Thermal Driven Giant Spin Dynamics at Three-Dimensional

the two dimensional interface that transfers stress/strain to achieve the large magnetoelectric (ME) .... effective method to determine the ME couplin...
1 downloads 9 Views 2MB Size
Subscriber access provided by UNIV OF NEW ENGLAND ARMIDALE

Article

Thermal Driven Giant Spin Dynamics at Three-Dimensional Heteroepitaxial Interface in Ni0.5Zn0.5Fe2O4/BaTiO3-Pillar Nanocomposites Guohua Dong, Ziyao Zhou, Mengmeng Guan, Xu Xue, Mingfeng Chen, Jing Ma, Zhongqiang Hu, Wei Ren, Zuo-Guang Ye, Ce-Wen Nan, and Ming Liu ACS Nano, Just Accepted Manuscript • DOI: 10.1021/acsnano.8b00962 • Publication Date (Web): 02 Mar 2018 Downloaded from http://pubs.acs.org on March 3, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Nano is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Nano

Thermal

Driven

Three-Dimensional

Giant

Spin

Dynamics

at

Interface

in

Heteroepitaxial

Ni0.5Zn0.5Fe2O4/BaTiO3-Pillar Nanocomposites Guohua Dong†, Ziyao Zhou*†, Mengmeng Guan†, Xu Xue†, Mingfeng Chen§, Jing Ma§, Zhongqiang Hu†, Wei Ren†, Zuo-Guang Ye‡, Ce-Wen Nan§, Ming Liu*†



Electronic Materials Research Laboratory, Key Laboratory of the Ministry of

Education & State Key Laboratory for Mechanical Behavior of Materials, Xi’an Jiaotong University, Xi’an 710049, China §

State Key Lab of New Ceramics and Fine Processing, School of Materials Science

and Engineering, Tsinghua University, Beijing, 100084, China ‡

Department of Chemistry and 4D LABS, Simon Fraser University, Burnaby, British

Columbia, V5A 1S6, Canada

*E-mail: [email protected]; [email protected]

ABSTRACT: Traditional magnetostrictive/piezoelectric laminated composites rely on the two dimensional interface that transfers stress/strain to achieve the large magnetoelectric (ME) coupling, nevertheless, they suffer from the theoretical limitation of the strain effect and of the substrate clamping effect in real ME applications. In this work, 3D NZFO/BTO-pillar nanocomposite films were grown on 1 ACS Paragon Plus Environment

ACS Nano 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

SrTiO3 by template-assisted pulsed laser deposition, where BaTiO3 (BTO) nanopillars appeared in an array with distinct phase transitions as the cores were covered by NiZn ferrite (NZFO) layer. The perfect 3D heteroepitaxial interface between BTO and NZFO phases can be identified without any edge dislocations, which allows effective strain transfer at the 3D interface. The 3D structure nanocomposites enable the strong two magnon scattering (TMS) effect that enhance ME coupling at the interface and reduce the clamping effect by strain relaxation. Thereby, a record high FMR field shift of 1866 Oe in NZFO/BTO-pillar nanocomposite was obtained at the TMS critical angle near the BTO nanopillars phase transition of 255 K, compared to the old highest record of FMR field shift in layered structure which is only 672 Oe.

KEYWORDS: multiferroic, magnetoelectric coupling, ferromagnetic resonance, spinel ferrite, 3D nanostructure

2 ACS Paragon Plus Environment

Page 2 of 29

Page 3 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Nano

Multiferroic composite materials with coexistence of the ferroelectric (FE) and ferromagnetic (FM) ordering have attracted great research interests due to their good magnetoelectric (ME) performance at room temperature,1–6 which lead to many compact, efficient and high-speed voltage tunable ME devices, including resonators, phase shifters, and tunable filters through electric field (E-field) modulation of magnetic properties or vice versa.7–9 In recent researches on the two dimensional (2D) thin film system such as Fe3O4/PMN-PT, FeGaB/PZN-PT and Metglass/PZT and so on, the stress/strain-mediated ME coupling still serves as one of the most promising ME mechanisms due to its wider effective range, greater ME tunability and fewer chemical damages.10–12 Nevertheless, the ME tunability of these heterostructures is fundamentally restricted by magnetostriction in FM phase and piezoelectricity in FE phase. Recently, an electric field control of two magnon scattering (TMS) effect has been

demonstrated

heterostructures.13,14

in

the

LSMO/PMN-PT

Interestingly,

the

ME

(011)

coupling

and

NZFO/PMN-PT

coefficients

of

these

heterostructures were enhanced upto 500% - 700% by the TMS effect, which overcomes the conventional limitation of strain effect by generating an additional spin lattice coupling (SLC) effect. TMS phenomenon is almost universal in magnetic systems, including ultrathin ferromagnetic films, and super-lattices, meanwhile it is related to the nanostructure of magnetic systems. Still, these TMS enhanced magnetostrictive thin film/piezoelectric bulk laminated composites usually require large external voltage (400 - 600 V), which is hardly achieved in integrated circuits, to drive the piezoelectric crystal. Additionally, in thin film FM/FE multiferroic 3 ACS Paragon Plus Environment

ACS Nano 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

heterostructures, the substrate clamping effect deteriorates the stress/strain transfer and limits their ME tunability as well as SLC – TMS strength accordingly in integrated circuit applications.15,16 In the past decade, significant efforts have been made to enhance the strain effect and to weaken the substrate clamping effect in the nanocomposite multiferroics.15,17,18 For example, the nanocomposite films consisting of epitaxial magnetic pillars in a single crystal perovskite matrix have attracted a lot of research interests due to the higher heteroepitaxial interfacial area between the ferromagnetic and ferroelectric phases and the significantly reduced clamping effect for strain relaxation.19,20 With the breakthrough in film development, various self-assembly three dimensional (3D) vertical nanostructures (1-3 type) have been explored from different perspectives, such as BaTiO3 (BTO)-Au, La0.7Sr0.3MnO3/ZnO and BiFeO3/CoFe2O4 (CFO).21–24 Moreover, we have revealed that the TMS effects are strongly related to nanostructures. Therefore, it is expected that the stronger scattering TMS effect may be found in the 3D nanostructure. As thus, the 3D nanostructured multiferroics may create a record-breaking ME coupling effect by reducing the substrate clamping effect and increasing the TMS effect respectively. Although applying E-field is direct ME coupling measurement, it is extremely difficult to apply an electric field due to the poor electrical insulation of most ferromagnetic materials in nanocomposite multiferroics. Therefore, the thermal driven ME coupling is also considered as an effective method to determine the ME coupling strength, especially when facing the challenge of adding E-field on nano-structures.2,15,25 4 ACS Paragon Plus Environment

Page 4 of 29

Page 5 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Nano

In our previous work, we obtained a significant ME coupling effect through BTO ferroelectric phase transition in the self-assembled vertically aligned 1-3 YIG/BTO nanostructure.26 Nevertheless, the high growth temperature of YIG and the big lattice mismatch between YIG and STO are still the impenetrable obstacles, leading to polycrystalline structure of YIG phases. The polycrystallinity of YIG with high density defects ruins the RF/microwave property with large FMR linewidth. More ominously, the cribriform architecture and polycrystalline behavior exceed the limitation of the TMS effect. In this work, fully epitaxial Ni0.5Zn0.5Fe2O4/BaTiO3 pillar nanocomposites were successfully fabricated by template-assisted pulsed laser deposition (PLD),25,27–29 which is much more convenience than electron beam lithography (EBL), ion milling or focused ion beam (FIB) patterning methods.30,31 An ultrathin AAO membrane as stencil mask has been applied to fabricate the well-ordered BTO-pillar array and then the NZFO layer was deposited directly onto the nano-pillar array by PLD. Here, the well-ordered epitaxial BTO nanopillars will act as phase transition cores and combine with NZFO layer to trigger the TMS effect by thermal variation. BTO is chosen as a typical lead-free piezo/ferroelectric phase with several phase transition points that can provide large strain change with a very small temperature variation. For FM phase, NZFO is selected due to its easy-to-growth, lattice similarity with BTO, relative large magnetostriction and great potential of replacing YIG in RF/microwave devices.11,32,33 The NZFO/BTO-pillar nanocomposite demonstrates well-established epitaxial 3D heterointerface, which allows effective strain transfer at the 3D interface. The 3D structure nanocomposites 5 ACS Paragon Plus Environment

ACS Nano 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

can enable stronger TMS effect than those of double layer structures. A record high FMR field shift of 1866 Oe in NZFO/BTO-pillar nanocomposite occurs at the TMS critical angle near the BTO nanopillars phase transition within a narrow temperature range, which overwhelms the state of the art bilayer and 3D multiferroic nanostructures.15,25 The TMS effect can greatly enhance ME coupling at the interface, and the clamping effect is reduced due to the 3D structure. The 3D NZFO/BTO multiferroic heterostructure can be applied in thermal/voltage controllable RF/microwave devices such as filters, shifters and thermal sensors from engineering perspectives. RESULTS AND DISCUSSION

Crystal phases and Micrographs for 3D NZFO/BTO Multiferroic Heterostructure. Figure 1a shows the typical θ-2θ XRD pattern of NZFO/BTO-pillar nanocomposite film. It is clear that only the NZFO (00l) and BTO (00l) peaks are observed along with the STO (00l) peaks, suggesting a preferential out-of-plane orientation of (00l) for the spinel NZFO and proverskite BTO phases. The average c value is determined to be ~ 4.04 Å for the BTO phase. Simultaneously, ϕ-scans were conducted to investigate the in-plane orientation. As shown in the inset of Figure 1a, a comparison of the ϕ-scans of the NZFO (404), BTO (202) and STO (202) planes shows four sets of sharp peaks, indicating the cube-on-cube epitaxial growth for the 3D heterostructure as expected. Figure 1b shows the SEM image of BTO pillars on STO substrate before depositing NZFO, confirming that the well-ordered epitaxial BTO-pillar is successfully patterned on STO. The reciprocal space mapping (RSM) 6 ACS Paragon Plus Environment

Page 6 of 29

Page 7 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Nano

was used to obtain more details of the structural information, such as lattice parameters and strain state. Figure 1c exhibits the RSM patterns around the asymmetric diffractions of NZFO (206), BTO (103) and STO (103) for the 3D epitaxial structure. The (103) spot of BTO exhibits some extent of broadening, implying some variation in lattice constant, which can be attributed to the multiple BTO crystalline domains. Meanwhile the shape of BTO should not be ignored, BTO pillars before deposition NZFO is standing on STO substrates. However, a narrower peak is observed in the RSM data for the NZFO (206), which demonstrates the construction of 3D epitaxial NZFO. From the RSM results, the strain state of NZFO is 0.26% in-plane (IP) and 2.38% out-of-plane (OOP) relative to the BTO. The more detailed lattice parameters and strain state is listed in the Table S1. The lower level of IP strain between NZFO and BTO may give rise to the nearly perfect interfaces, which introduces the effective strain-mediated ME coupling. It is worth to mention that a high c/a ratio of 1.034 indicates the existence of the tetragonality in NZFO film due to epitaxial strain from the substrate and BTO-pillar. Figure 1d presents the AFM image of NZFO/BTO-pillar nanocomposite film. The as-prepared composite film exhibits ordered morphology of nano-arrays similar to that of the underneath BTO pillars. The magnification of SEM image for BTO nanopillar (Figure S1) exhibits an average lateral size of ∼90 nm and neighboring dot-dot distance of ∼120 nm. These results provide a direct evidence of the 3D lattice.

7 ACS Paragon Plus Environment

ACS Nano 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 1 (a) HRXRD pattern of NZFO/BTO-pillar/STO (001). The inset is the ϕ-scans of the NZFO/BTO-pillar. (b) The SEM image of BTO pillars on STO before NZFO deposition. (c) Reciprocal space mapping (RSM) of NZFO/BTO-pillar adjacent to the STO (103) reflection. (d) The AFM image of NZFO/BTO-pillar nanocomposite film.

Microstructure of 3D Interface. Cross-sectional transmission electron microscopy (TEM)/scanning transmission electron microscopy (STEM) and selected-area electron diffraction (SAED) pattern have been conducted to understand the growth mechanism and the microstructure of NZFO/BTO-pillar nanocomposite film. Figure 2a shows a typical low magnification STEM image of the film with BTO pillars covered by NZFO layer. The dark contrast represents the NZFO phase, whereas the light contrast indicates the BTO phase because the atomic Z number of Ba is much larger than that of Ni/Zn/Fe. The composite film thickness of ~ 150 nm, 8 ACS Paragon Plus Environment

Page 8 of 29

Page 9 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Nano

sharp BTO/NZFO interface and well-ordered surface morphology are consistent with the results from XRD and AFM measurement. As shown in Figure 2b, the phase separation and well-epitaxial orientation relationship are further confirmed by SAED from the area including the BTO pillars, NZFO film and the STO substrate. The apparent large splitting diffraction spots of NZFO, BTO and STO can be identified. The epitaxial orientation relationship of [100]BTO/[100]NZFO/[100]STO (IP) and [001]BTO/[001]NZFO/[001]STO (OOP) can also be determined. Furthermore, through the SAED results of the 3D nanocomposite film with high epitaxial quality, the diffraction spots of NZFO show divergent dots, demonstrating the existence of multiply stress state due to 3D interface and limited lattice mismatch. Figure 2c presents a STEM image viewed along the [100] of STO at the triple junction, where both NZFO and BTO phases meet the substrate, showing the perfect construction of the three clean and sharp interfaces among NZFO, BTO and STO. The inter-diffusion among these oxides and substrate is not observed because the fabrication process involves two separated steps. Figure 2d represents the atomic-scale STEM image for the interface between NZFO and BTO. It is surprising that the almost perfect heterointerface is observed without any dislocation defects, suggesting the high quality of fully strained 3D nanocomposite film. Simultaneously, the top interface of BTO pillar is similar to that of one side interface. The Fourier-filtered image for Figure 2d shows Moiré patterns whose periodicities corresponded to the periodicity of misfit dislocation cores observed along the boundary, as shown in Figure 2e. The misfit dislocation is absent in this area. In this case, the high stress state of NZFO 9 ACS Paragon Plus Environment

ACS Nano 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

remains in the NZFO/BTO-pillar without the process of stress relaxation from the dislocations. Meanwhile, these results are the most convincing evidence that the well-epitaxial relationship between NZFO and BTO were constructed in 3D structure.

Figure 2 (a) Cross-sectional low magnification STEM image of NZFO/BTO-pillar. (b) SAED pattern along the [100] zone axis of STO. (c) STEM image viewed along the [100] of STO at the triple junction of NZFO/BTO/STO. (d) The atomic-scale STEM image for the interface between NZFO and BTO. (e) The fast-Fourier filtered image of Figure (d). (f) Atomic-scale STEM image of epitaxial BTO-pillar/STO.

Interestingly, through analyzing the atomic-scale STEM image of NZFO, we found that regions marked with X & Y display different atomic arrangements whose area as small as 5 nm. The ideal NZFO crystal has an inverse spinel structure and a face-centered cubic oxygen sublattice, where zinc and nickel cations occupy the 10 ACS Paragon Plus Environment

Page 10 of 29

Page 11 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Nano

tetrahedral sites (A-sites) and octahedral sites (B-sites), respectively; while iron cations are evenly distributed both within the A-sites and the B-sites with formula unit (Zn1–x2+Fex3+)tetra[Nix2+Fe2–x3+]octaO4. Therefore, the octahedral and tetrahedral atom positions of an ideal spinel structure along are indicated in the red square, whereas the interstitial atoms are absent from the A-sites in the yellow square. For clearer illustration of the structure difference between the two regions, schematic diagrams of the crystal structure model was presented in Figure S2b/e, corresponding to both two regions. In addition, the FFT of the images further demonstrates the structural difference between these two regions, as shown in Figure S2c/f. We note that four diffraction spots in the yellow square are missing from the red square. These similar structure separations have been observed in epitaxial spinel ferrite films, which are defined as nanoscale structural separation.34,35 In this work, the formation of structure separation is dominated by the high compression strain state from the lattice mismatch between NZFO, BTO-pillar and STO substrate. It is worthy to mention that the anomalous magnetic behavior originates from the exchange interactions across the nanoscale structural separation. This anomalous magnetic behavior will be further discussed in below. Moreover, Figure 2f presents the interface of BTO and STO, which further confirms the epitaxial growth of the BTO nanopillars on STO substrate. The atomic-scale analysis of interface structures demonstrates the well-ordered epitaxial 3D nanocomposite. The presence of separated BTO nanopillars and NZFO phase is clearly demonstrated by STEM and EDS mapping, as shown in Figure S3. 11 ACS Paragon Plus Environment

ACS Nano 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 29

Piezoelectric and Magnetic Characteristics. Figure 3a illustrates the 3D topography AFM image of NZFO/BTO-pillar nanocomposite film on STO substrate in 1×1 µm2 area. Piezoresponse force microscopy (PFM) measurements confirmed the ferroelectric nature of the pillar. The well-defined saturated phase-voltage hysteresis and butterfly like amplitude-voltage hysteresis are shown in Figure 3b, indicating the ferroelectric polarization states which can be reversed with the bias voltage of ±8 V. Figure 3c shows the M-H loops along the in-plane (IP) and out-of-plane (OOP) directions. Both two loops are very slim with small coercive fields of δHOOP or δHIP/δHOOP>>δH(G). Therefore, the inhomogeneous linewidth broadening can be eliminated. Here, we constructed the high quality of 3D nanocomposite film with NZFO ferromagnetic phase and BTO ferroelectric phase. Moreover, the BTO pillars have been perfected coated by NZFO layer and the fully epitaxial interface has been demonstrated by STEM images. The bulk BTO crystals exhibit three distinct structural

phase

transitions:

cubic-tetragonal,

tetragonal-orthorhombic

and

orthorhombic-rhombohedral, respectively. Therefore, BTO should cause interface strain in NZFO layer at the structural phase transition temperatures, leading to a possible strain induced magnetic anisotropy change in the 3D NZFO/BTO-pillar nanocomposite film. Figure 4c shows the temperature dependence of FMR fields of the NZFO/BTO-pillar when the magnetic field is parallel with film plane at 0°(IP), 50° (the TMS critical angle) and 90°(OOP), respectively. FMR fields show gradually approaching constant, which is a normal temperature dependent FMR fields for the magnetic film. Meanwhile, the magnetic anisotropy of the films will gradually disappear near the Néel temperature. FMR fields of the film will become the same value whatever along IP or OOP. More importantly, the two anomalous regions of FMR fields can be observed at 256 K and 187 K along IP and OOP with decreasing temperature. NZFO phase has no 15 ACS Paragon Plus Environment

ACS Nano 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

phase transformation or magnetic characteristic temperature. However, BTO crystallizes in a perovskite structure undergoes two structural phase transitions on cooling process near 280 K and 183 K. Therefore, in our work, the FMR fields of 3D NZFO/BTO-pillar nanocomposite film shows significant jumps-drops around 255 K and 187 K, corresponding to the structural phase transitions of BTO. It is a known fact that BTO exhibits different lattice strains corresponding to different ferroelectric phases and this could be the driving force for modulating the magnetization of the NZFO in 3D NZFO/BTO-pillar nanocomposite. In this case, the effective strain is generated due to shrinking or expanding of the unit cell and might be transferred to NZFO, causing a systematic change in magnetic anisotropy with clear jumps-drops at structural phase transitions of BTO.

Figure 4 (a) FMR absorption spectra of NZFO/BTO-pillar for IP and OOP at room 16 ACS Paragon Plus Environment

Page 16 of 29

Page 17 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Nano

temperature. (b) Angular dependence of FMR linewidth of NZFO/BTO-pillar. (c) Temperature dependence of the FMR fields of NZFO/BTO-pillar at 0°, 50° (the critical angle) and 90°, respectively.

As shown in Figure 4c, the FMR field shifts at 0° (IP), 50° (the critical angle) and 90° (OOP) are 487 Oe, 1866 Oe and 165 Oe near the 255 K, respectively. Obviously, a giant FMR field shift of 1866 Oe occurs at the TMS critical angle near the BTO nanopillars phase transition, meanwhile the FMR field shifts at 0°, 50° (the critical angle) and 90° present large difference (∆H50°>∆H0°>∆H90°). This interesting phenomenon can be attributed to several aspects. Firstly, the BTO nanopillars have been coated by NZFO, while BTO nanopillars show small length/diameter ratio. In this case, it is very limited role to realize strain mediated ME coupling along OOP due to the phase change stress of BTO nanopillars. Secondly, BTO nanopillars can greatly enhance clamping effect on the NZFO layer due to the high-density epitaxial interface. During the BTO phase transition processes, the phase transition stress will play a critical role on the NZFO ferromagnetic layer, which give rise to the stress induced effective magnetic field. Thirdly, the TMS effect has been confirmed by the broadening FMR linewidth at the TMS critical angle in this work. It is believed that the strain mediated ME coupling is also greatly assisted by the TMS effect. Simultaneously, the 3D interface between BTO and NZFO without any dislocations establishes an angle about 50° with STO substrate. For comparison to the 3D NZFO/BTO-pillar nanocomposite, the temperature dependence of FMR field shifts of 17 ACS Paragon Plus Environment

ACS Nano 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

NZFO/BTO double layer is shown in the Figure S6. Although the FMR field shifts at the critical angle is larger than that of IP and OOP in the double layer, 3D NZFO/BTO-pillar nanocomposite exhibits a superior the strain mediated ME coupling. Therefore, along this TMS critical angle, the transferring of phase transition stress will more effective, meanwhile it triggers a so called spin lattice coupling (SLC) - TMS effect that enhance the FMR field shift greatly in the 3D nanocomposite.13

Figure 5 (a) Angular dependence of FMR field of NZFO/BTO-pillar at different temperature. Angular dependence of FMR field shifts between 255.5 K and 256 K when the magnetic field is 360° rotating.

Magnetic Anisotropy Near the BTO O/T Phase Boundary. In order to clearly demonstrate the angular-resolving FMR field information, Figure 5a presents the angle dependence of the FMR fields for 3D NZFO/BTO-pillar nanocomposite film around 256 K near the BTO O/T phase boundary. It is noted that the angle dependence of the FMR fields at 300 K and 257 K are almost coincident due to the small shift caused by changes in temperature. This result can also be found in the 18 ACS Paragon Plus Environment

Page 18 of 29

Page 19 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Nano

figure 4c. More importantly, with decreasing temperature to 255.5 K, we observed a marked tendency to increase FMR fields along the 50°, while a tendency to decrease the FMR fields near the IP can also be found. Figure 5b shows the angular dependence of the FMR field changes between 255.5 K and 257 K. Obviously, the FMR field changes (∆Hr) show distinctive four-folder like anisotropy, which is very different from the traditional strain mediated ME coupling dipole patterns.40 The ∆Hr at the TMS critical angles (50°, 130°, 230° and 310°) shows the largest value, meanwhile the ∆Hr at 0°, 50°and 90° present large difference (∆Hr0° ~ -223 Oe, ∆Hr50° ~ 1323 Oe and ∆H90° ~ 57 Oe ). At these TMS critical angles, the huge ∆Hr of NZFO/BTO-pillar indicates much larger strain mediated ME couplings during the BTO phase transition. Simultaneously, the FMR field shift of NZFO/BTO-pillar presents opposite direction along 0° and 90°, which means the opposite stress states in the NZFO along the two directions. Therefore, what really happened in the 3D-interface-mediated ME coupling could be more complicated than this phenomenological explanation. The phase transition stress at the 3D-interface and the SLC - TMS contribution show co-occurrence in NZFO/BTO-pillar nanocomposite, and the results have nevertheless revealed a strain mediated ME coupling behavior at nanoscale to greatly enhance the ME coupling in the nanocomposite. From practical perspective, the outstanding ME effect promises functional device paradigms such as the spin dynamics modulation by small thermal variation.

19 ACS Paragon Plus Environment

ACS Nano 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

CONCLUSION In summary, NZFO/BTO-pillar nanocomposite films were constructed by AAO template-assisted PLD. The well-established epitaxial 3D heterointerface has been verified by atomic-scale STEM, meanwhile 3D interface of NZFO/BTO-pillar presents an almost perfect interface without any edge dislocations. The angle-dependent FMR demonstrates the sharply increase linewidth at TMS critical angle due to the SLC - TMS effects, which is originated from 3D interface and nanoscale structural separation observed by STEM. A record high FMR field shift of 1866 Oe in NZFO/BTO-pillar nanocomposite film occurs at the TMS critical angle near the BTO nanopillars phase transition.

METHODS Fabrication Procedures for the NZFO/BTO-pillar nanocomposite BTO and NZFO targets were prepared by a conventional ceramic sintering method. An ultrathin AAO membrane with ∼90 nm pore size was transferred onto the STO substrate in a liquid environment. The epitaxial BTO was grown using PLD with a KrF excimer laser (λ = 248 nm) at 3 Hz on the AAO template covered STO substrate in a low oxygen pressure ranging from 10 to 30 mTorr at 800 °C. The distance between the target and substrates was 6 cm and the background vacuum was