- Email: [email protected]
Prospects for nanostructured multiferroic composite materials
Available online at www.sciencedirect.com
ScienceDirect Scripta Materialia 74 (2014) 38–43 www.elsevier.com/locate/scriptamat
Viewpoint Paper
Prospects for nanostructured multiferroic composite materials Jennifer S. Andrew,⇑ Justin D. Starr and Maeve A.K. Budi Department of Materials Science and Engineering, University of Florida, Gainesville, FL 32611, USA Available online 29 September 2013
Abstract—Magnetoelectric multiferroic materials have the potential to transform a range of applications, including tunable microelectronics and multiphase memories. However, the coexistence of ferromagnetism and ferroelectricity in single-phase materials is quite rare, driving the development of composite multiferroic materials. In these composites, the coupling arises from strain transfer across a shared interface between the ferroelectric and ferromagnetic phase. This viewpoint article highlights recent work as well as future challenges in the synthesis and characterization of nanostructured multiferroic materials. Ó 2013 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved. Keywords: Magnetoelectrics; Nanocomposites; Multiferroics; Ferrites; Electrospinning
1. Introduction Multiferroic materials represent a novel class of material where multiple types of ferroic ordering coexist, and their coupling can lead to additional ordering parameters. These, in turn, can be leveraged to form new multifunctional devices. The ferroic orderings are ferroelectric, ferromagnetic, and ferroelastic and these materials exhibit a spontaneous polarization, magnetization or strain, which can be further controlled by an applied electric, magnetic, or stress field, respectively. In a multiferroic, when multiple types of ferroic ordering are coupled, additional functionalities can arise, including magnetoelectric, piezoelectric and magnetoelastic behaviors. Here, we are interested in magnetoelectric multiferroics that combine ferroelectricity with ferromagnetism. These materials will be the focus of this article, and herein the term multiferroic will refer solely to materials with magnetoeletric behavior. Single-phase multiferroics, although rare, do exist, but they tend to have both a low permittivity and a low permeability. This rare occurrence of single-phase multiferroics is a result of the often seemingly contradictory material requirements for ferroelectricity and ferromagnetism. Ferroelectric materials must be electrically insulating and require a non-centrosymmetric unit cell, while most ferromagnetic materials tend to be metals. A more complete discussion of these contrasting properties has been reviewed by Hill [1,2]. The single-phase multiferroics that do exist are classified into one of two types, designated as type 1 or type 2, based on the origin of multiferroicity. In type 1
⇑ Corresponding author. Tel.: +1 3528463345; e-mail: [email protected]fl.edu
multiferroics the sources for ferroelectricity and ferromagnetism are distinct and independent from one another [3–7]. In type 2 multiferroics the multiferroic behavior arises from magnetic ordering [8]. Although a rich area of research, single-phase multiferroics are plagued by low ordering temperatures, limiting their applications in devices. Single-phase multiferroics have been discussed thoroughly in several outstanding recent review articles and will not be further discussed here [9–13]. To overcome the limitations of these single-phase multiferroics, composite or multi-phase multiferroic materials have been developed. Composite multiferroics are typically heterostructures consisting of a magnetostrictive and a piezoelectric phase. These heterostructures can be readily prepared in a range of composite architectures defined by the connectivity between the two phases by a variety of methods, ranging from thin film to solid-state approaches [14] (Figure 1). For example, a 0-3 composite describes a particulate-based composite where zero-dimensional (0-D) particles are dispersed within a 3-D matrix. Particulate-based composites (0-3) [15], pillars within a 3-D matrix (1-3) [16] and layered heterostructures (2-2) [17–24] represent the most common geometries of magnetoelectric heterostructures (Fig. 1a). In these multiferroic composites the magnetoelectric coupling is mechanical in nature and is based on the fact that piezoelectric and magnetostrictive materials undergo a shape change in an applied electric or magnetic field, respectively. The magnetoelectric effect in these composites can be described by the direct magnetoelectric effect, which is the product of the magnetostrictive effect (magnetic/mechanical) and the piezoelectric effect (mechanical/electrical) [25].
1359-6462/$ - see front matter Ó 2013 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.scriptamat.2013.09.023
J. S. Andrew et al. / Scripta Materialia 74 (2014) 38–43
Figure 1. (a) Composite connectivities of composites prepared by traditional routes and (b) connectivities that could be realized with the development of nanostructured building blocks.
magnetic mechanical ð1Þ mechanical electric The figure of merit for these materials is the magnetoelectric coefficient, aE, which is determined experimentally by measuring the change in electric field, dE, generated by applying an ac magnetic field, dH, to a biased sample, Direct ME effect ¼
dE ð2Þ dH Although these magnetoeletric heterostructures have high theoretically predicted values for aE, actual experimental values tend to be much lower [20,26–31]. The fact that these experimental coupling values are lower than predicted can be attributed to a variety of factors including the formation of cracks or impurity phases at the interface between the two phases. These heterostructures are also challenged by the low resistivity ferrite phase and substrate effects, which can mechanically clamp the sample [16]. Because this coupling involves mechanical coupling across the interface between the magnetostrictive and piezoelectric phase, it follows that one route to enhance the magnetoelectric coupling is by maximizing the interfacial area between the two phases [32]. One way to accomplish this is to use nanoscale materials due to their large surface to volume ratio. The first example of coupling in multiferroic nanostructures was demonstrated by Zheng et al. [16], who embedded CoFe2O4 nanopillars epitaxially in a BaTiO3 thin film matrix. However, it is worth noting that here the CoFe2O4 rods are nanoscale in diameter only. Since this seminal work, much effort has been done to develop new nanostructured building blocks for the formation of the next generation of multiferroic materials. The focus of this viewpoint article will be to describe and review the development of nanostructured multiferroic building blocks that can be assembled into higher order multiferroic structures and devices, while also addressing the needs to develop and utilize novel characterization methods to fully study these systems. aE ¼
2. Synthesis and characterization of nanostructured multiferroic materials 2.1. Zero-dimensional (nanoparticles) One of the challenges in fabricating multiferroic composites with magnetoelectric coupling values approaching
39
the theoretically predicted values is the low resistivity of the ferrite phase. In many of the theoretical models the optimum volume fraction of ferrite is at high volume fractions (0.5–0.9) [26]. However, at such high loadings the leaky ferrite phase makes it challenging to pole the material. Therefore, much effort has been made to isolate the leaky ferrite phase within a ferroelectric matrix (0-3 connectivity). This can be accomplished via a multitude of routes with a range of synthetic complexities. The simplest methods involve mechanically mixing particles followed by densification via solid-state methods [33–37], or mixing solutions of ferroelectric and piezoelectric particles followed by spin-coating and calcination [38], and molten-salt based synthetic routes [39,40]. Thin film approaches such as pulsed laser deposition have also been used to fabricate particulate composites [41]. Novel one-pot synthesis strategies have also been developed to more fully control the synthesis and microstructure of composite multiferroics [33,42–46]. However, using many of these routes, it remains a challenge to isolate the ferrite phase at the high volume fractions required to obtain high performance magnetoelectric composites. In many of these routes the device performance drops off at volume fractions of ferrite between 0.2 and 0.5, corresponding with percolation of the ferrite phase [47–49]. Core–shell magnetoelectric particles have been synthesized to address this challenge of isolating the ferrite phase while maintaining a high overall volume fraction of ferrite. These novel particles consist of a ferrite core (CoFe2O4, Fe3O4, Ni0.5Zn0.5Fe2O4) surrounded by (i.e. electrically isolated by) a ferroelectric shell [51– 53,55–59]. The ferrite core is typically synthesized via either a hydrothermal or co-precipitation method, and is subsequently coated using sol–gel methods with a BaTiO3 shell. These particles are then calcined and/or sintered at elevated temperatures (650–1300 °C). Duong et al. have reported the longitudinal and transverse maximum magnetoelectric (ME) coupling coefficients as 3.4 mV cm1 Oe1 and 2 mV cm1 Oe1 respectively for sintered pellets of these core-shell nanoparticles [52,56,57]. However, they found that by varying pressure, sintering temperature and sintering duration, these values were further increased to 3.53 mV cm1 Oe1 and 2.23 mV cm1 Oe1. Duong et al. [57] also synthesized a structure using barium titanate as the core and cobalt ferrite as the shell. The barium titanate was synthesized using sol–gel methods, then suspended in a solution containing a chelating agent and the elements to form cobalt ferrite. As expected, in these materials where the CoFe2O4 was not isolated in the core, the ME coefficient was found to be 4.2 times lower than a comparable sample with cobalt ferrite as the core [57]. All the synthesis examples outlined above require calcining steps to produce crystalline tetragonal perovskite barium titanate from the amorphous barium titanium sol, which is needed for piezoelectric behavior. The exact temperatures varied from 650 to 1000 °C. All of the reported ME measurements were made on samples that underwent additional sintering for further densification [53,52,56–59]. Though some groups use transmission electron microscopy (TEM) and SEM to confirm the core–shell particles, the particles are agglomerated because of the high temperature processing [51,55]. One of
40
J. S. Andrew et al. / Scripta Materialia 74 (2014) 38–43
the challenges is to develop synthesis procedures that allow for the formation of isolated multiferroic particles that are free of agglomerates. These particles could then be assembled into higher order structures using colloidal processing routes. Additionally, the application of novel characterization methods to observe multiferroic behavior on a single nanostructured particle will provide a means to more fundamentally understand the magnetoelectric coupling and structure–property relationships in these systems. 2.2. One-dimensional (fibers) One-dimensional multiferroic nanostructures such as nanofibers provide another route to bulk multiferroic materials. Ideal nanofibers, free of substrate based constraints, have been predicted to display mulitferroic coupling that is an order of magnitude greater than that found in ideal thin films [60]. Further, composite nanofibers can be synthesized in a number of configurations by varying the orientation of the piezoelectric and magnetostrictive phases. These morphologies can be used as building blocks to enable additional connectivities in resultant bulk structures, including 1-1 (Fig. 1b). Multiferroic nanofibers can be synthesized via electrospinning [61,62]. Several groups have shown that, if two compatible phases are simultaneously electrospun, it is possible to produce biphasic structures as long as the phases remain segregated until the point of electrification [63–65]. This process relies on the fact that the kinetics of solvent evaporation and hydrolysis and condensation occur faster than mixing between the two miscible phases. Electrospinning multiphase structures depends on carefully matching the properties of precursor solutions, as similar solutions are most likely to cohesively jet. This technique can be used to form multiferroic fibers with a range of microstructures (Fig. 2). The simplest method involves synthesizing a nanofiber with a random distribution of ferroelectric and magnetostrictive phases. This can be achieved by mixing ceramic precursors for each phase together, forming a blended polymer solution. When electrospun and calcined, such a solution will produce biphasic nanofibers with a random distribution of grains of each phase (Fig. 2a). Randomly mixed multiferroic fibers have been synthesized for systems composed of lead zirconate titanate and cobalt or nickel ferrite [66–69]. Multiferroic nanofibers can also be synthesized in a coaxial or core–shell geometry. These nanofibers maintain a large contact area between phases, while segregating the ferroelectric and ferromagnetic phases. Core–shell nanofibers have been produced via electrospinning (Fig. 2b) for lead zirconate titanate with CoFe2O4 [70,71]. Core–shell nanofibers and nanotubes have also been fabricated via membrane template approaches, where a polycarbonate membrane was used as a template for the synthesis of CoFe2O4 nanotubes, which were subsequently added to a BaTiO3 precursor solution and calcined at 780 °C to form a BaTiO3 shell [54]. Both randomly dispersed and core–shell nanofibers can be assembled into bulk structures with connectivities similar to traditional methods, including 0-3 or 1-3, for randomly distributed or core–shell structures,
Figure 2. Electrospinning can produce multiferroic fibers in a range of geometries, including (a) random, (b) core–shell or (c) Janus-type arrangements between the two phases. TEM images reproduced with permission from Refs. [61,64,66].
respectively. Further, it is extraordinarily difficult to measure the magnetoelectric properties of a single isolated nanofiber. As a result, many reports of these mutiferroic nanofibers do not attempt to demonstrate multiferroic behavior or quantify the magnetoelectric coupling coefficient. Piezoresponse force microscopy (PFM) has been used to measure magnetoelectric coupling along a single nanofiber [71]. Using this method, Xie et al. [60] measured a magnetoelectric coupling coefficient of 2.95 104 mV cm1 Oe1 for CoFe2O4– Pb(Zr0.52Ti0.48)O3 core–shell nanofibers, which, as theoretically predicted, is two orders of magnitude greater than thin films of similar composition [70]. Recently, a new morphology of multiferroic nanofibers has been produced: the so-called Janus-type or hemispherical composite nanofiber (Fig. 2c). In these fibers, ferroelectric and ferromagnetic phases are arranged along the length of a nanofiber, with a shared interface running down the longitudinal axis [72]. These fibers retain external access to both phases as well as the interface, making them well suited to modular structures. For example, these Janus-type fibers can be assembled into composites with a 1-1 connectivity (Fig. 1b), which is not readily realized by conventional approaches. Thus far, these fibers have only been demonstrated for the barium titanate–cobalt ferrite system, but they have been further extended to trilayer structures [73]. Indirect evidence of magnetoelectric coupling has been shown in these Janus-type fibers by measuring the magnetization as a function of temperature. At the Curie temperature of barium titanate, a marked change in the magnetization of cobalt ferrite is present as the cobalt ferrite is strained when barium titanate undergoes a structural transition from tetragonal to cubic. 2.3. Two-dimensional (films) Layered structures formed from 2-2 laminate composites can overcome the leakage problem due to the low resistivity ferrite phase. To form nanostructured layered composites many efforts have focused on thin films. By using techniques such as pulsed laser deposition (PLD) or atomic layer deposition (ALD), precise heterostructures can be obtained. By carefully choosing
J. S. Andrew et al. / Scripta Materialia 74 (2014) 38–43
a substrate and growing layers of piezoelectric and magnetostrictive materials via PLD or similar techniques, it is possible to carefully control the alignment of crystallographic planes and achieve an epitaxial relationship across the shared interface. These epitaxial interfaces may reduce some of the losses inherent in bulk composite structures and lend themselves well to accurate study, as parameters can be carefully controlled and subsequently verified by simulation [74]. However, these routes are not without challenges because in many instances the substrates clamp the mechanical coupling, limiting magnetoelectric response. Biphasic composite film structures have been synthesized for a number of material systems, including barium titanate–cobalt ferrite, lead zirconate titanate– cobalt ferrite and barium titanate–lanthanum strontium manganite, as well as a variety of other systems [16,18,75–80]. Chang et al. [81] synthesized a variety of cobalt ferrite–barium titanate composites with a composition spread to determine the effects of film thickness on resultant properties. They grew a 300 nm thick film that consisted of wedges of cobalt ferrite and barium titanate. The thickness of each wedge ranged from as little as one unit cell to tens of unit cells. Using this approach, Chang et al. found that there was a region towards the middle of the spread with both a high dielectric constant and ferromagnetic behavior. Despite the large body of literature on the synthesis of magnetoelectric films of different compositions, comparatively very few studies have published direct measurements of the magnetoelectric coupling coefficient, as it can be difficult to make these measurements at the nanoscale. In the few cases where these values have been reported, the results have varied. Some lead zirconate titanate–nickel ferrite structures show magnetoelectric coupling values (16 mV cm1 Oe1) [41] that fall short of their bulk counterparts, while other barium titanate–nickel ferrite structures are more commensurate with accepted bulk values (12.1 and 7.9 mV cm1 Oe1 for out-of-plane and in-plane coupling, respectively) [80]. However, with careful refinement of the structure, converse magnetoelectric coupling coefficients (a = l0DM/DE) can be as large as those of Terfenol-D laminate systems (2.3 107 s m1), and provide evidence of the giant magnetoelectric effect [79]. As in the case for 1-D magnetoelectric structures, many groups verify multiferroic properties via indirect measurement (Fig. 3): either by observing a change in magnetization at the ferroelectric Curie temperature or by examining the changes in dielectric properties at the ferromagnetic or antiferromagnetic Curie/Nee´l temperatures (Fig. 3a) [16,72,73]. Other groups examine how polarization or magnetization hysteresis loops change under the presence of an applied magnetic or electric field, respectively (Fig. 3b) [82,83]. While thin films have received considerable attention, they are not without drawbacks: techniques like PLD and ALD can be slow and expensive. Further, the use of a substrate can reduce the available degrees of freedom and create clamping effects, lessening strain transfer and negating potential magnetoelectric coupling [60]. Some multilayer thin films fail to show indirect evidence of coupling due to these limitations [16].
41
Figure 3. Indirect methods for determining the existence of magnetoelectric coupling: (a) observation of anomalous magnetization behavior at the ferroelectric Curie temperature where the piezoelectric material undergoes a structural change leading to a change in the magnetization of the magnetostrictive material through strain coupling (Reprinted with permission from Ref. [16]) and (b) Measurement of changes in the electronic polarization loop of a magnetoelectric composite under an applied magnetic field. Reprinted with permission from Ref. [75].
Thin films are also limited to 2-2, 1-3 and 0-3 type connectivities [14,84]. 3. Looking forward 3.1. Characterization methods Bulk multiferroic composites typically exhibit magnetoelectric coupling values that are significantly lower than theoretically predicted values. For the case of BaTiO3–CoFe2O4 composites typical experimental values are 1.5–130 mV cm1 Oe1 [26,53,55,57,50], whereas, depending on the assumptions made, theory predicts values that range from 1.5 to 5 V cm1 Oe1 [26,27]. These decreased values can be attributed to the presence of cracks, impurity phases or leakage due to the low resistivity ferrite phase. Some of these limitations can be overcome by the development and characterization of the novel nanostructured multiferroic building blocks discussed in this article and their assembly into bulk structures. However, to date, studies of magnetoelectric coupling at the nanoscale are limited, but are necessary to develop composite multiferroics with enhanced coupling. Piezoeresponse force microscopy (PFM) has emerged as a promising technique for measuring the properties of ferroic materials, including measuring the magnetoelectric coefficients of nanostructured materials [70,85,86]. Pan et al. [87] used electron tomography in conjunction with vector-PFM (v-PFM) to study CoFe2O4–PbTiO3 nanocomposites prepared via metal organic chemical vapor deposition. By combining these two methods, they were able to determine the 3-D structure of the nanocomposite at the nanoscale via electron tomography and correlate it with the 3-D polarization domains, determined with v-PFM. Futhermore, this data could be correlated to the magnetic properties of the material, including hysteresis and anisotropy data. The application of these tools to nanostructured multiferroic materials will allow for a thorough understanding of structure–property relationships in these complex heterostructures. Nanoscale structure–property relationships for magnetoelectric coupling behavior can then
42
J. S. Andrew et al. / Scripta Materialia 74 (2014) 38–43
be used to guide the assembly of these into bulk materials with enhanced properties. 3.2. Idealized structures The generation of new nanostructured multiferroic materials allows the fabrication of composites with a range of connectivities, including new connectivities not readily synthesized by traditional routes. For example, the Janus-type nanofibers provide an anisotropic structure that can be leveraged for self-assembly into more complex structures, including 1-1 type-connectivity (Fig. 1b) [88–90], while isolated magnetoelectric particles can be assembled into multiferroic composites with unique architectures using colloidal processing routes. In many cases, to obtain a multiferroic device it will be necessary to consolidate these structures into dense composites. In order to maintain the nanostructure of the building blocks, fast firing approaches can be utilized, including microwave sintering and spark plasma sintering [91–94]. 4. Summary Much research has focused on the development of nanostructured multiferroic materials. This body of research includes the development of many novel synthetic routes as well as the thorough structural characterization of the materials developed. However, in many cases the magnetoelectric properties of these materials are not fully described. Therefore, more work needs to be done to develop and employ tools capable of measuring the magnetoelectric properties of these nanostructured elements. These new nanostructured elements can then be assembled into unique architectures, and their structures can be preserved using fast firing techniques. By understanding structure–property relationships at the nanoscale, novel magnetoelectric materials with enhanced properties can be realized. Acknowledgement This work was supported by the National Science Foundation through a CAREER Award (DMR1150665). References [1] N. Hill, J. Phys. Chem. B 104 (2000) 6694. [2] N. Hill, Annu. Rev. Mater. Sci. 32 (2002) 1. [3] A. Moreira dos Santos, S. Parashar, A. Raju, Y. Zhao, A. Cheetham, C. Rao, Solid State Commun. 122 (2002) 49. [4] J. Wang, J. Neaton, H. Zheng, V. Nagarajan, S. Ogale, B. Liu, D. Viehland, V. Vaithyanathan, D. Schlom, U. Waghmare, N. Spaldin, K. Rabe, M. Wuttig, R. Ramesh, Science 299 (2003) 1719. [5] R. Seshadri, N. Hill, Chem. Mater. 13 (2001) 2892.
[6] M. Fiebig, T. Lottermoser, D. Frohlich, A. Goltsev, R. Pisarev, Nature 419 (2002) 818. [7] J. Van Den Brink, D.I. Khomskii, J. Phys.: Condens. Matter 20 (2008) 434217. [8] T. Kimura, T. Goto, H. Shintani, K. Ishizaka, T. Arima, Y. Tokura, Nature 426 (2003) 55. [9] L.W. Martin, D.G. Schlom, Curr. Opin. Solid State Mater. Sci. 16 (2012) 199. [10] R. Ramesh, N. Spaldin, Nat. Mater. 6 (2007) 21. [11] N.A. Spaldin, S.-W. Cheong, R. Ramesh, Phys. Today 63 (2010) 38. [12] G. Catalan, J.F. Scott, Adv. Mater. 21 (2009) 2463. [13] L.W. Martin, S.P. Crane, Y.H. Chu, M.B. Holcomb, M. Gajek, M. Huijben, C.H. Yang, N. Balke, R. Ramesh, J. Phys.: Condens. Matter 20 (2008) 434220. [14] R. Newnham, D. Skinner, L. Cross, Mater. Res. Bull. 13 (1978) 525. [15] J. van den Boomgaard, R. Born, J. Mater. Sci. 13 (1978) 1538. [16] H. Zheng, J. Wang, S. Lofland, Z. Ma, L. MohaddesArdabili, T. Zhao, L. Salamanca-Riba, S. Shinde, S. Ogale, F. Bai, D. Viehland, Y. Jia, D. Schlom, M. Wuttig, A. Roytburd, R. Ramesh, Science 303 (2004) 661. [17] J. Ryu, S. Priya, A. Vazquez Carazo, K. Uchino, J. Am. Ceram. Soc. 84 (2001) 2905. [18] G. Srinivasan, E.T. Rasmussen, J. Gallegos, R. Srinivasan, Y. Bokhan, V. Laletin, Phys. Rev. B 64 (2001) 214408. [19] G. Srinivasan, E.T. Rasmussen, B.J. Levin, R. Hayes, Phys. Rev. B 65 (2002) 134402. [20] M. Bichurin, V. Petrov, G. Srinivasan, J. Appl. Phys. 92 (2002) 7681. [21] C. Nan, M. Li, J. Huang, Phys. Rev. B 63 (2001) 144415. [22] G. Srinivasan, E. Rasmussen, R. Hayes, Phys. Rev. B 67 (2003) 014418. [23] M. Bichurin, D. Filippov, V. Petrov, V. Laletsin, N. Paddubnaya, G. Srinivasan, Phys. Rev. B 68 (2003) 132408. [24] X. Lu, H. Li, J. Appl. Phys. 109 (2011) 074112. [25] J. Ma, J. Hu, Z. Li, C.-W. Nan, Adv. Mater. 25 (2011) 1062. [26] C.-W. Nan, Undefined author, S. Dong, D. Viehland, G. Srinivasan, J. Appl. Phys. 103 (2008) 031101. [27] J. Ryu, S. Priya, K. Uchino, H.-E. Kim, J. Electroceram. 8 (2002) 107. [28] M.I. Bichurin, V.M. Petrov, Adv. Condens Matter Phys 2012 (2012) 1. [29] M. Bichurin, V. Petrov, G. Srinivasan, Phys. Rev. B 68 (2003) 054402. [30] V.M. Petrov, G. Srinivasan, M.I. Bichurin, A. Gupta, Phys. Rev. B 75 (2007) 224407. [31] M. Avellaneda, G. Harshe´, J. Intell. Mater. Syst. Struct. 5 (1994) 501. [32] R. Liu, Y. Zhao, R. Huang, Y. Zhao, H. Zhou, J. Mater. Chem. 20 (2010) 10665. [33] A. Testino, L. Mitoseriu, V. Buscaglia, M.T. Buscaglia, I. Pallecchi, A.S. Albuquerque, V. Calzona, D. Marre, A.S. Siri, P. Nanni, J. Eur. Ceram. Soc. 26 (2006) 3031–3036. [34] J. Ryu, A. Carazo, K. Uchino, H.-E. Kim, J. Electroceram. 7 (2001) 17. [35] G. Srinivasan, C. DeVreugd, C. Flattery, V. Laletsin, N. Paddubnaya, Appl. Phys. Lett. 85 (2004) 2550. [36] A. Khamkongkaeo, P. Jantaratana, C. Sirisathitkul, T. Yamwong, S. Maensiri, Trans. Nonferrous Met. Soc. China 21 (2011) 2438. [37] S.G. Lu, Z.K. Xu, Y.P. Wang, S.S. Guo, H. Chen, T.L. Li, S.W. Or, J. Electroceram. 21 (2007) 398.
J. S. Andrew et al. / Scripta Materialia 74 (2014) 38–43
[38] A. McDannald, M. Staruch, G. Sreenivasulu, C. Cantoni, G. Srinivasan, M. Jain, Appl. Phys. Lett. 102 (2013) 122905. [39] J.-W. Nie, G.-Y. Xu, X.-J. Qi, Trans. Nonferrous Met. Soc. China 19 (2009) 1593. [40] J. Nie, G. Xu, Y. Yang, C. Cheng, Mater. Chem. Phys. 115 (2009) 400. [41] H. Ryu, P. Murugavel, J.H. Lee, S.C. Chae, T.W. Noh, Y.S. Oh, H.J. Kim, K.H. Kim, J.H. Jang, M. Kim, Appl. Phys. Lett. 89 (2006) 102907. [42] S.Q. Ren, L.Q. Weng, S.-H. Song, F. Li, J.G. Wan, M. Zeng, J. Mater. Sci. 40 (2005) 4375. [43] D. Wu, W. Gong, H. Deng, M. Li, J. Phys. D Appl. Phys. 40 (2007) 5002. [44] A.R. Iordan, M. Airimioaiei, M.N. Palamaru, C. Galassi, A.V. Sandu, C.E. Ciomaga, F. Prihor, L. Mitoseriu, A. Ianculescu, J. Eur. Ceram. Soc. 29 (2009) 2807. [45] L. Weng, Y. Fu, S. Song, J. Tang, J. Li, Scripta Mater. 56 (2007) 465. [46] J.-G. Wan, H. Zhang, X. Wang, D. Pan, J.-M. Liu, G. Wang, Appl. Phys. Lett. 89 (2006) 122914. [47] A. Lakhtakia, T. Mackay, Int. J. Electron. Commun. 58 (2004) 1. [48] C.A.F. Vaz, J. Hoffman, C.H. Anh, R. Ramesh, Adv. Mater. 22 (2010) 2900. [49] L.P. Curecheriu, M.T. Buscaglia, V. Buscaglia, L. Mitoseriu, P. Postolache, A. Ianculescu, P. Nanni, J. Appl. Phys. 107 (2010) 104106. [50] V. Corral-Flores, D. Bueno-Baques, R.F. Ziolo, Acta Mater. 58 (2010) 764. [51] M. Etier, Y. Gao, V.V. Shvartsman, A. Elsukova, J. Landers, H. Wende, D.C. Lupascu, Ferroelectrics 438 (2012) 115. [52] G.V. Duong, R. Groessinger, R.S. Turtelli, IEEE Trans. Magn. 42 (2006) 3611. [53] G.V. Duong, R.S. Turtelli, R. Groessinger, J. Magn. Magn. Mater. 322 (2010) 1581. [54] K. Raidongia, A. Nag, A. Sundaresan, C.N.R. Rao, Appl. Phys. Lett. 97 (2010) 062904. [55] V.V. Shvartsman, F. Alawneh, P. Borisov, D. Kozodaev, D.C. Lupascu, Smart Mater. Struct. 20 (2011) 075006. [56] G.V. Duong, R. Groessinger, R. Sato Turtelli, J. Magn. Magn. Mater. 310 (2007) 1157. [57] G.V. Duong, R. Groessinger, R. Sato Turtelli, J. Magn. Magn. Mater. 310 (2007) e361. [58] G.V. Duong, R. Groessinger, J. Magn. Magn. Mater. 316 (2007) e624. [59] V. Corral-Flores, D. Bueno-Baques, D. Carrillo-Flores, J.A. Matutes-Aquino, J. Appl. Phys. 99 (2006) 0803. [60] C.L. Zhang, W.Q. Chen, S.H. Xie, J.S. Yang, J.Y. Li, Appl. Phys. Lett. 94 (2009) 102907. [61] W. Sigmund, J. Yuh, H. Park, V. Maneeratana, G. Pyrgiotakis, A. Daga, J. Taylor, J. Nino, J. Am. Ceram. Soc. 89 (2006) 395. [62] S. Ramakrishna, K. Fujihara, W. Teo, T. Lim, Z. Ma, An introduction to electrospinning and nanofibers, World Scientific, New Jersey, 2005. [63] K.-H. Roh, D.C. Martin, J. Lahann, Nat. Mater. 4 (2005) 759.
43
[64] Z. Liu, D.D. Sun, P. Guo, J.O. Leckie, Nano Lett. 7 (2007) 1081. [65] S. Bhaskar, J. Hitt, S. Chang, J. Lahann, Angew. Chem. Int. Ed. 48 (2009) 4589. [66] S.H. Xie, J.Y. Li, Y.Y. Liu, L.N. Lan, G. Jin, Y.C. Zhou, J. Appl. Phys. 104 (2008) 024115. [67] J.-C. Zheng, X.-Q. Shen, C.-Y. Min, X.-F. Meng, Q.-R. Liang, J. Compos. Mater. 44 (2010) 2135. [68] S.H. Xie, J.Y. Li, Y. Qiao, Y.Y. Liu, L.N. Lan, Y.C. Zhou, S.T. Tan, Appl. Phys. Lett. 92 (2008). [69] S. Xiangqian, Z. Jincai, M. Xianfeng, L. Qingrong, J. Wuhan Univ. Technol., Mater. Sci. Ed. 26 (2011) 384. [70] S. Xie, F. Ma, Y. Liu, J. Li, Nanoscale 3 (2011) 3152. [71] S. Xie, Y.-Y. Liu, J.-Y. Li, Front. Phys. 7 (2012) 399. [72] J.D. Starr, J.S. Andrew, Chem. Commun. 49 (2013) 4151. [73] J.D. Starr, J.S. Andrew, J. Mater. Chem. C 1 (2013) 2529. [74] W. Eerenstein, N.D. Mathur, J.F. Scott, Nature 442 (2006) 759. [75] Y. Yang, J. Gao, J.-F. Li, D. Viehland, J. Electrochem. Soc. 158 (2011) K149. [76] J. Li, I. Levin, J. Slutsker, V. Provenzano, P.K. Schenck, R. Ramesh, J. Ouyang, A.L. Roytburd, Appl. Phys. Lett. 87 (2005) 072909. [77] M. Lee, T. Nath, C. Eom, M. Smoak, F. Tsui, Appl. Phys. Lett. 77 (2000) 3547. [78] S.-K. Kim, J.-W. Lee, S.-C. Shin, H.W. Song, C.H. Lee, K. No, J. Magn. Magn. Mater. 267 (2003) 127. [79] W. Eerenstein, M. Wiora, J. Prieto, J. Scott, N. Mathur, Nat. Mater. 6 (2007) 348. [80] C. Deng, Y. Zhang, J. Ma, Y. Lin, C.-W. Nan, J. Appl. Phys. 102 (2007) 074114. [81] K.S. Chang, M.A. Aronova, C.L. Lin, M. Murakami, M.H. Yu, J. Hattrick-Simpers, O.O. Famodu, S.Y. Lee, R. Ramesh, M. Wuttig, I. Takeuchi, C. Gao, L.A. Bendersky, Appl. Phys. Lett. 84 (2004) 3091. [82] N. Ortega, P. Bhattacharya, R.S. Katiyar, P. Dutta, A. Manivannan, M.S. Seehra, I. Takeuchi, S.B. Majumder, J. Appl. Phys. 100 (2006) 126105. [83] J.-P. Zhou, H. He, Z. Shi, C.-W. Nan, Appl. Phys. Lett. 88 (2006) 013111. [84] C.-W. Nan, Prog. Mater Sci. 37 (1993) 1. [85] B.J. Rodriguez, S. Jesse, A.P. Baddorf, T. Zhao, Y.H. Chu, R. Ramesh, E.A. Eliseev, A.N. Morozovska, S.V. Kalinin, Nanotechnology 18 (2007) 405701. [86] A. Gruverman, S.V. Kalinin, J. Mater. Sci. 41 (2006) 107. [87] M. Pan, Y. Liu, G. Bai, S. Hong, V.P. Dravid, A.K. Petford-Long, J. Appl. Phys. 110 (2011) 034103. [88] S. Glotzer, Science 306 (2004) 419. [89] S. Glotzer, M. Solomon, Nat. Mater. 6 (2007) 557. [90] S. Jiang, Q. Chen, M. Tripathy, E. Luijten, K. Schweizer, S. Granick, Adv. Mater. 22 (2010) 1060. [91] D. Ghosh, H. Han, J.C. Nino, G. Subhash, J.L. Jones, J. Am. Ceram. Soc. 95 (2012) 2504. [92] Q.H. Jiang, Z.J. Shen, J.P. Zhou, Z. Shi, C.-W. Nan, J. Eur. Ceram. Soc. 27 (2007) 279. [93] Q. Jiang, F. Liu, H. Yan, H. Ning, Z. Libor, Q. Zhang, M. Cain, M.J. Reece, J. Am. Ceram. Soc. 94 (2011) 2311. [94] Y. Liu, X. Ruan, B. Zhu, S. Chen, Z. Lu, J. Shi, R. Xiong, J. Am. Ceram. Soc. 94 (2011) 1695.
ScienceDirect Scripta Materialia 74 (2014) 38–43 www.elsevier.com/locate/scriptamat
Viewpoint Paper
Prospects for nanostructured multiferroic composite materials Jennifer S. Andrew,⇑ Justin D. Starr and Maeve A.K. Budi Department of Materials Science and Engineering, University of Florida, Gainesville, FL 32611, USA Available online 29 September 2013
Abstract—Magnetoelectric multiferroic materials have the potential to transform a range of applications, including tunable microelectronics and multiphase memories. However, the coexistence of ferromagnetism and ferroelectricity in single-phase materials is quite rare, driving the development of composite multiferroic materials. In these composites, the coupling arises from strain transfer across a shared interface between the ferroelectric and ferromagnetic phase. This viewpoint article highlights recent work as well as future challenges in the synthesis and characterization of nanostructured multiferroic materials. Ó 2013 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved. Keywords: Magnetoelectrics; Nanocomposites; Multiferroics; Ferrites; Electrospinning
1. Introduction Multiferroic materials represent a novel class of material where multiple types of ferroic ordering coexist, and their coupling can lead to additional ordering parameters. These, in turn, can be leveraged to form new multifunctional devices. The ferroic orderings are ferroelectric, ferromagnetic, and ferroelastic and these materials exhibit a spontaneous polarization, magnetization or strain, which can be further controlled by an applied electric, magnetic, or stress field, respectively. In a multiferroic, when multiple types of ferroic ordering are coupled, additional functionalities can arise, including magnetoelectric, piezoelectric and magnetoelastic behaviors. Here, we are interested in magnetoelectric multiferroics that combine ferroelectricity with ferromagnetism. These materials will be the focus of this article, and herein the term multiferroic will refer solely to materials with magnetoeletric behavior. Single-phase multiferroics, although rare, do exist, but they tend to have both a low permittivity and a low permeability. This rare occurrence of single-phase multiferroics is a result of the often seemingly contradictory material requirements for ferroelectricity and ferromagnetism. Ferroelectric materials must be electrically insulating and require a non-centrosymmetric unit cell, while most ferromagnetic materials tend to be metals. A more complete discussion of these contrasting properties has been reviewed by Hill [1,2]. The single-phase multiferroics that do exist are classified into one of two types, designated as type 1 or type 2, based on the origin of multiferroicity. In type 1
⇑ Corresponding author. Tel.: +1 3528463345; e-mail: [email protected]fl.edu
multiferroics the sources for ferroelectricity and ferromagnetism are distinct and independent from one another [3–7]. In type 2 multiferroics the multiferroic behavior arises from magnetic ordering [8]. Although a rich area of research, single-phase multiferroics are plagued by low ordering temperatures, limiting their applications in devices. Single-phase multiferroics have been discussed thoroughly in several outstanding recent review articles and will not be further discussed here [9–13]. To overcome the limitations of these single-phase multiferroics, composite or multi-phase multiferroic materials have been developed. Composite multiferroics are typically heterostructures consisting of a magnetostrictive and a piezoelectric phase. These heterostructures can be readily prepared in a range of composite architectures defined by the connectivity between the two phases by a variety of methods, ranging from thin film to solid-state approaches [14] (Figure 1). For example, a 0-3 composite describes a particulate-based composite where zero-dimensional (0-D) particles are dispersed within a 3-D matrix. Particulate-based composites (0-3) [15], pillars within a 3-D matrix (1-3) [16] and layered heterostructures (2-2) [17–24] represent the most common geometries of magnetoelectric heterostructures (Fig. 1a). In these multiferroic composites the magnetoelectric coupling is mechanical in nature and is based on the fact that piezoelectric and magnetostrictive materials undergo a shape change in an applied electric or magnetic field, respectively. The magnetoelectric effect in these composites can be described by the direct magnetoelectric effect, which is the product of the magnetostrictive effect (magnetic/mechanical) and the piezoelectric effect (mechanical/electrical) [25].
1359-6462/$ - see front matter Ó 2013 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.scriptamat.2013.09.023
J. S. Andrew et al. / Scripta Materialia 74 (2014) 38–43
Figure 1. (a) Composite connectivities of composites prepared by traditional routes and (b) connectivities that could be realized with the development of nanostructured building blocks.
magnetic mechanical ð1Þ mechanical electric The figure of merit for these materials is the magnetoelectric coefficient, aE, which is determined experimentally by measuring the change in electric field, dE, generated by applying an ac magnetic field, dH, to a biased sample, Direct ME effect ¼
dE ð2Þ dH Although these magnetoeletric heterostructures have high theoretically predicted values for aE, actual experimental values tend to be much lower [20,26–31]. The fact that these experimental coupling values are lower than predicted can be attributed to a variety of factors including the formation of cracks or impurity phases at the interface between the two phases. These heterostructures are also challenged by the low resistivity ferrite phase and substrate effects, which can mechanically clamp the sample [16]. Because this coupling involves mechanical coupling across the interface between the magnetostrictive and piezoelectric phase, it follows that one route to enhance the magnetoelectric coupling is by maximizing the interfacial area between the two phases [32]. One way to accomplish this is to use nanoscale materials due to their large surface to volume ratio. The first example of coupling in multiferroic nanostructures was demonstrated by Zheng et al. [16], who embedded CoFe2O4 nanopillars epitaxially in a BaTiO3 thin film matrix. However, it is worth noting that here the CoFe2O4 rods are nanoscale in diameter only. Since this seminal work, much effort has been done to develop new nanostructured building blocks for the formation of the next generation of multiferroic materials. The focus of this viewpoint article will be to describe and review the development of nanostructured multiferroic building blocks that can be assembled into higher order multiferroic structures and devices, while also addressing the needs to develop and utilize novel characterization methods to fully study these systems. aE ¼
2. Synthesis and characterization of nanostructured multiferroic materials 2.1. Zero-dimensional (nanoparticles) One of the challenges in fabricating multiferroic composites with magnetoelectric coupling values approaching
39
the theoretically predicted values is the low resistivity of the ferrite phase. In many of the theoretical models the optimum volume fraction of ferrite is at high volume fractions (0.5–0.9) [26]. However, at such high loadings the leaky ferrite phase makes it challenging to pole the material. Therefore, much effort has been made to isolate the leaky ferrite phase within a ferroelectric matrix (0-3 connectivity). This can be accomplished via a multitude of routes with a range of synthetic complexities. The simplest methods involve mechanically mixing particles followed by densification via solid-state methods [33–37], or mixing solutions of ferroelectric and piezoelectric particles followed by spin-coating and calcination [38], and molten-salt based synthetic routes [39,40]. Thin film approaches such as pulsed laser deposition have also been used to fabricate particulate composites [41]. Novel one-pot synthesis strategies have also been developed to more fully control the synthesis and microstructure of composite multiferroics [33,42–46]. However, using many of these routes, it remains a challenge to isolate the ferrite phase at the high volume fractions required to obtain high performance magnetoelectric composites. In many of these routes the device performance drops off at volume fractions of ferrite between 0.2 and 0.5, corresponding with percolation of the ferrite phase [47–49]. Core–shell magnetoelectric particles have been synthesized to address this challenge of isolating the ferrite phase while maintaining a high overall volume fraction of ferrite. These novel particles consist of a ferrite core (CoFe2O4, Fe3O4, Ni0.5Zn0.5Fe2O4) surrounded by (i.e. electrically isolated by) a ferroelectric shell [51– 53,55–59]. The ferrite core is typically synthesized via either a hydrothermal or co-precipitation method, and is subsequently coated using sol–gel methods with a BaTiO3 shell. These particles are then calcined and/or sintered at elevated temperatures (650–1300 °C). Duong et al. have reported the longitudinal and transverse maximum magnetoelectric (ME) coupling coefficients as 3.4 mV cm1 Oe1 and 2 mV cm1 Oe1 respectively for sintered pellets of these core-shell nanoparticles [52,56,57]. However, they found that by varying pressure, sintering temperature and sintering duration, these values were further increased to 3.53 mV cm1 Oe1 and 2.23 mV cm1 Oe1. Duong et al. [57] also synthesized a structure using barium titanate as the core and cobalt ferrite as the shell. The barium titanate was synthesized using sol–gel methods, then suspended in a solution containing a chelating agent and the elements to form cobalt ferrite. As expected, in these materials where the CoFe2O4 was not isolated in the core, the ME coefficient was found to be 4.2 times lower than a comparable sample with cobalt ferrite as the core [57]. All the synthesis examples outlined above require calcining steps to produce crystalline tetragonal perovskite barium titanate from the amorphous barium titanium sol, which is needed for piezoelectric behavior. The exact temperatures varied from 650 to 1000 °C. All of the reported ME measurements were made on samples that underwent additional sintering for further densification [53,52,56–59]. Though some groups use transmission electron microscopy (TEM) and SEM to confirm the core–shell particles, the particles are agglomerated because of the high temperature processing [51,55]. One of
40
J. S. Andrew et al. / Scripta Materialia 74 (2014) 38–43
the challenges is to develop synthesis procedures that allow for the formation of isolated multiferroic particles that are free of agglomerates. These particles could then be assembled into higher order structures using colloidal processing routes. Additionally, the application of novel characterization methods to observe multiferroic behavior on a single nanostructured particle will provide a means to more fundamentally understand the magnetoelectric coupling and structure–property relationships in these systems. 2.2. One-dimensional (fibers) One-dimensional multiferroic nanostructures such as nanofibers provide another route to bulk multiferroic materials. Ideal nanofibers, free of substrate based constraints, have been predicted to display mulitferroic coupling that is an order of magnitude greater than that found in ideal thin films [60]. Further, composite nanofibers can be synthesized in a number of configurations by varying the orientation of the piezoelectric and magnetostrictive phases. These morphologies can be used as building blocks to enable additional connectivities in resultant bulk structures, including 1-1 (Fig. 1b). Multiferroic nanofibers can be synthesized via electrospinning [61,62]. Several groups have shown that, if two compatible phases are simultaneously electrospun, it is possible to produce biphasic structures as long as the phases remain segregated until the point of electrification [63–65]. This process relies on the fact that the kinetics of solvent evaporation and hydrolysis and condensation occur faster than mixing between the two miscible phases. Electrospinning multiphase structures depends on carefully matching the properties of precursor solutions, as similar solutions are most likely to cohesively jet. This technique can be used to form multiferroic fibers with a range of microstructures (Fig. 2). The simplest method involves synthesizing a nanofiber with a random distribution of ferroelectric and magnetostrictive phases. This can be achieved by mixing ceramic precursors for each phase together, forming a blended polymer solution. When electrospun and calcined, such a solution will produce biphasic nanofibers with a random distribution of grains of each phase (Fig. 2a). Randomly mixed multiferroic fibers have been synthesized for systems composed of lead zirconate titanate and cobalt or nickel ferrite [66–69]. Multiferroic nanofibers can also be synthesized in a coaxial or core–shell geometry. These nanofibers maintain a large contact area between phases, while segregating the ferroelectric and ferromagnetic phases. Core–shell nanofibers have been produced via electrospinning (Fig. 2b) for lead zirconate titanate with CoFe2O4 [70,71]. Core–shell nanofibers and nanotubes have also been fabricated via membrane template approaches, where a polycarbonate membrane was used as a template for the synthesis of CoFe2O4 nanotubes, which were subsequently added to a BaTiO3 precursor solution and calcined at 780 °C to form a BaTiO3 shell [54]. Both randomly dispersed and core–shell nanofibers can be assembled into bulk structures with connectivities similar to traditional methods, including 0-3 or 1-3, for randomly distributed or core–shell structures,
Figure 2. Electrospinning can produce multiferroic fibers in a range of geometries, including (a) random, (b) core–shell or (c) Janus-type arrangements between the two phases. TEM images reproduced with permission from Refs. [61,64,66].
respectively. Further, it is extraordinarily difficult to measure the magnetoelectric properties of a single isolated nanofiber. As a result, many reports of these mutiferroic nanofibers do not attempt to demonstrate multiferroic behavior or quantify the magnetoelectric coupling coefficient. Piezoresponse force microscopy (PFM) has been used to measure magnetoelectric coupling along a single nanofiber [71]. Using this method, Xie et al. [60] measured a magnetoelectric coupling coefficient of 2.95 104 mV cm1 Oe1 for CoFe2O4– Pb(Zr0.52Ti0.48)O3 core–shell nanofibers, which, as theoretically predicted, is two orders of magnitude greater than thin films of similar composition [70]. Recently, a new morphology of multiferroic nanofibers has been produced: the so-called Janus-type or hemispherical composite nanofiber (Fig. 2c). In these fibers, ferroelectric and ferromagnetic phases are arranged along the length of a nanofiber, with a shared interface running down the longitudinal axis [72]. These fibers retain external access to both phases as well as the interface, making them well suited to modular structures. For example, these Janus-type fibers can be assembled into composites with a 1-1 connectivity (Fig. 1b), which is not readily realized by conventional approaches. Thus far, these fibers have only been demonstrated for the barium titanate–cobalt ferrite system, but they have been further extended to trilayer structures [73]. Indirect evidence of magnetoelectric coupling has been shown in these Janus-type fibers by measuring the magnetization as a function of temperature. At the Curie temperature of barium titanate, a marked change in the magnetization of cobalt ferrite is present as the cobalt ferrite is strained when barium titanate undergoes a structural transition from tetragonal to cubic. 2.3. Two-dimensional (films) Layered structures formed from 2-2 laminate composites can overcome the leakage problem due to the low resistivity ferrite phase. To form nanostructured layered composites many efforts have focused on thin films. By using techniques such as pulsed laser deposition (PLD) or atomic layer deposition (ALD), precise heterostructures can be obtained. By carefully choosing
J. S. Andrew et al. / Scripta Materialia 74 (2014) 38–43
a substrate and growing layers of piezoelectric and magnetostrictive materials via PLD or similar techniques, it is possible to carefully control the alignment of crystallographic planes and achieve an epitaxial relationship across the shared interface. These epitaxial interfaces may reduce some of the losses inherent in bulk composite structures and lend themselves well to accurate study, as parameters can be carefully controlled and subsequently verified by simulation [74]. However, these routes are not without challenges because in many instances the substrates clamp the mechanical coupling, limiting magnetoelectric response. Biphasic composite film structures have been synthesized for a number of material systems, including barium titanate–cobalt ferrite, lead zirconate titanate– cobalt ferrite and barium titanate–lanthanum strontium manganite, as well as a variety of other systems [16,18,75–80]. Chang et al. [81] synthesized a variety of cobalt ferrite–barium titanate composites with a composition spread to determine the effects of film thickness on resultant properties. They grew a 300 nm thick film that consisted of wedges of cobalt ferrite and barium titanate. The thickness of each wedge ranged from as little as one unit cell to tens of unit cells. Using this approach, Chang et al. found that there was a region towards the middle of the spread with both a high dielectric constant and ferromagnetic behavior. Despite the large body of literature on the synthesis of magnetoelectric films of different compositions, comparatively very few studies have published direct measurements of the magnetoelectric coupling coefficient, as it can be difficult to make these measurements at the nanoscale. In the few cases where these values have been reported, the results have varied. Some lead zirconate titanate–nickel ferrite structures show magnetoelectric coupling values (16 mV cm1 Oe1) [41] that fall short of their bulk counterparts, while other barium titanate–nickel ferrite structures are more commensurate with accepted bulk values (12.1 and 7.9 mV cm1 Oe1 for out-of-plane and in-plane coupling, respectively) [80]. However, with careful refinement of the structure, converse magnetoelectric coupling coefficients (a = l0DM/DE) can be as large as those of Terfenol-D laminate systems (2.3 107 s m1), and provide evidence of the giant magnetoelectric effect [79]. As in the case for 1-D magnetoelectric structures, many groups verify multiferroic properties via indirect measurement (Fig. 3): either by observing a change in magnetization at the ferroelectric Curie temperature or by examining the changes in dielectric properties at the ferromagnetic or antiferromagnetic Curie/Nee´l temperatures (Fig. 3a) [16,72,73]. Other groups examine how polarization or magnetization hysteresis loops change under the presence of an applied magnetic or electric field, respectively (Fig. 3b) [82,83]. While thin films have received considerable attention, they are not without drawbacks: techniques like PLD and ALD can be slow and expensive. Further, the use of a substrate can reduce the available degrees of freedom and create clamping effects, lessening strain transfer and negating potential magnetoelectric coupling [60]. Some multilayer thin films fail to show indirect evidence of coupling due to these limitations [16].
41
Figure 3. Indirect methods for determining the existence of magnetoelectric coupling: (a) observation of anomalous magnetization behavior at the ferroelectric Curie temperature where the piezoelectric material undergoes a structural change leading to a change in the magnetization of the magnetostrictive material through strain coupling (Reprinted with permission from Ref. [16]) and (b) Measurement of changes in the electronic polarization loop of a magnetoelectric composite under an applied magnetic field. Reprinted with permission from Ref. [75].
Thin films are also limited to 2-2, 1-3 and 0-3 type connectivities [14,84]. 3. Looking forward 3.1. Characterization methods Bulk multiferroic composites typically exhibit magnetoelectric coupling values that are significantly lower than theoretically predicted values. For the case of BaTiO3–CoFe2O4 composites typical experimental values are 1.5–130 mV cm1 Oe1 [26,53,55,57,50], whereas, depending on the assumptions made, theory predicts values that range from 1.5 to 5 V cm1 Oe1 [26,27]. These decreased values can be attributed to the presence of cracks, impurity phases or leakage due to the low resistivity ferrite phase. Some of these limitations can be overcome by the development and characterization of the novel nanostructured multiferroic building blocks discussed in this article and their assembly into bulk structures. However, to date, studies of magnetoelectric coupling at the nanoscale are limited, but are necessary to develop composite multiferroics with enhanced coupling. Piezoeresponse force microscopy (PFM) has emerged as a promising technique for measuring the properties of ferroic materials, including measuring the magnetoelectric coefficients of nanostructured materials [70,85,86]. Pan et al. [87] used electron tomography in conjunction with vector-PFM (v-PFM) to study CoFe2O4–PbTiO3 nanocomposites prepared via metal organic chemical vapor deposition. By combining these two methods, they were able to determine the 3-D structure of the nanocomposite at the nanoscale via electron tomography and correlate it with the 3-D polarization domains, determined with v-PFM. Futhermore, this data could be correlated to the magnetic properties of the material, including hysteresis and anisotropy data. The application of these tools to nanostructured multiferroic materials will allow for a thorough understanding of structure–property relationships in these complex heterostructures. Nanoscale structure–property relationships for magnetoelectric coupling behavior can then
42
J. S. Andrew et al. / Scripta Materialia 74 (2014) 38–43
be used to guide the assembly of these into bulk materials with enhanced properties. 3.2. Idealized structures The generation of new nanostructured multiferroic materials allows the fabrication of composites with a range of connectivities, including new connectivities not readily synthesized by traditional routes. For example, the Janus-type nanofibers provide an anisotropic structure that can be leveraged for self-assembly into more complex structures, including 1-1 type-connectivity (Fig. 1b) [88–90], while isolated magnetoelectric particles can be assembled into multiferroic composites with unique architectures using colloidal processing routes. In many cases, to obtain a multiferroic device it will be necessary to consolidate these structures into dense composites. In order to maintain the nanostructure of the building blocks, fast firing approaches can be utilized, including microwave sintering and spark plasma sintering [91–94]. 4. Summary Much research has focused on the development of nanostructured multiferroic materials. This body of research includes the development of many novel synthetic routes as well as the thorough structural characterization of the materials developed. However, in many cases the magnetoelectric properties of these materials are not fully described. Therefore, more work needs to be done to develop and employ tools capable of measuring the magnetoelectric properties of these nanostructured elements. These new nanostructured elements can then be assembled into unique architectures, and their structures can be preserved using fast firing techniques. By understanding structure–property relationships at the nanoscale, novel magnetoelectric materials with enhanced properties can be realized. Acknowledgement This work was supported by the National Science Foundation through a CAREER Award (DMR1150665). References [1] N. Hill, J. Phys. Chem. B 104 (2000) 6694. [2] N. Hill, Annu. Rev. Mater. Sci. 32 (2002) 1. [3] A. Moreira dos Santos, S. Parashar, A. Raju, Y. Zhao, A. Cheetham, C. Rao, Solid State Commun. 122 (2002) 49. [4] J. Wang, J. Neaton, H. Zheng, V. Nagarajan, S. Ogale, B. Liu, D. Viehland, V. Vaithyanathan, D. Schlom, U. Waghmare, N. Spaldin, K. Rabe, M. Wuttig, R. Ramesh, Science 299 (2003) 1719. [5] R. Seshadri, N. Hill, Chem. Mater. 13 (2001) 2892.
[6] M. Fiebig, T. Lottermoser, D. Frohlich, A. Goltsev, R. Pisarev, Nature 419 (2002) 818. [7] J. Van Den Brink, D.I. Khomskii, J. Phys.: Condens. Matter 20 (2008) 434217. [8] T. Kimura, T. Goto, H. Shintani, K. Ishizaka, T. Arima, Y. Tokura, Nature 426 (2003) 55. [9] L.W. Martin, D.G. Schlom, Curr. Opin. Solid State Mater. Sci. 16 (2012) 199. [10] R. Ramesh, N. Spaldin, Nat. Mater. 6 (2007) 21. [11] N.A. Spaldin, S.-W. Cheong, R. Ramesh, Phys. Today 63 (2010) 38. [12] G. Catalan, J.F. Scott, Adv. Mater. 21 (2009) 2463. [13] L.W. Martin, S.P. Crane, Y.H. Chu, M.B. Holcomb, M. Gajek, M. Huijben, C.H. Yang, N. Balke, R. Ramesh, J. Phys.: Condens. Matter 20 (2008) 434220. [14] R. Newnham, D. Skinner, L. Cross, Mater. Res. Bull. 13 (1978) 525. [15] J. van den Boomgaard, R. Born, J. Mater. Sci. 13 (1978) 1538. [16] H. Zheng, J. Wang, S. Lofland, Z. Ma, L. MohaddesArdabili, T. Zhao, L. Salamanca-Riba, S. Shinde, S. Ogale, F. Bai, D. Viehland, Y. Jia, D. Schlom, M. Wuttig, A. Roytburd, R. Ramesh, Science 303 (2004) 661. [17] J. Ryu, S. Priya, A. Vazquez Carazo, K. Uchino, J. Am. Ceram. Soc. 84 (2001) 2905. [18] G. Srinivasan, E.T. Rasmussen, J. Gallegos, R. Srinivasan, Y. Bokhan, V. Laletin, Phys. Rev. B 64 (2001) 214408. [19] G. Srinivasan, E.T. Rasmussen, B.J. Levin, R. Hayes, Phys. Rev. B 65 (2002) 134402. [20] M. Bichurin, V. Petrov, G. Srinivasan, J. Appl. Phys. 92 (2002) 7681. [21] C. Nan, M. Li, J. Huang, Phys. Rev. B 63 (2001) 144415. [22] G. Srinivasan, E. Rasmussen, R. Hayes, Phys. Rev. B 67 (2003) 014418. [23] M. Bichurin, D. Filippov, V. Petrov, V. Laletsin, N. Paddubnaya, G. Srinivasan, Phys. Rev. B 68 (2003) 132408. [24] X. Lu, H. Li, J. Appl. Phys. 109 (2011) 074112. [25] J. Ma, J. Hu, Z. Li, C.-W. Nan, Adv. Mater. 25 (2011) 1062. [26] C.-W. Nan, Undefined author, S. Dong, D. Viehland, G. Srinivasan, J. Appl. Phys. 103 (2008) 031101. [27] J. Ryu, S. Priya, K. Uchino, H.-E. Kim, J. Electroceram. 8 (2002) 107. [28] M.I. Bichurin, V.M. Petrov, Adv. Condens Matter Phys 2012 (2012) 1. [29] M. Bichurin, V. Petrov, G. Srinivasan, Phys. Rev. B 68 (2003) 054402. [30] V.M. Petrov, G. Srinivasan, M.I. Bichurin, A. Gupta, Phys. Rev. B 75 (2007) 224407. [31] M. Avellaneda, G. Harshe´, J. Intell. Mater. Syst. Struct. 5 (1994) 501. [32] R. Liu, Y. Zhao, R. Huang, Y. Zhao, H. Zhou, J. Mater. Chem. 20 (2010) 10665. [33] A. Testino, L. Mitoseriu, V. Buscaglia, M.T. Buscaglia, I. Pallecchi, A.S. Albuquerque, V. Calzona, D. Marre, A.S. Siri, P. Nanni, J. Eur. Ceram. Soc. 26 (2006) 3031–3036. [34] J. Ryu, A. Carazo, K. Uchino, H.-E. Kim, J. Electroceram. 7 (2001) 17. [35] G. Srinivasan, C. DeVreugd, C. Flattery, V. Laletsin, N. Paddubnaya, Appl. Phys. Lett. 85 (2004) 2550. [36] A. Khamkongkaeo, P. Jantaratana, C. Sirisathitkul, T. Yamwong, S. Maensiri, Trans. Nonferrous Met. Soc. China 21 (2011) 2438. [37] S.G. Lu, Z.K. Xu, Y.P. Wang, S.S. Guo, H. Chen, T.L. Li, S.W. Or, J. Electroceram. 21 (2007) 398.
J. S. Andrew et al. / Scripta Materialia 74 (2014) 38–43
[38] A. McDannald, M. Staruch, G. Sreenivasulu, C. Cantoni, G. Srinivasan, M. Jain, Appl. Phys. Lett. 102 (2013) 122905. [39] J.-W. Nie, G.-Y. Xu, X.-J. Qi, Trans. Nonferrous Met. Soc. China 19 (2009) 1593. [40] J. Nie, G. Xu, Y. Yang, C. Cheng, Mater. Chem. Phys. 115 (2009) 400. [41] H. Ryu, P. Murugavel, J.H. Lee, S.C. Chae, T.W. Noh, Y.S. Oh, H.J. Kim, K.H. Kim, J.H. Jang, M. Kim, Appl. Phys. Lett. 89 (2006) 102907. [42] S.Q. Ren, L.Q. Weng, S.-H. Song, F. Li, J.G. Wan, M. Zeng, J. Mater. Sci. 40 (2005) 4375. [43] D. Wu, W. Gong, H. Deng, M. Li, J. Phys. D Appl. Phys. 40 (2007) 5002. [44] A.R. Iordan, M. Airimioaiei, M.N. Palamaru, C. Galassi, A.V. Sandu, C.E. Ciomaga, F. Prihor, L. Mitoseriu, A. Ianculescu, J. Eur. Ceram. Soc. 29 (2009) 2807. [45] L. Weng, Y. Fu, S. Song, J. Tang, J. Li, Scripta Mater. 56 (2007) 465. [46] J.-G. Wan, H. Zhang, X. Wang, D. Pan, J.-M. Liu, G. Wang, Appl. Phys. Lett. 89 (2006) 122914. [47] A. Lakhtakia, T. Mackay, Int. J. Electron. Commun. 58 (2004) 1. [48] C.A.F. Vaz, J. Hoffman, C.H. Anh, R. Ramesh, Adv. Mater. 22 (2010) 2900. [49] L.P. Curecheriu, M.T. Buscaglia, V. Buscaglia, L. Mitoseriu, P. Postolache, A. Ianculescu, P. Nanni, J. Appl. Phys. 107 (2010) 104106. [50] V. Corral-Flores, D. Bueno-Baques, R.F. Ziolo, Acta Mater. 58 (2010) 764. [51] M. Etier, Y. Gao, V.V. Shvartsman, A. Elsukova, J. Landers, H. Wende, D.C. Lupascu, Ferroelectrics 438 (2012) 115. [52] G.V. Duong, R. Groessinger, R.S. Turtelli, IEEE Trans. Magn. 42 (2006) 3611. [53] G.V. Duong, R.S. Turtelli, R. Groessinger, J. Magn. Magn. Mater. 322 (2010) 1581. [54] K. Raidongia, A. Nag, A. Sundaresan, C.N.R. Rao, Appl. Phys. Lett. 97 (2010) 062904. [55] V.V. Shvartsman, F. Alawneh, P. Borisov, D. Kozodaev, D.C. Lupascu, Smart Mater. Struct. 20 (2011) 075006. [56] G.V. Duong, R. Groessinger, R. Sato Turtelli, J. Magn. Magn. Mater. 310 (2007) 1157. [57] G.V. Duong, R. Groessinger, R. Sato Turtelli, J. Magn. Magn. Mater. 310 (2007) e361. [58] G.V. Duong, R. Groessinger, J. Magn. Magn. Mater. 316 (2007) e624. [59] V. Corral-Flores, D. Bueno-Baques, D. Carrillo-Flores, J.A. Matutes-Aquino, J. Appl. Phys. 99 (2006) 0803. [60] C.L. Zhang, W.Q. Chen, S.H. Xie, J.S. Yang, J.Y. Li, Appl. Phys. Lett. 94 (2009) 102907. [61] W. Sigmund, J. Yuh, H. Park, V. Maneeratana, G. Pyrgiotakis, A. Daga, J. Taylor, J. Nino, J. Am. Ceram. Soc. 89 (2006) 395. [62] S. Ramakrishna, K. Fujihara, W. Teo, T. Lim, Z. Ma, An introduction to electrospinning and nanofibers, World Scientific, New Jersey, 2005. [63] K.-H. Roh, D.C. Martin, J. Lahann, Nat. Mater. 4 (2005) 759.
43
[64] Z. Liu, D.D. Sun, P. Guo, J.O. Leckie, Nano Lett. 7 (2007) 1081. [65] S. Bhaskar, J. Hitt, S. Chang, J. Lahann, Angew. Chem. Int. Ed. 48 (2009) 4589. [66] S.H. Xie, J.Y. Li, Y.Y. Liu, L.N. Lan, G. Jin, Y.C. Zhou, J. Appl. Phys. 104 (2008) 024115. [67] J.-C. Zheng, X.-Q. Shen, C.-Y. Min, X.-F. Meng, Q.-R. Liang, J. Compos. Mater. 44 (2010) 2135. [68] S.H. Xie, J.Y. Li, Y. Qiao, Y.Y. Liu, L.N. Lan, Y.C. Zhou, S.T. Tan, Appl. Phys. Lett. 92 (2008). [69] S. Xiangqian, Z. Jincai, M. Xianfeng, L. Qingrong, J. Wuhan Univ. Technol., Mater. Sci. Ed. 26 (2011) 384. [70] S. Xie, F. Ma, Y. Liu, J. Li, Nanoscale 3 (2011) 3152. [71] S. Xie, Y.-Y. Liu, J.-Y. Li, Front. Phys. 7 (2012) 399. [72] J.D. Starr, J.S. Andrew, Chem. Commun. 49 (2013) 4151. [73] J.D. Starr, J.S. Andrew, J. Mater. Chem. C 1 (2013) 2529. [74] W. Eerenstein, N.D. Mathur, J.F. Scott, Nature 442 (2006) 759. [75] Y. Yang, J. Gao, J.-F. Li, D. Viehland, J. Electrochem. Soc. 158 (2011) K149. [76] J. Li, I. Levin, J. Slutsker, V. Provenzano, P.K. Schenck, R. Ramesh, J. Ouyang, A.L. Roytburd, Appl. Phys. Lett. 87 (2005) 072909. [77] M. Lee, T. Nath, C. Eom, M. Smoak, F. Tsui, Appl. Phys. Lett. 77 (2000) 3547. [78] S.-K. Kim, J.-W. Lee, S.-C. Shin, H.W. Song, C.H. Lee, K. No, J. Magn. Magn. Mater. 267 (2003) 127. [79] W. Eerenstein, M. Wiora, J. Prieto, J. Scott, N. Mathur, Nat. Mater. 6 (2007) 348. [80] C. Deng, Y. Zhang, J. Ma, Y. Lin, C.-W. Nan, J. Appl. Phys. 102 (2007) 074114. [81] K.S. Chang, M.A. Aronova, C.L. Lin, M. Murakami, M.H. Yu, J. Hattrick-Simpers, O.O. Famodu, S.Y. Lee, R. Ramesh, M. Wuttig, I. Takeuchi, C. Gao, L.A. Bendersky, Appl. Phys. Lett. 84 (2004) 3091. [82] N. Ortega, P. Bhattacharya, R.S. Katiyar, P. Dutta, A. Manivannan, M.S. Seehra, I. Takeuchi, S.B. Majumder, J. Appl. Phys. 100 (2006) 126105. [83] J.-P. Zhou, H. He, Z. Shi, C.-W. Nan, Appl. Phys. Lett. 88 (2006) 013111. [84] C.-W. Nan, Prog. Mater Sci. 37 (1993) 1. [85] B.J. Rodriguez, S. Jesse, A.P. Baddorf, T. Zhao, Y.H. Chu, R. Ramesh, E.A. Eliseev, A.N. Morozovska, S.V. Kalinin, Nanotechnology 18 (2007) 405701. [86] A. Gruverman, S.V. Kalinin, J. Mater. Sci. 41 (2006) 107. [87] M. Pan, Y. Liu, G. Bai, S. Hong, V.P. Dravid, A.K. Petford-Long, J. Appl. Phys. 110 (2011) 034103. [88] S. Glotzer, Science 306 (2004) 419. [89] S. Glotzer, M. Solomon, Nat. Mater. 6 (2007) 557. [90] S. Jiang, Q. Chen, M. Tripathy, E. Luijten, K. Schweizer, S. Granick, Adv. Mater. 22 (2010) 1060. [91] D. Ghosh, H. Han, J.C. Nino, G. Subhash, J.L. Jones, J. Am. Ceram. Soc. 95 (2012) 2504. [92] Q.H. Jiang, Z.J. Shen, J.P. Zhou, Z. Shi, C.-W. Nan, J. Eur. Ceram. Soc. 27 (2007) 279. [93] Q. Jiang, F. Liu, H. Yan, H. Ning, Z. Libor, Q. Zhang, M. Cain, M.J. Reece, J. Am. Ceram. Soc. 94 (2011) 2311. [94] Y. Liu, X. Ruan, B. Zhu, S. Chen, Z. Lu, J. Shi, R. Xiong, J. Am. Ceram. Soc. 94 (2011) 1695.