce an internal magnetic bias and thereby finite remanent αME at zero bias.11 SME response has been modeled by taking into account the nature of magnetization and its tunability through varying the demagnetization state and the resultant differential magnetic flux density distribution in the magnetostrictive layer.12,13 For a finite non-spherical ferromagnet, the demagnetization factor has been observed to greatly depend on its structural parameters (size and geometry). Optimization of these parameters results in effective control over the position and magnitude of maximum αME of the composite.14−16 However, a comprehensive understanding of the key factors influencing the SME behavior in composites is still lacking. Nickel (Ni) has been the most utilized material in SME composites, based on hysteretic magnetostrictive effect.17−20 However, in majority of the studies, the ME composites were prepared by epoxy bonding of the piezoelectric layer with the Ni layer, which has limited relevance for device fabrication. Few studies have pursued the deposition of piezoelectric films on Ni, employing techniques that were complex, limited to a small area, and required careful selection of deposition parameters and material compositions.21,22 Besides, it is also necessary to optimize the post-deposition thermal treatment, adopted to enhance the properties of the piezoelectric films, for minimizing the thermal expansion mismatch between the film and substrate and for mitigating the interfacial reactions during fabrication. This study provides a fundamental understanding of the SME phenomenon in a piezoelectric Pb(Zr,Ti)O3 (PZT) thick film deposited on a flexible magnetostrictive Ni foil. The choice of PZT was based on its high piezoelectric voltage constant (gij), because, αME ∝ gij.23 Generally, the crystallization of PZT films requires thermal annealing at temperatures above 600 °C.24,25 However, such high temperatures could lead to oxidation of Ni and interfacial chemical reactions between PZT and Ni. Although use of buffer layers between PZT and Ni and post-deposition treatment by rapid thermal annealing (RTA) could improve the mechanical adhesion and crystallization and inhibit interfacial reactions, sometimes these approaches have detrimental effects on the interfacial strain transfer and ME coupling.21,22,26,27 In this work, direct deposition of PZT film (4 μm-thick) was carried out using granule spray in vacuum (GSV) at room temperature (RT).28,29 Crystallization of the PZT film was induced through laser annealing (Figure 1), which addressed the challenges in achieving interfacial compatibility of the film with metal substrates. By doing so, we obtained a highly enhanced SME response of 3.15 V/cm·Oe in the PZT/Ni composite, which is the highest magnitude among similar ME composite systems reported in literature. Experimental results on SME response in

Figure 1. (a, b) Laser irradiation of the PZT film on Ni substrate and the laser annealing sequence, respectively. Laser spot size, power, scan speed, and annealed area were 50 μm, 965 mW, 0.03 mm/s, and 10 × 5 mm2, respectively. Both the deposition and laser treatment of the PZT film were conducted at room temperature.

the PZT/Ni composite are explained by modeling the mesoscale phenomenon and its tunability.

2. EXPERIMENTAL METHODS Fabrication of the ME Composite. A 4 μm-thick PZT film was deposited on a thin flexible Ni foil (50 μm-thick) using the GSV process at room temperature, followed by localized annealing of the film with continuous-wave 560 nm ytterbium fiber laser radiation. PZT granules (d50 values of primary particle and granule were ∼1.3 μm and ∼100 μm, respectively, JA-1, JK Precision Electric, Korea) mixed with medical grade-dried air were sprayed (230 L/min flow rate) on to the Ni substrate (99.5% metals basis, Alfa Aesar, Ward Hill, MA) through a laval-type nozzle (400 mm-slit length) in a deposition chamber under vacuum (∼4 Torr). The details of the GSV process and the stages involved in film growth are described in our prior studies.30,31 Figure 1 illustrates the scheme of laser annealing of the PZT film. During annealing, the laser beam with an effective diameter of 50 μm was focused and illuminated over the target area (10 × 5 mm2) of the sample mounted on an X-Y linear stage. The incident laser power and the sample scanning speed were fixed at 965 mW and 0.03 mm/s, respectively. Characterization of the ME Composite. The as-deposited (AD) and laser-annealed (LA) PZT films on Ni were characterized by X-ray diffraction (XRD, D/Max 2200, Rigaku Corporation, Japan), Raman spectroscopy (LabRam HR800, Horiba Ltd., Japan), and transmission electron microscopy (TEM, JEM-2100F, JEOL Ltd., Japan), to understand the structural differences between them. The crosssectional features of PZT/Ni and the elemental distributions near the interfacial regions were examined by TEM and energy-dispersive X-ray spectroscopy (EDS) equipped with it. The magnetization of the AD and LA PZT/Ni samples was evaluated along the in-plane direction using a physical properties measurement system (PPMS, Quantum Design, San Diego, CA). For electrical and ME characterization, patterned circular Pt top electrodes (0.5 mm diameter) were sputter deposited on the PZT film while the Ni substrate served as the bottom electrode. Dielectric properties of the films were measured by an impedance analyzer (4294A, Agilent Technologies, Santa Clara, CA), and the polarization hysteresis behavior was evaluated by a ferroelectric test system (Precision LC II, Radiant Technologies, Duluth, GA). For ME measurement, the samples were corona poled at 130 °C for 20 min using a DC potential of 12 kV. The ME output voltage from the PZT/Ni composite was measured by lock-in amplifier (SR-850, Stanford Research Systems, Sunnyvale, CA) under offresonance conditions. The ME signal from the sample was obtained in its thickness direction (transverse αME) by subjecting it to superimposed AC (Hac = 1 Oe, f = 1 kHz) and DC (Hdc) magnetic fields along its in-plane direction. Finite Element Modeling. Magnetic flux density distribution of Ni was estimated by a finite element model using COMSOL Multiphysics (version 5.2) program. In this model, rectangular Ni sheets with different thicknesses (t = 50, 100, 200, 500 μm) but the same planar dimension of 10 × 5 mm2 (identical to that used in experiments) were considered. A magnetostatic insulating boundary condition was applied around the Ni sheet, which is assumed to have a 11019

DOI: 10.1021/acsami.7b16706 ACS Appl. Mater. Interfaces 2018, 10, 11018−11025

Research Article

ACS Applied Materials & Interfaces

Figure 2. (a) XRD patterns, (b) Raman spectra, and (c, d) SAED patterns of the as-deposited and laser-annealed PZT films on Ni, respectively. Radial intensity profiles of SAED patterns are also shown for both the samples (insets of c and d). All these structural characterization results suggest the improvement in crystallinity of the PZT film after laser irradiation.

Figure 3. (a−c) Cross-sectional TEM micrograph and the corresponding EDS elemental mapping of the laser-annealed PZT/Ni, respectively. (d) Magnetization hysteresis loops of the as-deposited and laser-annealed PZT/Ni composites. No apparent diffusion or chemical reaction was observed across the PZT/Ni interface after laser irradiation, which is further supported by the identical magnetization behavior of the composite before and after laser annealing. Phase-Field Modeling. Details of the phase-field modeling of the PZT/Ni ME composites are provided in Supporting Information.32,33 During the simulation process, the PZT layer is poled along its thickness direction, and an additional mismatch strain ε0ij(r) due to thermal expansion mismatch was introduced to the PZT layer. An inplane magnetic field ΔH is applied along the Ni layer length direction,

relative permeability of 600 when placed in air and subjected to a magnetic field of 1 Oe (Hac) along its in-plane direction.12,13 In each case, the Ni sheet was meshed with 20000 points and the in-plane magnetic field strength in response to zero DC magnetic bias was visualized along the center plane of Ni sheet. 11020

DOI: 10.1021/acsami.7b16706 ACS Appl. Mater. Interfaces 2018, 10, 11018−11025

Research Article

ACS Applied Materials & Interfaces and the generated magnetostrictive strain elastically interacts with the poled piezoelectric layer, leading to a change in ferroelectric domain structure and polarization response ΔP. The ME coefficient is determined as αME = (1/ε0εr) ΔP/ΔH.

3. RESULTS AND DISCUSSION The X-ray diffraction (XRD) patterns of the AD and LA PZT films on Ni foil (Figure 2(a)) showed peaks corresponding to a typical perovskite polycrystalline PZT. The LA PZT film exhibited higher intensities of the XRD peaks as compared to the AD film, which can be attributed to the improved crystallinity of the PZT by laser annealing. Figure 2(b) shows the Raman spectra obtained from the AD and LA PZT films. It can be observed that both the samples display similar peaks located at 210, 270, 560, 700, and 735 cm−1, which correspond to the E(2TO), B1+E, A1(3TO), A1(3LO), and E(3LO) modes, respectively.34 These spectra represent the typical perovskite phase of PZT with composition close to the morphotropic phase boundary between PbZrO3 and PbTiO3. The higher intensity of Raman peaks of the LA PZT film further confirmed its greater degree of crystallinity as compared to the AD PZT film.35 The selected area electron diffraction (SAED) patterns collected from the AD and LA PZT films on Ni are shown in Figure 2, parts (c) and (d), respectively. The improvement in crystallinity of PZT by laser annealing is also indicated by the higher intensity of the diffraction rings in the SAED pattern, observed from corresponding radial profile, of the LA PZT film. During GSV deposition, the powder granules are sprayed onto the substrate, with high kinetic energy, in a vacuum chamber. Consequently, the fracturing and deformation of primary particles, due to high-speed collision, results in a dispersed nanocrystalline phase in an amorphous matrix of the deposited film. The nanocrystalline regions in the AD PZT film may act as heterogeneous nucleation sites for the crystallization of amorphous phase under laser irradiation.28−31 The cross-sectional TEM image of the laser-annealed PZT/ Ni composite (Figure 3 (a)) indicates that the PZT film is well adhered to the Ni foil. Such a good interfacial bonding is favorable for efficient elastic strain transfer and enhanced ME coupling between PZT and Ni. The interfacial chemical homogeneity of LA PZT/Ni composite was investigated through EDS analysis. The corresponding elemental distribution maps (Figure 3(b,c)) confirm that there is no apparent diffusion or chemical reaction across the PZT/Ni interface after the laser annealing. The magnetization vs magnetic field (M-H) behavior of the AD and LA PZT/Ni composites is shown in Figure 3(d). Both samples exhibited almost identical but asymmetric M-H hysteresis loops (shifting of the hysteresis loop on field axis). The superposition of the anisotropic magnetization field and the applied AC magnetic field, during ME measurement, results in an internal magnetic bias in Ni. This gives rise to the self-biased ME coupling in the PZT/Ni composite. The EDS and M-H results suggest that the laserinduced heating was localized to PZT layer and did not cause any damage to the Ni substrate. This clearly demonstrates the advantage of laser radiation for localized thermal treatment of piezoelectric films on heat-sensitive substrates. The dielectric properties of the AD and LA PZT films are compared in Figure 4(a). Both films exhibited low dielectric loss (tan δ) of 0.07−0.08 over a frequency range of 1 kHz to 1 MHz. The dielectric constant (εr) of the LA PZT film (∼780 at 1 kHz) is much higher than that of the AD PZT film (∼220 at 1 kHz). Figure 4(b) shows the polarization−electric field (P-E)

Figure 4. (a) Dielectric properties and (b) Polarization hysteresis loops of the as-deposited and laser-annealed PZT films on Ni. (c) ME responses of the corresponding PZT/Ni composites. The electrical and ME properties of the PZT/Ni heterostructure were significantly improved due to the increase in crystallinity of the PZT film by laser irradiation.

hysteresis loops of the AD and LA PZT films on Ni foil. The LA PZT film showed significantly better ferroelectric polarization than the AD PZT film. The measured remanent polarization (Pr) value of the LA PZT film was 40 μC/cm2, whereas the AD film exhibited a Pr value of 7 μC/cm2. This increased Pr after laser annealing indicates the enhanced piezoelectric response of the LA PZT film, as the piezoelectric constant is expressed as dij = 2εQPr, where Q is the electrostriction constant.36 The higher dielectric and polarization properties of the LA PZT film compared with that of the AD film can be attributed to the improved crystallinity due to laser irradiation.31,37 The ME responses obtained from the AD and LA PZT/Ni composite samples are plotted in Figure 4(c). The values of maximum αME of the AD and LA composites were measured to be 0.16 and 3.15 V/cm·Oe, respectively. The superior ME performance of the LA PZT/Ni composite can be mainly attributed to its improved electrical properties due to increased crystallinity of the PZT film. Furthermore, minimized mechanical damping at the PZT/Ni interface, due to matching mechanical impedance of PZT (25−30 MRayl) and Ni (27 MRayl), can facilitate an efficient interfacial elastic strain transfer, leading to better ME coupling in PZT/Ni 11021

DOI: 10.1021/acsami.7b16706 ACS Appl. Mater. Interfaces 2018, 10, 11018−11025

Research Article

ACS Applied Materials & Interfaces

Table 1. Comparison of Reported ME Voltage Coefficients of PZT Films Deposited on Ni-Based Substrates and Their Annealing Conditionsa

a

ME composite

PZT/Ni thickness (μm)

annealing condition

maximum αME (V/cm·Oe)

self-biased αME (V/cm·Oe)

% SME

reference

PZT/Pt/NiZnFe2O4 PZT/Ni PZT/Pt/Ni PZT/LaNiO3/Ni PZT/Ni

1/550 1/200 0.4/200 20/500 4/50

FA, 650 °C RTA, 650 °C FA, 650 °C FA, 700 °C LA, 965 mW

0.14 0.22 0.772 1 3.15

0 0.2 0 0.2 3.15

nil 90 nil 20 100

39 21 22 27 this work

FA: furnance annealing, RTA: rapid thermal annealing, and LA: laser annealing.

Figure 5. (a) Variation of demagnetization factor with the thickness of the Ni sheet. (b) FEM results of in-plane magnetic flux density distribution along the center plane of Ni sheets of varying thickness. A thinner Ni sheet (50 μm-thick) was found to exhibit a higher magnetic flux concentration due to its lower demagnetization factor, compared to other thicker samples. The high flux density facilitates the realization of maximum αME under a low magnetic bias.

in the magnetostrictive layer.12,13 A high flux concentration in the magnetic phase was observed to facilitate enhanced ME coupling under a low magnetic bias.40,41 In ferromagnetic materials, the demagnetization field (Hd) is directly proportional to demagnetization factor (Nd) via Hd = MNd. Demagnetization in the magnetostrictive Ni layer depends mainly on its thickness, which is much smaller as compared to its transverse dimensions. By considering the influence of demagnetization field, the effective magnetic field (Heff) and the corresponding magnetic flux density (Beff) in the magnetic phase can be expressed using the following relations:12,13,42

composites.27,38 The reported ME coefficient values of PZT films deposited on Ni-based magnetostrictive substrates and their fabrication conditions are summarized in Table 1. From this table, it can be observed that the laser-annealed PZT/Ni composite outperforms the furnace-annealed (FA) and rapid thermal-annealed (RTA) PZT films on Ni-based substrates.21,22,27,39 As can be seen from Figure 4(c) and Table 1, the LA PZT film on Ni displayed a fully self-biased ME response, while the FA and RTA PZT/Ni samples exhibited a lower SME response than their respective maximum αME values. Because the laserinduced heat is localized to the PZT layer only, the hysteretic magnetostrictive properties of the Ni substrate remained intact (Figure 3(c)), leading to undiminished self-biased ME response, which is similar in both the as-deposited and laserannealed PZT films on Ni. The absence of such a large SME response in FA and RTA PZT films on Ni can be ascribed to the degraded magnetic properties of Ni at high annealing temperatures.21,22 The lower demagnetization in the thin Ni foil substrate, used in this study, might also have played a significant role in achieving the enhanced ME coupling at zero bias. The origin of internal magnetic bias generated in Ni was investigated by Zhou et al.12 using magnetic force microscopic analysis. It was revealed that the presence of macrosized domains with long-range ordering, in polycrystalline Ni, results in a sizable coercive field. Upon reorientation of the magnetic domains, a higher field is required to switch back to the random state, leading to asymmetry in the magnetization (M) hysteresis of Ni, and further to an anisotropic magnetostriction and piezomagnetic coefficient, as λij ∝ M2 and qij ∝ dM2/dH.39 While the occurrence of self-biased ME coupling in Ni-based composites is related to the anisotropic λij of Ni and the corresponding remanent qij at Hdc = 0, the tunability of SME response has been found to depend on size-induced demagnetization and the resultant differential magnetic flux distribution

Heff = Hbias − Hd = Hbias − MNd

(1)

Beff = μ0 (Heff + M ) = μ0 (Hbias + M ) − μ0 MNd

(2)

Here μ0 is the permeability of free space. Figure 5(a) shows the variation of Nd as a function of Ni foil thickness (50−500 μm).43 It is clear that a thinner Ni substrate exhibits a smaller Nd, which could lead to lower Hd and thus higher Heff. This implies that for achieving the same magnitude of Heff, one needs to apply smaller Hbias under lower Hd, and vice versa. Accordingly, a maximum value of αME for a thinner Ni substrate can be obtained at a lower magnetic bias in comparison to its thicker counterpart. To verify the above correlation, the magnetic flux density distribution was simulated, using a finite element model (FEM),12,13,42 for Ni sheets with different thicknesses (50− 500 μm), and the results are shown in Figure 5(b). As seen from these results, the flux concentration in the Ni sheet significantly depends on its thickness. A thinner Ni sheet could display much stronger magnetic induction due to the lower Hd and higher Heff. Consequently, a larger value of αME was observed in the thinner Ni sheet under a lower Hbias. The shifting of the αME peak position toward the zero bias with a decrease in thickness of the magnetostrictive layer was also 11022

DOI: 10.1021/acsami.7b16706 ACS Appl. Mater. Interfaces 2018, 10, 11018−11025

Research Article

ACS Applied Materials & Interfaces

Figure 6. Simulated domain structures (under zero magnetic bias), M-H hysteresis, and ME voltage coefficient αME for laminate PZT/Ni composites (a−c) under zero mismatch strain εmis = 0 with varying demagnetization factors in the Ni film plane: (a) Nd = 0.008, (b) Nd = 0.014, and (c) Nd = 0.026, and (d, e) under mismatch strain εmis = 0.003 with the same demagnetization factor Nd = 0.008 in (d) flexural (PZT/Ni) and (e) non-flexural (Ni/PZT/Ni) composites with 50 μm-thick Ni layer. Domain patterns are visualized by color maps with red, green, blue (RGB) components proportional to Px, Py, Pz in the PZT layer and Mx, My, Mz in the Ni layer, respectively.

observed in the case of Ni-based bulk ME laminates.12,13 On the basis of the above findings, it can be inferred that the low Hd and high Heff in the thin Ni foil (50 μm-thick) could be responsible for achieving a fully self-biased ME response in the PZT/Ni composite. In addition to the improved crystallinity of the PZT film, the strong magnetic induction in the thin Ni foil could also facilitate the enhanced ME coupling in the composite. The larger differences between maximum αME and SME responses, reported in the literature (Table 1) for PZT/ Ni film composites with thicker Ni substrates, could be attributed to the higher demagnetization and the resultant decrease in the magnetic flux concentration. Nevertheless, there seems to be a discrepancy between the SME responses of the PZT/Ni samples reported in the literature22,27 (Table 1), contrary to the trend expected, with respect to the Ni layer thickness. As explained in the following section, through phase-field simulations of M-H and ME behavior of the PZT/Ni composite, besides demagnetization in the magnetostrictive layer, other factors such as mismatch strain (thermal and poling) in the laminate composite and its structural configuration (bilayer/multilayer, flexural/non-flexural) will also significantly influence the SME behavior. The differences in quantities of the above parameters of the PZT/Ni samples fabricated under different experimental conditions in other studies22,27 (Table 1) might have led to the observed differences between their SME responses with respect to Ni

layer thickness. In fact, it was observed in our phase-field simulations that at zero mismatch strain, the PZT/Ni composite with 100 μm-thick Ni layer exhibited a better SME response than the composite having a 50 μm-thick Ni layer (Figure 6(a,b)), which is in contrast to the trend expected. Further, introducing some mismatch strain in the PZT/Ni composite with a 50 μm-thick Ni layer led to fully self-biased ME response (in flexural state) (Figure 6(d)). We believe that an optimized combination of magnetostrictive layer thickness (demagnetization/magnetic flux concentration), mismatch strain, and laminate configuration will yield a highly enhanced self-biased ME response in the composite. To elucidate the underlying mechanisms responsible for the SME behavior, domain-level phase-field modeling and computer simulation were employed. The phase-field model for the ME composites developed in our previous work, which explicitly addresses the domain-level strain-mediated coupling between magnetization and polarization, was adopted.32,33 The influence of demagnetization and mismatch strain on the shape of the M-H loop and magnitude of remanent magnetization (M r ) and magnetic susceptibility (χ r ), which directly determines the SME behavior, were investigated. Results of the simulations indicate that the PZT/Ni heterostructures with 200 μm-thick Ni substrate (Nd = 0.026) show reduced magnetic hysteresis and a lower Mr, thereby displaying a smaller αME at zero magnetic bias, as shown in Figure 6(c). Decreasing the Ni foil thickness to 100 μm (Nd = 0.014) was 11023

DOI: 10.1021/acsami.7b16706 ACS Appl. Mater. Interfaces 2018, 10, 11018−11025

Research Article

ACS Applied Materials & Interfaces

The laser-induced crystallization of PZT film significantly contributed to the improvement of dielectric, ferroelectric, and ME properties of the PZT/Ni heterostructure. The laserannealed PZT film on Ni exhibited a highly enhanced and fully self-biased ME coupling of 3.15 V/cm·Oe, which is the highest among all the reported values in the literature for similar systems. The demagnetization in the Ni foil, mismatch strain, and flexural moment in the PZT/Ni bilayer composite were found to greatly influence its self-biased ME response.

found to increase the magnetic hysteresis as well as the SME response (Figure 6(b)). However, a further decrease in Ni foil thickness (50 μm, Nd = 0.008) diminished the SME coupling due to the decreased magnetic susceptibility χr (curve slope) at zero bias, despite an improved magnetic hysteresis and larger Mr (Figure 6(a)). These results indicate that there exists an optimal thickness of the Ni foil to realize the full SME response in the PZT/Ni composite. In the laminated PZT/Ni ME composite system, the total mismatch strain between the two layers can be determined by εmis = εth − εp, where εth is the thermal mismatch strain (resulting from different thermal expansion coefficients of the two layers and depends on the temperature change during the LA treatment), and εp is the poling mismatch strain (caused by the deformation of PZT layer due to the electrical poling). In an appropriate approximation, a planar mismatch strain of εmis = 0.003 was considered in our modeling, which corresponds to an experimental deposition/annealing temperature of 600−700 °C. Two representative configurations of the laminate composites were considered in modeling of the mismatch strain effect: a flexural composite corresponding to bilayer PZT/Ni that could bend under the mismatch strain and a nonflexural composite corresponding to trilayer Ni/PZT/Ni (and multilayer structures) that is rigid and could not bend. Figure 6(d,e) shows the corresponding simulation results for these two types of composites with the same demagnetization factor Nd = 0.008 (50 μm-thick Ni). A significant difference was observed between the M-H loops as well as SME behaviors of the two composites, which might have resulted from different stress/ strain distributions in the Ni layer. For the non-flexural composite, a homogeneous in-plane tensile stress is distributed within the Ni layer, and stripe domain structures were formed in the Ni layer. The positive stress-induced perpendicular uniaxial anisotropy and the negative magnetostriction of Ni (λij = −40 ppm) significantly softened the M-H hysteresis and resulted in a nearly zero Mr, thereby resulting in almost nil SME response (Figure 6(e)). For the flexural composite, both tensile (near the PZT/Ni interface) and compressive (far from the interface) stress are distributed along the length direction in the Ni layer due to bending, resulting in a mixing of in-plane (length direction) and perpendicular (thickness direction) magnetizations to produce non-zero magnetization remanence and thus a larger SME response (Figure 6(d)). These above findings suggest that the position and magnitude of maximum αME, and therefore the SME response could be tailored by tuning demagnetization factor, mismatch strain, and flexural moment in the PZT/Ni heterostructure. An optimum combination of these factors should be considered to realize a fully self-biased ME response in the composite. It is interesting to see that both the AD and LA PZT/Ni samples displayed fully self-biased ME coupling, which implies that the crystallinity of the PZT film or thermal stress mismatch does not critically affect the SME behavior of the PZT/Ni heterostructure.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsami.7b16706. Phase-field modeling of laminate PZT/Ni ME composites (PDF)

■

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. *E-mail: [email protected]. ORCID

Hyun-Cheol Song: 0000-0001-5563-9088 Yoon Seok Oh: 0000-0001-8233-1898 Jungho Ryu: 0000-0002-4746-5791 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This research work was supported by the National Research Foundation of Korea (grant no. NRF-2016R1A2B4011663), Korea Institute of Materials Science (KIMS) internal R&D program (grant no. PNK5061), and the U.S. Office of Naval Research Global (grant no. N62909-16-1-2135). D.M. acknowledges support from Office of Basic Energy Science, Department of Energy (grant no. DE-FG02-06ER46290). S.P. acknowledges support from Office of Naval Research (grant no. N00014-161-3043). Y.S.O. was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT & Future Planning (NRF-2015R1C1A1A01055964 and NRF2016K1A3A7A09005338).

■

REFERENCES

(1) Nan, C.-W.; Bichurin, M. I.; Dong, S.; Viehland, D.; Srinivasan, G. Multiferroic magnetoelectric composites: Historical perspective, status, and future directions. J. Appl. Phys. 2008, 103, 031101. (2) Zhou, Y.; Kim, J.-W.; Dong, S.; Priya, S.; Wang, J.; Ryu, J. Applications of Multiferroic Magnetoelectric Composites. In Multiferroic Materials; Wang, J., Eds.; CRC Press: Boca Raton, FL, 2016; Ch. 8, p215. (3) Palneedi, H.; Annapureddy, V.; Priya, S.; Ryu, J. Status and Perspectives of Multiferroic Magnetoelectric Composite Materials and Applications. Actuators 2016, 5, 9. (4) Hu, J.-M.; Nan, T.; Sun, N. X.; Chen, L.-Q. Multiferroic magnetoelectric nanostructures for novel device applications. MRS Bull. 2015, 40, 728. (5) Ortega, N.; Kumar, A.; Scott, J. F.; Katiyar, R. S. Multifunctional magnetoelectric materials for device applications. J. Phys.: Condens. Matter 2015, 27, 504002. (6) Ryu, J.; Priya, S.; Uchino, K.; Kim, H.-E. Magnetoelectric Effect in Composites of Magnetostrictive and Piezoelectric Materials. J. Electroceram. 2002, 8, 107.

4. CONCLUSION In summary, a magnetoelectric heterostructure of PZT/Ni was fabricated by depositing PZT thick film on thin flexible Ni foil. Deposition of the PZT film at room temperature using granule spray in vacuum process combined with localized annealing of the film through laser radiation provided a feasible fabrication approach toward overcoming the issues related to the high temperature processing of ceramic films on metal substrates. 11024

DOI: 10.1021/acsami.7b16706 ACS Appl. Mater. Interfaces 2018, 10, 11018−11025

Research Article

ACS Applied Materials & Interfaces (7) Chopra, I.; Sirohi, J. Smart structures theory; Cambridge University Press, New York, 2013; Ch. 6, p 604. (8) Srinivasan, G. Magnetoelectric characterization techniques. In Composite Magnetoelectrics; Srinivasan, G.; Priya, S.; Sun, N. X., Eds.; Woodhead Publishing: Cambridge, UK, 2015; Ch. 2, p 42. (9) Lage, E.; Kirchhof, C.; Hrkac, V.; Kienle, L.; Jahns, R.; Knöchel, R.; Quandt, E.; Meyners, D. Exchange biasing of magnetoelectric composites. Nat. Mater. 2012, 11, 523. (10) Zhou, Y.; Maurya, D.; Yan, Y.; Srinivasan, G.; Quandt, E.; Priya, S. Self-Biased Magnetoelectric Composites: An Overview and Future Perspectives. Energy Harvest. Syst. 2015, 3, 1. (11) Sreenivasulu, G.; Mandal, S. K.; Bandekar, S.; Petrov, V. M.; Srinivasan, G. Low-frequency and resonance magnetoelectric effects in piezoelectric and functionally stepped ferromagnetic layered composites. Phys. Rev. B: Condens. Matter Mater. Phys. 2011, 84, 144426. (12) Zhou, Y.; Yang, S. C.; Apo, D. J.; Maurya, D.; Priya, S. Tunable self-biased magnetoelectric response in homogenous laminates. Appl. Phys. Lett. 2012, 101, 232905. (13) Zhou, Y.; Priya, S. Near-flat self-biased magnetoelectric response in geometry gradient composite. J. Appl. Phys. 2014, 115, 104107. (14) Silva, M.; Reis, S.; Lehmann, C. S.; Martins, P.; LancerosMendez, S.; Lasheras, A.; Gutiérrez, J.; Barandiarán, J. M. Optimization of the Magnetoelectric Response of Poly(vinylidene fluoride)/Epoxy/ Vitrovac Laminates. ACS Appl. Mater. Interfaces 2013, 5, 10912. (15) Lasheras, A.; Gutiérrez, J.; Barandiarán, J. M. Quantification of size effects in the magnetoelectric response of metallic glass/PVDF laminates. Appl. Phys. Lett. 2016, 108, 222903. (16) Wang, Y.; Li, M.; Hasanyan, D.; Gao, J.; Li, J.; Viehland, D. Geometry-induced magnetoelectric effect enhancement and noise floor reduction in Metglas/piezofiber sensors. Appl. Phys. Lett. 2012, 101, 092905. (17) Patil, D. R.; Zhou, Y.; Kang, J.-E.; Sharpes, N.; Jeong, D.-Y.; Kim, Y.-D.; Kim, K. H.; Priya, S.; Ryu, J. Anisotropic self-biased dualphase low frequency magneto-mechano-electric energy harvesters with giant power densities. APL Mater. 2014, 2, 046102. (18) Tang, C.; Lu, C. Strong self-biased magnetoelectric charge coupling in a homogenous laminate stack for magnetic sensor. J. Alloys Compd. 2016, 686, 723. (19) Ryu, J.-G.; Jeon, S.-J. Magnetoelectric Characteristics of Layered Ni-PZT-Ni, Co, Fe Composites for Magnetic Field Sensor. J. Korean Inst. Electr. Electron. Mater. Eng. 2015, 28, 92. (20) Zhou, Y.; Apo, D. J.; Priya, S. Dual-phase self-biased magnetoelectric energy harvester. Appl. Phys. Lett. 2013, 103, 192909. (21) Gupta, R.; Tomar, M.; Gupta, V.; Zhou, Y.; Chopra, A.; Priya, S.; Bhalla, A. S.; Guo, R. Giant Magnetoelectric Effect in PZT Thin Film Deposited on Nickel. Energy Harvesting and Syst. 2016, 3, 181. (22) Feng, M.; Wang, J.-j.; Hu, J.-M.; Wang, J.; Ma, J.; Li, H.-B.; Shen, Y.; Lin, Y.-H.; Chen, L.-Q.; Nan, C.-W. Optimizing direct magnetoelectric coupling in Pb(Zr,Ti)O3/Ni multiferroic film heterostructures. Appl. Phys. Lett. 2015, 106, 072901. (23) Ryu, J.; Carazo, A. V.; Uchino, K.; Kim, H.-E. Magnetoelectric Properties in Piezoelectric and Magnetostrictive Laminate Composites. Jpn. J. Appl. Phys. 2001, 40, 4948. (24) Im, I.-H.; Chung, K.-H.; Kim, D.-H. Dependence of Annealing Temperature on Properties of PZT Thin Film Deposited onto SGGG Substrate. Trans. Electr. Electron. Mater. 2014, 15, 253. (25) Park, C.-K.; Kang, D.-K.; Lee, S.-H.; Kong, Y.-M.; Jeong, D.-Y. Phase Evolution and Electrical Properties of PZT Films by AerosolDeposition Method. J. Korean Inst. Electr. Electron. Mater. Eng. 2017, 30, 541. (26) Jung, J.; Annapureddy, V.; Hwang, G.-T.; Song, Y.; Lee, W.; Kang, W.; Ryu, J.; Choi, H. 31-mode piezoelectric micromachined ultrasonic transducer with PZT thick film by granule spraying in vacuum process. Appl. Phys. Lett. 2017, 110, 212903. (27) Kambale, R. C.; Ryu, J.; Patil, D. R.; Chai, Y. S.; Kim, K. H.; Yoon, W.-H.; Jeong, D.-Y.; Park, D.-S.; Kim, J.-W.; Choi, J.-J.; Ahn, C.W. Colossal magnetoelectric response of PZT thick films on Ni substrates with a conductive LaNiO3 electrode. J. Phys. D: Appl. Phys. 2013, 46, 092002.

(28) Park, D. S.; Kim, J.-W.; Ryu, J.; Yoon, W. H.; Choi, J. J.; Hahn, B. D. Granules of a Brittle Material for Room Temperature Granule Spray in Vacuum, and Method for Forming a Coating Film Using Same. US Patent US20130295272A1, 2013. (29) Kim, Y.; Ahn, C.-W.; Choi, J.-J.; Ryu, J.; Kim, J.-W.; Yoon, W.H.; Park, D.-S.; Yoon, S.-Y.; Ma, B.; Hahn, B.-D. Next Generation Ceramic Substrate Fabricated at Room Temperature. Sci. Rep. 2017, 7, 6637. (30) Palneedi, H.; Choi, I.; Kim, G.-Y.; Annapureddy, V.; Maurya, D.; Priya, S.; Kim, J.-W.; Lee, K. J.; Choi, S.-Y.; Chung, S.-Y.; Kang, S.-J. L.; Ryu, J. Tailoring the Magnetoelectric Properties of Pb(Zr,Ti)O3 Film Deposited on Amorphous Metglas Foil by Laser Annealing. J. Am. Ceram. Soc. 2016, 99, 2680. (31) Palneedi, H.; Maurya, D.; Kim, G.-Y.; Annapureddy, V.; Noh, M.-S.; Kang, C.-Y.; Kim, J.-W.; Choi, J.-J.; Choi, S.-Y.; Chung, S.-Y.; Kang, S.-J. L.; Priya, S.; Ryu, J. Unleashing the Full Potential of Magnetoelectric Coupling in Film Heterostructures. Adv. Mater. 2017, 29, 1605688. (32) Ma, F. D.; Jin, Y. M.; Wang, Y. U.; Kampe, S. L.; Dong, S. Phase field modeling and simulation of particulate magnetoelectric composites: Effects of connectivity, conductivity, poling and bias field. Acta Mater. 2014, 70, 45−55. (33) Geng, L. D.; Yan, Y.; Priya, S.; Wang, Y. U. Theoretical model and computer simulation of Metglas/PZT magnetoelectric composites for voltage tunable inductor applications. Acta Mater. 2017, 140, 97− 106. (34) Li, Z.; Gao, Y.; Yang, B.; Lin, Y.; Yu, R.; Nan, C.-W. Influence of Stress and Orientation on Magnetoelectric Coupling of Pb(Zr,Ti)O3− CoFe2O4 Bilayer Films. J. Am. Ceram. Soc. 2011, 94, 1060. (35) Kim, J.; Yang, S. A.; Choi, Y. C.; Han, J. K.; Jeong, K. O.; Yun, Y. J.; Kim, D. J.; Yang, S. M.; Yoon, D.; Cheong, H.; Chang, K.-S.; Noh, T. W.; Bu, S. D. Ferroelectricity in Highly Ordered Arrays of UltraThin-Walled Pb(Zr,Ti)O3 Nanotubes Composed of Nanometer-Sized Perovskite Crystallites. Nano Lett. 2008, 8, 1813. (36) Ryu, J.; Priya, S.; Park, C.-S.; Kim, K.-Y.; Choi, J.-J.; Hahn, B.-D.; Yoon, W.-H.; Lee, B.-K.; Park, D.-S.; Park, C. Enhanced domain contribution to ferroelectric properties in freestanding thick films. J. Appl. Phys. 2009, 106, 024108. (37) Palneedi, H.; Maurya, D.; Kim, G.-Y.; Priya, S.; Kang, S.-J. L.; Kim, K.-H.; Choi, S.-Y.; Ryu, J. Enhanced off-resonance magnetoelectric response in laser annealed PZT thick film grown on magnetostrictive amorphous metal substrate. Appl. Phys. Lett. 2015, 107, 012904. (38) Wan, J. G.; Liu, J.-M.; Chand, H. L. W.; Choy, C. L.; Wang, G. H.; Nan, C. W. Giant magnetoelectric effect of a hybrid of magnetostrictive and piezoelectric composites. J. Appl. Phys. 2003, 93, 9916. (39) Park, C.-S.; Khachaturyan, A.; Priya, S. Giant magnetoelectric coupling in laminate thin film structure grown on magnetostrictive substrate. Appl. Phys. Lett. 2012, 100, 192904. (40) Fang, Z.; Lu, S. G.; Li, F.; Datta, S.; Zhang, Q. M.; El Tahchi, M. Enhancing the magnetoelectric response of Metglas/polyvinylidene fluoride laminates by exploiting the flux concentration effect. Appl. Phys. Lett. 2009, 95, 112903. (41) Gao, J.; Gray, D.; Shen, Y.; Li, J.; Viehland, D. Enhanced dc magnetic field sensitivity by improved flux concentration in magnetoelectric laminates. Appl. Phys. Lett. 2011, 99, 153502. (42) Chu, Z.; Shi, H.; Shi, W.; Liu, G.; Wu, J.; Yang, J.; Dong, S. Enhanced Resonance Magnetoelectric Coupling in (1−1) Connectivity Composites. Adv. Mater. 2017, 29, 1606022. (43) Chen, D. X.; Pardo, E.; Sanchez, A. IEEE Trans. IEEE Trans. Magn. 2002, 38, 1742−1752.

11025

DOI: 10.1021/acsami.7b16706 ACS Appl. Mater. Interfaces 2018, 10, 11018−11025