Exchange Bias Effect along Vertical Interfaces in La0.7Sr0.3MnO3:NiO

Subscriber access provided by UNIV OF DURHAM

Article

Exchange Bias Effect along Vertical Interfaces in La0.7Sr0.3MnO3:NiO Vertically Aligned Nanocomposite Thin Films Integrated on Silicon Substrates Jijie Huang, Adam Gellatly, Alexander Kauffmann, Xing Sun, and Haiyan Wang Cryst. Growth Des., Just Accepted Manuscript • DOI: 10.1021/acs.cgd.8b00366 • Publication Date (Web): 15 Jun 2018 Downloaded from http://pubs.acs.org on June 21, 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.

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 21 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

Crystal Growth & Design

Exchange Bias Effect along Vertical Interfaces in La0.7Sr0.3MnO3:NiO Vertically Aligned Nanocomposite Thin Films Integrated on Silicon Substrates

Jijie Huang1, Adam Gellatly1, Alexander Kauffmann1, Xing Sun1 and Haiyan Wang1, 2* 1

School of Materials Engineering, Purdue University, West Lafayette, IN 47907, USA School of Electrical and Computer Engineering, Purdue University, West Lafayette, IN 47907, USA Email: ([email protected]) 2

Keywords: Si integration, exchange bias, vertically aligned nanocomposite (VAN)

Silicon (Si) integration is a critical step towards future applications of multifunctional oxides as nanoscale electronics and spintronic devices, because of the low cost and scalability of silicon substrates. As a demonstration, self-assembled (La0.7Sr0.3MnO3)x: NiO1-x (LSMO:NiO) vertically aligned nanocomposite (VAN) thin films with exchange bias (EB) properties have been successfully deposited on buffered Si substrates. To enable the epitaxial growth of LSMO:NiO VAN, SrRuO3/TiN was first grown as the buffer layers on Si substrates. The composition of the 2-phases has been varied with x=0.25, 0.5, 0.75 and 1 to explore the electrical transport and magnetic properties of the VAN system on Si. The irreversible temperature Tirr was found to increase with increasing NiO composition, with the highest for (LSMO)0.25(NiO)0.75 of ~275 K, when the field was applied in the out-of-plane direction. In addition, the EB effect has been observed for all the nanocomposite films, with the highest HEB value of 300 Oe for (LSMO)0.25(NiO)0.75. The integration of the VAN system on Si with pronounced EB properties presents a promising approach towards future practical devices using oxide VANs on Si.

ACS Paragon Plus Environment

Crystal Growth & Design 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 2 of 21

INTRODUCTION Exchange bias (EB) is the phenomenon that the magnetization curve of a ferromagnetic thin film shifts from origin by the spin interactions between ferromagnetic (FM) and antiferromagnetic (AFM) layers.1-3 EB has been widely used in various applications, such as magnetic memory, magnetic field sensors and recording heads for hard disk drive.4-6 To date, most of the exchange bias effect has been demonstrated as in-plane FM-AFM pinning effect in bilayer or multilayer fashion.7-11 For example, apparent EB effect has been detected in various FM-AFM material systems fabricated by different techniques, such as magnetron sputtered Co/CuMn,8 pulsed laser deposition (PLD) deposited La0.7Sr0.3MnO3/BiFeO3,7, 9 as well as the molecular beam epitaxy (MBE) deposited Fe/FeO.10 However such in-plane exchange bias effect has strong thickness dependence, i.e., as predicted by the traditional theory of ‫ܪ‬ா஻ ∝ 1ൗ‫ݐ‬

ிெ

(tFM

is the thickness of the FM layer).12 For example, HEB value of IrMn/Co bilayer was found to follow a smooth 1/tCo curve where the HEB reduces with increasing Co layer thickness and to near zero beyond particular thickness.13 Taking advantages of the unique vertical interface structure of vertically aligned nanocomposites (VAN),14-16 the concept of EB along vertical interface has been introduced with its potential to overcome such thickness limitation by coupling a FM oxide and an AFM oxide along their vertical interfaces.16, 17 EB effects along vertical interface have been demonstrated in several VAN systems17-20 and are considered more attractive for the next-generation high-density non-volatile memory devices, which requires good thermal stability and high density.21, 22 However, most of the VAN systems were obtained on single crystal oxide substrates (e.g. SrTiO3),17-20 which are expensive and limited for large-scale device integration. Silicon is an ideal substrate for future device applications and the integration of oxide-based devices with


Page 3 of 21 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


complementary metal-oxide semiconductor (CMOS) devices. To enable the integration of oxides on Si-substrates, much effort has been devoted to overcome the challenges of the epitaxial growth of oxides on Si. Different buffer layers have been developed to establish the epitaxial relationship under such large mismatch between the oxides and the Si substrates and to minimize the inter-diffusion between the oxides and Si.23-25 Specifically, BiFeO3-CoFe2O4 nanocomposite thin film has been demonstrated on Si substrate with PLD deposited Sr(Ti0.65Fe0.35)O3/CeO2/YSZ buffer layers, or molecular beam epitaxy (MBE) grown STO.23 In this work, FM La0.7Sr0.3MnO3 (LSMO) and AFM NiO are selected as the novel FMAFM VAN system for the demonstration of EB oxides integrated with Si substrates. The LSMO:NiO system on STO substrate has been demonstrated to have high epitaxial quality and obvious exchange bias effect.17, 20 Through the proper design of the buffer layer stack, it is believed that this system could be a good candidate to integrate on Si to explore its exchange bias effect. A bilayer buffer structure SrRuO3 (SRO)/TiN was adopted to establish the epitaxial growth of oxides on Si. TiN is selected because of its domain epitaxy growth with Si, i.e., 4 of TiN (002) matching with 3 of Si (004) enables a direct cube-on-cube growth despite the very large lattice mismatch (f~24%),26 as illustrated in the right panel of Figure 1. SRO is selected to further reduce the lattice parameter from 4.24 Å to 3.93 Å to enable the high quality growth of the oxide VAN. The SRO/TiN buffer stack also provides a conducting bottom electrode for the electrical measurements. Illustrated in the left panel of Figure 1 is the overall 3-D structure of the FM-AFM VAN with NiO nanopillar embedded in LSMO matrix on buffered Si. The exchange coupling occurs at the LSMO (FM) /NiO (AFM) vertical interface areas, and produces the EB effect along the vertical interface. A composition dependent study is conducted to explore



the composition of AFM phase on the overall microstructural, magnetic and electrical transport properties of (LSMO)x:(NiO)1-x VANs on Si.

Figure 1. Schematic illustration of LSMO-NiO nanocomposite integrated on SRO/TiN buffered Si substrate; the right panel shows the lattice / domain matching between the buffers and Si substrate, and the upper panel shows the out-of-plane matching between the LSMO and NiO phases. Experimental Section Thin Film Growth: Pure LSMO thin film, as well as LSMO:NiO with different compositions were deposited by a PLD system with a KrF excimer laser (Lambda Physik Compex Pro 205, λ=248 nm, 5 Hz) on SRO/TiN buffered Si substrates. The laser energy density was 3 J/cm2. The deposition temperature was kept at 700°C, and the target-substrate distance was 4.5 cm for all


Page 4 of 21

Page 5 of 21 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


the depositions. The base pressure for all the depositions reached less than 1×10-6 Torr in vacuum. The multilayer thin films were prepared by alternative laser ablation of the TiN, SRO and LSMO:NiO nanocomposite targets. Microstructure and Physical Property Characterizations: The microstructure of the ﬁlms was characterized by X-ray diffraction (XRD) (Panalytical X’Pert X-ray diffractometer) and transmission electron microscopy (TEM) (FEI Talos-200X). Resistance in dependence of temperature (R-T) measurement was conducted by physical property measurement system (PPMS: Quantum Design). Temperature dependence of ZFC and FC magnetization and magnetic hysteresis curves were done by a SQUID magnetometer (MPMS: Quantum Design).

RESULTS and DISCUSSION Figure 2 shows the typical θ-2θ XRD scans of (LSMO)x:(NiO)1-x nanocomposite thin films with different LSMO compositions (x=0.25, 0.5 and 0.75), as well as a pure LSMO film as reference, all on SRO/TiN buffered Si substrates. First, for the SRO/TiN buffer layer, SRO (00l) peaks can be identified, which indicates the highly textured growth of the SRO layer on Si. However, no obvious TiN peak was observed, while TiO2 (004) peak was identified, possibly due to the partial oxidation of TiN layer during deposition of SRO buffer layer and the following VAN layer. Both NiO (002) and LSMO (002) peaks were identified in (LSMO)x:(NiO)1-x nanocomposite thin films, which indicates the formation of both textured NiO and LSMO phases. This further demonstrates that the epitaxial growth of SRO/TiN buffer can provide an effective template for the following VAN epitaxial growth. Furthermore, the peak intensity of NiO (002) increases with decreasing x, while LSMO (002) peak intensity decreases accordingly.



Figure 2. θ-2θ XRD scans of LSMO:NiO thin films with different compositions, as well as a pure LSMO sample as the reference.

To explore the microstructure characteristics of the LSMO:NiO nanocomposites on Si substrate, a detailed TEM/STEM analysis was carried out. Figure 3(a) shows a lowmagnification cross-sectional TEM image of a (LSMO)0.25(NiO)0.75 film. Two buffer layers and the VAN layer can be clearly identified and NiO nanopillars are embedded in LSMO matrix structure. The corresponding selected area electron diffraction (SAED) pattern is shown in the inset of Figure 3(a) to identify the epitaxial growth and the orientation relationships between LSMO and NiO in the nanocomposite and the underlying SRO layer, which are LSMO (002) ∥ NiO (002) ∥ SRO (002) and LSMO [200] ∥ NiO [200] ∥ SRO [200]. The high-resolution TEM image of LSMO/NiO interface area in Figure 3(b) confirms the epitaxial matching relationship and the high epitaxial quality of the nanocomposite film. Scanning transmission electron microscopy (STEM) image in Figure 3(c) and its corresponding EDX mapping image shown in


Page 6 of 21

Page 7 of 21 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 3(d) further demonstrate the well separated NiO and LSMO phases without obvious interdiffusion. Microstructure of the (LSMO)0.5(NiO)0.5 film was shown in Figure S1, lower density NiO nanopillars with smaller size were observed, because of the less amount of NiO in the sample compared to the case of x=0.25 above.

Figure 3. TEM micrographs of a typical (LSMO)0.25(NiO)0.75 nanocomposite thin film on Si substrate. (a) Cross-sectional low magnification TEM image to show the overall film stack and its corresponding selected area diffraction pattern as the inset; (b) high resolution TEM image taken at the interface area to show the epitaxial quality of both LSMO and NiO phases; (c) STEM image and (d) its corresponding EDS mapping to show the 2-phase VAN structure.



Figure 4. Magnetic hysteresis curves of the pure LSMO and LSMO:NiO films after 1 T FC to 5 K in (a) OP and (c) IP directions. (b) and (d) are the enlarged regimes of L0.25N0.75 and L0.75N0.25, to show the bias shift; (e) Heb and (f) Hc in dependence of NiO ratio in both OP and IP directions. To explore the EB effect of the LSMO:NiO nanocomposite thin films, magnetization hysteresis measurement was carried out for all the nanocomposite films, and the pure LSMO film for comparison. The measurements were taken after a field cooling of 1 T down to 5 K, both


Page 8 of 21

Page 9 of 21 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


OP and IP measurements were conducted, as shown in Figure 4(a) and 4(c), respectively. Figure 4(b) and 4(d) are the enlarged plot near zero field regime for L0.75N0.25 and L0.25N0.75, to show the horizontal shift of the hysteresis loop. Here, the HEB value can be calculated by HEB= |H+ + H−|/2 (where H+ and H− represent the positive and negative values of coercivity as the magnetization goes to zero). Specifically, L0.25N0.75 exhibits high HEB values of 300 Oe and 175 Oe for OP and IP measurements, respectively. L0.5N0.5 obtains a HEB (OP) value of 275 Oe, which is higher than the previous reported value of 77 Oe for the L0.5N0.5 on STO substrate.17 It is noted that the measurement of the HEB values was conducted under 5 K for the samples on Si substrates in this work, while the measurement was done at 10 K for the STO ones.17 The results suggest that the Si integration could even lead to enhanced physical properties in certain nanocomposites compared to those on STO substrates. The HEB values of all the samples can be obtained from the M-H curves in Figure S2, and the results are plotted and compared in Figure 4(e). When the applied field was perpendicular to the film surface, as illustrated in the inset in Figure 4(e), the M-H curves of all the nanocomposite films exhibit an obvious horizontal shift, with the highest value of 300 Oe for L0.25N0.75 and no shift for the pure LSMO film. The general trend is that the HEB value increases with increasing NiO composition, as more vertical FM/AFM interface was generated with increasing NiO ratio and thus leads to a stronger FM-AFM coupling. Furthermore, with more NiO, an increased out-of-plane tensile strain is applied on LSMO because of the larger lattice parameter of NiO (a=4.17 Å) compared to that of LSMO (a=3.88 Å). Such enhanced tensile strain on the LSMO domains could stabilize the FM ordering in LSMO and results in higher exchange bias value.27 However, no EB shift was observed when the measurement was taken after zero-field cooling (ZFC), as shown in Figure S3. Furthermore, the



IP HEB values of the nanocomposite films are lower than their OP HEB values, regardless of the different compositions. The reason is that the spins of NiO align along [112ത] direction,28 as illustrated in Figure 5(a) and 5(b), and therefore the OP spin component is larger than the IP spin component, which results in larger OP HEB values. The result demonstrates the anisotropic nature of the LSMO:NiO films and the EB effect dominated along the vertical interface. Furthermore, coercivity (HC) values of all the samples were also compared in Figure 4(f), which was calculated by HC= |H+ - H−|/2. The general trend is that OP HC is higher than its corresponding IP HC, which further confirms the anisotropic nature of the nanocomposite thin films with the preferred out-of-plane anisotropy. Overall, the LSMO:NiO nanocomposite thin films on Si exhibit obvious EB effect by the FM-AFM vertical interface coupling and demonstrate enhanced coercivity, which suggests the VAN on Si is very promising for potential applications such as read heads in magnetic storage devices with reduced dimensions. To further understand the magnetic spin structures, the magnetization as a function of temperature (M-T) under both zero-field cooling (ZFC) and 1000 Oe field cooling (FC) was measured for all the samples, in both OP and IP directions, as shown in Figure 5(c) and 5(d), respectively. First, the Curie temperature TC for pure LSMO is ~355 K, for both OP and IP directions, which is close to previous reports.29, 30 The TC value of LSMO:NiO VANs slightly decreases to ~315 K, possibly due to the lower degree of magnetic ordering after introducing NiO in the film,31 as well as the lattice mismatch strain from LSMO-NiO interfaces, which causes the distortion of MnO6 octahedra and thus suppresses the TC.32 More interestingly, when the field was applied OP, two bifurcations between the FC and ZFC curves (which determines the irreversible temperature Tirr) were observed in the VAN films, while only one bifurcation appears in that of the pure LSMO film. Obviously, the lower one at ~140 K is attributed to the


Page 10 of 21

Page 11 of 21 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


LSMO phase, which indicates that the LSMO phase is lack of long-range FM ordering and becomes possible spin-glass states, spin clusters or superparamagnets under such temperature. The higher Tirr in the nanocomposite films is believed to be originated from the spin-frustration in the FM-AFM interface area, evidenced from that its value increases with increasing NiO composition, e.g., (LSMO)0.25(NiO)0.75 of ~275 K. With more NiO included, the motion of the domain wall in LSMO under magnetic field was hindered and thus Tirr increases.33 Such interface ferromagnetic contributions could be from the following: (1) the finite-size effect could result in magnetic NiO due to the ultra-thin NiO nanopillars (

Exchange Bias Effect along Vertical Interfaces in La0.7Sr0.3MnO3:NiO

Recommend Documents