Role of the magnetic anisotropy in organic spin valves 2017 Journal of Science Advanced Materials and Devices

Role of the magnetic anisotropy in organic spin valves 2017 Journal of Science Advanced Materials and Devices tài liệu,...

Role of the magnetic anisotropy in organic spin valves

V Kalappattil, R Geng, S.H Liang, D Mukherjee, J Devkota, A Roy, M.H Luong,

N.D Lai, L.A Hornak, T.D Nguyen, W.B Zhao, X.G Li, N.H Duc, R Das, S.

Chandra, H Srikanth, M.H Phan

PII: S2468-2179(17)30131-4

DOI: 10.1016/j.jsamd.2017.07.010

Reference: JSAMD 114

To appear in: Journal of Science: Advanced Materials and Devices

Received Date: 28 July 2017

Revised Date: 2468-2179 2468-2179

Accepted Date: 31 July 2017

Please cite this article as: V Kalappattil, R Geng, S.H Liang, D Mukherjee, J Devkota, A Roy, M.H Luong, N.D Lai, L.A Hornak, T.D Nguyen, W.B Zhao, X.G Li, N.H Duc, R Das, S Chandra,

H Srikanth, M.H Phan, Role of the magnetic anisotropy in organic spin valves, Journal of Science:

Advanced Materials and Devices (2017), doi: 10.1016/j.jsamd.2017.07.010.

This is a PDF file of an unedited manuscript that has been accepted for publication As a service to our customers we are providing this early version of the manuscript The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Role of the magnetic anisotropy in organic spin valves

V Kalappattil1,#, R Geng2,#, S.H Liang2, D Mukherjee1, J Devkota2, A Roy3, M.H Luong2,4, N.D Lai4, L.A Hornak5, T.D Nguyen2,*, W.B Zhao6, X.G Li6, N.H Duc7, R Das1, S

Chandra1, H Srikanth1, and M.H Phan1,*

(STO) substrate was manipulated by reducing film thickness from 200 nm to 20 nm

Keywords: LSMO; Magnetic anisotropy; Magnetoresistance; Organic Spintronics

#

Equal contributions to the work

* Corresponding authors: ngtho@uga.edu (T.D Nguyen); phanm@usf.edu (M.H Phan)

SrTiO3 (STO) substrate was manipulated by reducing film thickness from 200 nm to 20 nm

Substrate-induced compressive strain was shown to drastically increase the bulk in-plane magnetic anisotropy when the LSMO became thinner In contrast, the MR response of LSMO/OSC/Co OSVs for many organic semiconductors (OSCs) does not depend on either the in-plane magnetic anisotropy of the LSMO electrodes or their bulk magnetization All the studied OSV devices show a similar temperature dependence of MR, indicating a similar temperature-dependent spinterface effect irrespective of LSMO thickness, resulting from the orbital hybridization of carriers at the OSC/LSMO interface

Keywords: LSMO; Magnetic anisotropy; Magnetoresistance; Organic Spintronics

1 Introduction

La1-xSrxMnO3 (LSMO, x~0.33) is a very promising candidate material for spintronic

devices applications due to its chemical stability and intriguing physical properties [1-6] In particular, LSMO is a half-metallic ferromagnet that acts as an excellent spin injector/detector due to near 100% spin polarization at low temperatures [3,4] This material also provides manufacturing flexibility and cost-effectiveness, which are of practical importance [3]

Organic spintronics based on LSMO have attracted growing interest since Xiong et al

reported in 2004 a giant magneto-resistance (MR) of ~40% at 11 K in a LSMO/Alq3/Co spin

nanorods [9] This unusually large MR effect caused by a large effective Co spin polarization has been attributed to the spinterface effect, which is generally accepted to be strong in OSVs [8,10,11] The accomplished MR is also associated with the long life spin of electrons in organic materials As compared to inorganic semiconductor-based devices, the organic ones are cheaper and more mechanically flexible [3] However, it has been reported by several research groups that the MR of the LSMO/Alq3/Co device decreases drastically with an

increase in temperature (T < ~200 K) and reaches a relatively small value at room temperature (0.15−9%), rendering it undesirable for practical use [3,9,12-14] The MR temperature dependence of the LSMO/Alq3/Co device has remained an issue of long-lasting debate

[3,7,9,12-15] It was attributed to the reduction of the spin relaxation rate in the Alq3 layer [7]

or/and the weakening of spin polarization of ferromagnetic electrodes at high temperature

range (T > ~200 K) [14,16] Since the Curie temperature of Co (TC~1388 K) is much higher

than that of LSMO (TC~360 K), a considerably reduced spin polarization of LSMO near 300

K was naturally thought of as a possible cause for the observed small MR values [14] While the spin-1/2 photoluminescence detected magnetic resonance (PLDMR) study on Alq3

revealed a weak temperature dependence of spin lattice relaxation time [14], Drew et al

employed a low energy muon spin rotation method to study the temperature dependence of spin diffusion length, demonstrating that the spin relaxation in Alq3 dominated the MR

temperature dependence [17] Recently, Chen et al related the MR temperature dependence

to the spin relaxation in Alq3 for T < ~100 K, but to the surface spin polarization (or

spinterface) of LSMO for T > ~100 K [12] This seems to be supported by Majumdar et al who also observed a noticeable difference in the MR ratio for T > ~100 K between the two

LSMO/Alq3/Co devices made of LSMO films grown on SrTiO3 and MgO substrates [18]

Due to a larger lattice mismatch, a larger strain (9%) was reported in the LSMO film grown

on MgO as compared to the LSMO film grown on SrTiO3 (1%) [18] This suggests that the

observed MR difference for T > 100 K should not be simply related to the loss of spin

polarization carriers at the LSMO/Alq3 interface [18], but also due to the substrate-induced

strain effect [19] On the other hand, inorganic spintronics based on LSMO has been under investigation for a long time [1,2,4,20] and the magnetic anisotropy of LSMO has been reported to play a crucial role in controlling the performance of these devices [2,19,20] Unfortunately, to our best knowledge, no work has dealt with the role of magnetic anisotropy

of LSMO electrodes in organic spin valves

To address this important issue, we have performed the first comparative study of the bulk and surface magnetic properties of LSMO films with distinct thicknesses (20 nm and 200 nm), as well as the MR responses of LSMO/Alq3/Co devices using these LSMO films as

electrodes By reducing film thickness from 200 nm to 20 nm, the magnetic anisotropy of LSMO was drastically increased, due to the enhanced SrTiO3 (STO) substrate-induced strain

effect, allowing for probing effects of the magnetic anisotropy of the LSMO film on the MR response of OSV devices Our results indicate that instead of the in-plane magnetic anisotropy

of the LSMO electrode, the effective surface spin polarization at the OSCs/LSMO interface or

a spinterface effect plays an important role in the spin injection and transport in LSMO/Alq3/Co spin valve devices Our observation of the negligible influence of STO

substrate strain on the MR response indicates that the variation in substrate strain is not of significant concern while producing large numbers of OSV devices For achieving a larger

MR at room temperature, however, other types of ferromagnetic half-metals needs to be employed

2 Experiment

LSMO films of various thicknesses (200 nm, 50 nm, 20 nm) were grown epitaxially

on single-crystal SrTiO3 (STO) (100) substrates at 750 °C using magnetron sputtering

technique, with Ar and O2 flux in the ratio of 1:1 in a pressure 4 Pa The films were

subsequently annealed at 800 °C for 2 hours in flowing O2 atmosphere before slowly cooled

to room temperature For the OSV device fabrication, the LSMO films were subsequently patterned using standard photolithography and chemical etching techniques The Alq3 spacer

was thermally evaporated using an organic evaporation furnace with the evaporation rate of 0.5 Å /s at the base pressure of 2 × 10−7 torr; 15 nm cobalt (capped by 50 nm Al) top electrode

was deposited onto the Alq3 spacer using a shadow mask The obtained active device area was

typically about 0.2 × 0.4 mm2

2.2 Properties Characterization

The crystallinity and crystallographic orientations in the heterostructures were characterized by X-ray diffraction (XRD) (Bruker AXS D8 diffractometer equipped with high-resolution Lynx Eye position-sensitive detector using Cu Kα radiation) The in-plane epitaxy was determined from XRD azimuthal (ϕ) scans (Philips X’pert Diffractometer) The surface morphologies were observed using an atomic force microscope (AFM) (Digital Instruments III) Magnetic measurements were performed at different temperatures (10-350K) using a commercial Physical Property Measurement System (PPMS) (Quantum Design, Inc.)

in magnetic fields up to 5T Temperature dependent magnetic anisotropy measurements were measured by the radio-frequency transverse susceptibility (TS) using a RF tunnel diode oscillator (TDO) integrated into the PPMS The surface magnetic properties of LSMO films were studied by regular magneto-optic Kerr effect (MOKE) and balanced magneto-optic Kerr effect (B-MOKE) Magnetoresistance (MR) measurements on the OSV devices were conducted using the four probe technique in the presence of an in-plane magnetic field up to 3 kOe

3 Results and Discussion

First we examined the structure of the grown films and performed a substrate-induced strain analysis on them using the XRD technique Figures 1a and 1b show the XRD θ-2θ

patterns for the 200 nm and 20 nm LSMO films, respectively In both cases, only (00l) (l = 1,

2, and 3) diffraction peaks of the pseudo-cubic perovskite LSMO phase (JCPDS 4461) are observed along with the (00l) peaks of the STO substrates No secondary phase formations are present within the resolution limits of the instrument The small lattice

01-089-mismatch between bulk pseudo-cubic LSMO (lattice parameter, a = 0.389 nm) and cubic STO (a = 0.3905 nm) allows for the epitaxial growth of LSMO on STO as evident from the

close proximity of the XRD peak positions (inset to Fig 1a) Atomic force microscopy (AFM) images of the LSMO films are shown in insets of Fig 1a and 1b, with a similar root mean squared (RMS) surface roughness of ~0.6 nm While the double exchange mechanism alone cannot explain the magnetism of LSMO, a strong electron lattice coupling due to Jahn-Teller distortions via MnO6 octahedral deformation has been demonstrated to play an

important role [21] Although bulk manganites show small magnetic anisotropy, in thin films

it differs substantially from the bulk because of epitaxial strain in the films [22]

The average out-of-plane (a⊥) and in-plane (aװ) lattice parameters of the LSMO films with 200 nm and 20 nm thicknesses grown on STO (100) substrates were calculated from the symmetric θ-2θ scans (performed about the LSMO (100), (200), (300) pseudo-cubic planes) and the asymmetric 2θ-ω (or detector) scans (performed about the LSMO (110) and (211) planes), following the same method as detailed in previous works [23,24] A representative detector scan is shown in the inset of Fig 1b for the 20 nm LSMO film performed about the LSMO (211) plane The lattice parameters obtained from the XRD analysis were calculations

were aװ =0.381 (±0.003) nm and a⊥= 0.393 nm (±0.001) for the 20 nm film and those for the

200 nm films were aװ = 0.392 (±0.002) nm and a⊥= 0.391 nm, respectively Difference

between the in-plane and out-of-plane lattice parameters of each film indicates the lattice

distortion during their growing In-plane strain was calculated by using the formula (aװ –

a o )/ao, where ao is the bulk lattice parameter of LSMO as measured from the XRD powder

diffraction [23] A large compressive strain of – 0.020 is found in the 20 nm LSMO film, while a relatively small tensile strain of 0.005 is seen in the 200 nm LSMO film Also by

comparing the aװ and a ⊥ values, we see that the 20 nm LSMO film undergoes a larger

tetragonal distortion due to lattice induced strain This is in accordance with the dependent epitaxial strain study conducted by the other research group [19] In the present study, we aimed to understand how this strain variation would affect the surface and bulk magnetic properties of the LSMO films and hence the MR responses of the SV devices using these LSMO films as ferromagnetic electrodes

thickness-To understand how surface magnetic properties of LSMO films differ from their bulk ones, magneto-optical Kerr effect (MOKE) measurement and vibrating sample magnetometer (VSM) measurements were performed over the temperature range of 10-350K We recall that the MOKE technique is an excellent tool for studying the surface magnetization as it is highly sensitive to the magnetization within the skin depth region of 10-15nm in most materials Figures 2a and 2b display the MOKE data at two selected temperatures (127 K and 215 K) for the 200 nm and 20 nm LSMO films, respectively The coercive field (HC) of each film was

determined from the MOKE loop and the temperature dependence of HC for both films are

plotted in Fig 2c As compared to the 20 nm LSMO film, the values of HC (close to values of

the surface magnetic anisotropy field) are larger for the 200 nm LSMO film, especially in the high temperature range Figures 2d and 2e show the VSM loops at two selected temperatures (50 K and 300 K) for the 200 nm and 20 nm LSMO films, respectively The temperature dependences of HC for both films are also plotted in Fig 2f We note that the HC value of the

200 nm LSMO film obtained from MOKE (Fig 2c) is similar to that obtained from VSM at

To quantify the effective anisotropy field value (HK) and its temperature evolution, we

measured the radio-frequency (RF) transverse susceptibility (TS) of both LSMO films.25 This

TS method has been validated by us as an excellent tool for studying the anisotropic magnetic properties of a variety of systems from thin films [26] to single crystals [27] and nanoparticles [28] Here the sample is kept inside an LC circuit which is self-resonant around 12 MHz with

a sensitivity of 1-10 Hz in 10 MHz When a small RF field is applied perpendicular to the sweeping DC field, change of the resonant frequency (∆fres) is directly proportional to the change of the magnetic susceptibility (∆χT) in the transverse direction: ∆χT/χT ∝∆fres/fres As

theoretically predicted by Aharoni et al [29] for a Stoner-Wohlfarth particle with its magnetic

hard axis aligned parallel to the DC field, TS spectra should yield peaks at the anisotropy fields (±HK) and switching fields (±HS) as the DC field is swept from positive to negative saturation TS curves taken at different temperatures for the 200 nm and 20 nm LSMO films are displayed in Fig 3a and 3b, respectively For comparison of the ∆χT and HK, the TS spectra taken at the same temperature of 20 K are plotted in Fig 3c for both films We note that for the present LSMO films the switching peak is merged with the anisotropy peak, thus causing a difference in the positive and negative HK values as well as a slight asymmetry in