Effect of Orbital Hybridization on Spin-Polarized Tunneling across Co

Subscriber access provided by Northern Illinois University

Article

The effect of orbital hybridization on spinpolarized tunneling across Co/C interfaces 60

Kai Wang, Elia Strambini, Johnny G. M. Sanderink, Thijs Bolhuis, Wilfred G. van der Wiel, and Michel P. de Jong ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.6b08313 • Publication Date (Web): 14 Sep 2016 Downloaded from http://pubs.acs.org on September 21, 2016

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 free 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 accessible to all readers and 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 Applied Materials & Interfaces 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 27

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 Applied Materials & Interfaces

The Effect of Orbital Hybridization on Spin-Polarized Tunneling across Co/C60 Interfaces †,*

Kai Wang, , Elia Strambini, Johnny G. M. Sanderink, Thijs Bolhuis, Wilfred G. van der Wiel and Michel P. de Jong* NanoElectronics (NE) Group, MESA+ Institute for Nanotechnology, University of Twente, P.O. Box 217, Enschede, 7500AE, The Netherlands [†]

Present Address Key Laboratory of Luminescence and Optical Information, Ministry of Education, Beijing Jiaotong University, No.3 Shang Yuan Cun, Haidian District, Beijing, China

AUTHORS INFORMATION Corresponding Authors *Kai Wang E-Mail: [email protected] *Michel P. de Jong E-Mail: [email protected]

KEYWORDS C60, organic spintronics, spinterfaces, hybrid interfacial density of states, tunneling anisotropic magnetoresistance (TAMR), interfacial magnetic moments, Co/C60 hybrid interfacial states, spin-dependent density of states.

1 ACS Paragon Plus Environment


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

ABSTRACT The interaction between ferromagnetic surfaces and organic semiconductors leads to the formation of hybrid interfacial states. As a consequence, the local magnetic moment is altered, a hybrid interfacial density of states (DOS) is formed, and spin-dependent shifts of energy levels occur. Here, we show that this hybridization affects spin transport across the interface significantly. We report spin-dependent electronic transport measurements for tunnel junctions comprising C60 molecular thin films grown on top of face-centered-cubic (fcc) epitaxial Co electrodes, an AlOx tunnel barrier, and an Al counter electrode. Since only one ferromagnetic electrode (Co) is present, spin-polarized transport is due to tunneling anisotropic magnetoresistance (TAMR). An in-plane TAMR ratio of approximately 0.7% has been measured at 5 K under application of a magnetic field of 800 mT. The magnetic switching behavior shows some remarkable features, which are attributed to the rotation of interfacial magnetic moments. This behavior can be ascribed to the magnetic coupling between the Co thin films and the newly formed Co/C60 hybridized interfacial states. Using the Tedrow-Meservey technique, the tunnel spin polarization of the Co/C60 interface was found to be 43%.


Page 2 of 27

Page 3 of 27

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


INTRODUCTION The introduction of organic semiconductors (OSCs) in spintronics has triggered a large number of studies into the electronic and magnetic properties of OSC-ferromagnet hybrid interfaces. It is well established by now that it is crucial to properly engineer the hybrid interfaces in order to achieve robust spin injection.1-3 Such interfaces are characterized by a unique hybrid interfacial electronic structure due to the strong overlap of molecular orbitals with the electronic wave functions of the ferromagnetic surface. Owing to the net spin polarization of electronic states of the ferromagnet, the hybridization process can transfer spin-polarized electrons from the ferromagnet to the molecules. These newly formed spindoped interfacial states exhibit magnetic properties and a net spin polarization. Tuning the interfacial properties via the selection of appropriate combinations of ferromagnetic surfaces and molecules is the essence of a new research field coined “spinterface science”.4 The pure carbon allotrope buckminsterfullerene (C60) has been considered as a promising candidate for spintronics.5,6 This is due to the expectation of a long spin lifetime, resulting from very weak spin-orbit coupling (SOC), and zero nuclear spin of the majority 12C isotope (98%), such that the hyperfine interaction is minute. Recent experimental and theoretical studies have shown that, in addition to these intrinsic attributes, various metal/C60 interfaces show extraordinary magnetic properties that may be exploited for spintronics.7-11 A salient example is the observation of magnetism in multilayers of thin films of C60 and nonmagnetic metals, Cu and Mn.7 In addition, x-ray absorption spectroscopy (XAS) and x-ray magnetic circular dichroism (XMCD) measurements of ferromagnet/C60 interfaces have revealed quite remarkable results.10,12,13 The C K-edge XAS and XMCD spectra of a C60 monolayer grown on a thin epitaxial film (several nanometers) of body-centered-cubic (bcc) Fe (001) have shown significant hybridization of Fe and C60 continuum states.12 Further investigations of a monolayer C60/5-monolayer Fe(001)/W(001) multi-layer structure showed a reduction of 6% of the spin magnetic moments of the top Fe surface, while antiparallel 3 ACS Paragon Plus Environment


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

magnetic moments of -0.2µB on the C60 molecules were induced.13 Recently, Moorsom and co-authors reported that C60 becomes ferromagnetic as a result of spin doping from Co,10 while Bairagi et al. showed that orbital hybridization between C60 and Co (0001) leads to a considerable increase of the perpendicular magnetic anisotropy of Co thin films.9 Very interestingly, spin-polarized scanning tunneling microscopy (SP-STM) was utilized to demonstrate a spin-filter effect for C60 molecules adsorbed on a Cr (001) surface.11 Regarding organic spintronic devices, early studies have demonstrated sizable tunneling magnetoresistance (TMR) in organic spin valves even at room temperature.5 The properties of the hybrid interfaces in such devices play an important role in their operation. This has been suggested by, for example, a study of organic spin valves with metallic nanocontacts made by a nano-indentation method, which showed a sign reversal of spin polarization due to spin-dependent hybridization at the organic/ferromagnet interface.14 In addition to spin valves, another class of spin tunneling devices has been studied that relies on a single ferromagnetic contact, an organic spacer (with or without additional tunnel barrier), and a non-magnetic counter-electrode.15 In this configuration, the conventional TMR effect is absent, but tunneling anisotropic magnetoresistance (TAMR), i.e. a change in resistance upon changing the direction of magnetization of the ferromagnetic contact, can be observed under appropriate conditions. TAMR has been mainly studied in inorganic spintronic systems, such as Co/AlOx/Al, GaAs/(Ga,Mn)As/AlOx/Ti/Au and Fe/GaAs/Au heterojunctions.16-18 The origin of the TAMR effect is attributed to the anisotropy of the SOC, such that the spindependent density of states (DOS) is modulated by variation of the magnetization direction along different crystallographic axes of a magnetic layer. In some systems, the superposition of Bychkov-Rashba and Dresselhaus SOC, generated by device structural asymmetry along the growth direction and bulk inversion asymmetry of the semiconductor spacer (e.g. GaAs), respectively, can be applied to enhance the TAMR effect.16


Page 4 of 27

Page 5 of 27

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


In this work, we show that the hybrid electronic states formed at the interface between a ferromagnet, fcc-Co (111), and a molecular semiconductor, C60, produce distinct tunneling anisotropic magnetoresistance when incorporated into a tunnel junction. The newly formed Co/C60

interfacial

layer

produces

a

robust

magnetoresistance

effect

(interfacial

magnetoresistance, IMR). Figure 1 shows the device structure, more details about its fabrication can be found in the experimental part. The spin transport phenomena will be investigated based on two experimental characterization methods: (1) TAMR measurements, and (2) measurements of the tunnel spin polarization using the Tedrow-Meservey technique. We will demonstrate a distinct IMR effect, showing hysteretic magnetoresistance exhibiting several metastable resistance states, and TAMR ratios up to 0.7%. The content of this paper is organized as follows. The following section contains experimental details. After this, experimental results and discussions of the spin transport phenomena in sapphire(substrate)/fcc-Co(8 nm)/C60(4 nm)/AlOx(3.3 nm)/Al(35 nm) systems and Tedrow-Meservey measurements of Si(substrate)/SiO2(300 nm)/Al(6.7 nm)/AlOx(3.3 nm)/C60(4 nm)/fcc-Co(30 nm) devices will be provided. A conclusion is given in the last section.

EXPERIMENTAL SECTION Device Fabrication Tunnel junctions were fabricated following a procedure described in detail elsewhere.18,19 In short, the devices were patterned via shadow masking. Co electrodes with thickness of 8 nm were deposited by electron beam (e-beam) evaporation onto singlecrystalline sapphire (0001) substrates held at room temperature. A 4 nm C60 layer was thermally evaporated onto the Co electrodes, and a 3.3 nm AlOx tunneling barrier (TB) was deposited on top of the C60 layer by e-beam deposition of Al2O3 source material. The AlOx 5 ACS Paragon Plus Environment


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

was used as a buffer layer, to prevent the formation of metallic shorts between the counterelectrodes separated by a C60 layer of merely 4 nm. To define 250 µm wide strips on the Co/C60/AlOx structures where the junction area is to be formed, 30 nm thick AlOx slabs were deposited through a shadow mask. Finally, 35 nm thick and 300 µm wide Al strips were deposited on top, orthogonal to the slabs, forming the counter-electrodes. The structure thus contains a junction area of 250 µm × 300 µm. The overall structure is depicted in Fig.1 (a) and (b). Similar devices were prepared, for which, after depositing a 3.3 nm AlOx tunneling barrier, the devices were transferred into an integrated load-lock system and subjected to 2 min oxygen plasma at room temperature. This improves the quality of the AlOx tunnel barrier, but at the same time drastically affects the C60/Co interface (see Figure S3, S4 and S5 of the Supplementary Information). The Tedrow-Meservey technique relies on tunneling into a superconducting Al thin film. Zeeman splitting of the superconducting DOS is achieved by applying an in-plane magnetic field of a few Tesla, such that the electronic DOS of Al is completely spin polarized near the edges of the superconducting energy gap. Owing to the critical requirement for obtaining an ultrathin (i.e. < 10 nm) and very smooth superconducting Al layer, such that the (in-plane) critical field is sufficiently high (i.e., around Bc = 3 T),20 the device structure was modified by reversing the growth order, as: Si(substrate)/SiO2(300 nm)/Al(6.7 nm)/AlOx(3.3 nm)/C60(4 nm)/Co(30 nm). The junction area was 50 µm × 50 µm. During the fabrication, the Al thin film, with a thickness of 10 nm, was deposited onto the Si/SiO2(300 nm) substrate held at a temperature close to that of liquid nitrogen (i.e. ~77 K, maintained by keeping the sample holder in the vicinity of a liquid-nitrogen cooled baffle). Then, the sample was transferred to the load-lock chamber, where it was subjected to an oxygen plasma for 30 min under an oxygen pressure of approximately 100 mTorr. This produces a ~3.3 nm thick AlOx tunneling barrier. The sample was transferred back into the growth chamber, and after the


Page 6 of 27

Page 7 of 27

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


substrate temperature approached 77 K again the 4 nm C60 and 30 nm Co layers were deposited.

Magnetotransport Measurements The spin transport measurements were performed using a liquid-helium flow cryostat, equipped with an electromagnet (maximum field strength 1 T). A four-wire measuring method was utilized to minimize the contributions from the electrode resistances. The devices were mounted onto a sample holder that enables 360o in-plane rotation. Spin-valve-like magnetoresistance signals were measured by sweeping the magnetic field within a ±100 mT range, while applying a constant dc current through the junction. Angle-dependent TAMR measurements were carried out by applying a sufficiently large in-plane magnetic field (800 mT) to saturate the magnetization of the fcc-Co layer along the external field direction; meanwhile, V-I curves were recorded at different angles (i.e., θ) ranging from 0o to 360o with a nanovolt-meter at 5 K. The corresponding TAMR as a function of θ has been calculated from: =

− 0 × 100% , 1 0

where the reference direction (i.e., 0 degrees) is defined as the [110] direction of the fcc-Co thin film, which is also its magnetic easy axis.18,19

Tedrow-Meservey Measurements For Tedrow-Meservey measurements a 3He refrigerator was used (i.e., Oxford Instruments Heliox VL System) which can reach a base temperature of 260 mK. This satisfies the criterion ⁄ < 1, i.e. the temperature is sufficiently far below the critical temperature to maintain the superconducting state under application of an in-plane magnetic field of a few Tesla, in order to achieve a Zeeman splitting of a few hundred µeV at B = 3 T. Lock-in techniques, 7 ACS Paragon Plus Environment


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

combined with a low noise electronics setup (see Reference 21)21 were used to maximize the signal to noise ratio. During the measurements, the magnetic field (generated by a superconducting magnet) was swept with a step-size of 100 mT. For each increment of the magnetic field, the corresponding differential conductance ⁄ was recorded within a bias window of ± 1 mV.

RESULTS AND DISCUSSION Figure 2 shows the temperature dependence of the I-V curves for the device with structure sapphire(substrate)/Co(8 nm)/C60(4 nm)/AlOx(3.3 nm)/Al(35 nm). Positive (negative) bias corresponds to electrons flowing from occupied (unoccupied) Co/C60 hybrid states to unoccupied (occupied) Al states (this sign convention is applied for all measurements). All the I-V curves of Figure 2 show clear quasi-symmetric and non-linear behavior. The junction resistance is moderately influenced by the temperature indicating that below 100 K tunneling is the dominant electronic transport mechanism in these junctions. TAMR measurements, performed at different temperatures with an in-plane sweeping magnetic field applied along the easy axis (i.e.,[110]) of the fcc-Co thin film, are shown in Figure 3(a)–(d). Rotation of the magnetization away from the easy axis during magnetization reversal produces the TAMR effect.17 The voltage was measured while a constant bias current of 20 µA, which corresponds to a bias voltage of approximately 28 mV, was applied across the junction for all measurements. Even though the device contains only a single ferromagnetic Co layer, a spin-valve-like signal can be clearly observed at temperatures below 20 K due to the TAMR effect. At higher temperature (i.e., ≥ 50 K), the signal disappears. This might be in part attributed to the thermal broadening of the Co/C60 interfacial DOS, which is consistent with the strong bias dependence (the spin-valve-like signal is highest at low bias), indicating that the effect mainly stems from states close to the Fermi8 ACS Paragon Plus Environment

Page 8 of 27

Page 9 of 27

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


energy. By comparison, the temperature-dependent measurements for a pristine epitaxial Co thin film are displayed in Figure S1, supplementary information, showing similar AMR effects. Therefore, it can be concluded that the magnetization reversal process of the Co electrode is nearly temperature independent. Sharp peaks are observed upon reversal of the magnetization and hence rotation of the magnetization vector with respect to the direction of current flow. The measurements show that the magnetization of the fcc-Co can be easily saturated at small (in-plane) magnetic fields, due to its soft ferromagnetic properties. Remarkably, the magnetic field sweeps of Figure 3 show abrupt changes in the resistance at magnetic field values that are much larger than the coercive field of the Co electrode. We propose that interfacial magnetic moments, residing in hybrid Co/C60 states, may give rise to these effects. To unambiguously identify the signature of the hybrid Co/C60 interface, however, it is necessary to compare the magnetoresistance with that of other related systems, such as the tunnel junctions without C60 (i.e., Co(8 nm)/AlOx(3.3 nm)/Al(35 nm)), and the one using AlOx TB to decouple the interaction between Co and C60 at their interfaces (i.e., Co(8 nm)/AlOx(3.3 nm)/C60(2 to 8 nm)/Al(35 nm)).18,19 A comparison of the magnetoresistance of these devices with that of the present junctions is shown in Figure 4. Note that the resistance of the device containing the Co/C60 interface is comparatively low, due to the oxygen deficient AlOx barrier (Figures 4 (e)). Improving the quality of the AlOx barrier by plasma oxidation strongly affects the Co/C60 interface (see Supplementary Information) and is therefore omitted. Figure 4(a) shows a typical anisotropic magnetoresistance (AMR) measurement of an epitaxial fcc-Co layer on a single crystalline sapphire substrate, measured at 5 K and with bias current of 1 µA. The TAMR for sapphire(substrate)/fcc-Co(8 nm)/AlOx(3.3 nm)/Al(35 nm) and sapphire(substrate)/fcc-Co(8 nm)/AlOx(3.3 nm)/C60(2 nm)/Al(35 nm) devices are shown in Figures 4(b) and (c), respectively. Readers are referred to reference 18 and 19 for details. These TAMR measurements reflect magnetic switching 9 ACS Paragon Plus Environment


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

behavior that is consistent with the AMR measurement of the fcc-Co electrode: magnetoresistance features are observed as the Co layer reverses its magnetization. This is in agreement with the notion that the TAMR effect originates from the Co/AlOx interface, and the magnetic switching behavior is, as expected, determined by the magnetic properties of the fcc-Co thin films. Figure 4(d) shows an AMR measurement of the Co/C60 electrode of the device of Figure 3. Compared to the “pristine” fcc-Co of Figure 4(a), the AMR clearly changes due to the adsorbed C60 molecules, indicating that the Co/C60 hybrid interface significantly affects the (bulk) magnetic properties of the Co layer. As it has already been shown in Figure 3, the tunneling magnetoresistance measured through the junction, which probes the interfacial electronic structure, shows more remarkable features (Figure 4(e)). Figure 4(e) is a zoom-in of Figure 3(a), showing TAMR features at small magnetic fields. Evidently, the different features cannot be explained by the switching behavior of the fcc-Co electrode alone. In addition to features at the coercive field of the Co-electrode (indicated by Hc1), several other switching events are detected, at 20 mT (Hc2), 105 mT (Hc3) and 155 mT (Figure 3(d)). A clear metastable resistance state is found between Hc1 and Hc2 of Figure 4(e). This might be due to interfacial magnetic moments that assume metastable configurations. The metastable states were consistently observed with the same Hc1 and Hc2 upon repetition of the magnetization reversal cycle (all measurements were repeated at least three times for each bias). From sample to sample, variations in Hc2 were observed, most probably due to varying interfacial properties over the fairly large junction area in our devices. Similar findings were reported for C14H10O2Zn (ZMP) molecules grown on polycrystalline Co films.22 So far, we have addressed spin-valve-like behavior measured by sweeping the magnetic field through zero, resulting in a complex hysteretic resistance as shown above. We now discuss TAMR measurements of the same junction for different in-plane magnetic field orientations, performed at a sufficiently high magnetic field (800 mT) to achieve saturation 10 ACS Paragon Plus Environment

Page 10 of 27

Page 11 of 27

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


magnetization of the Co electrode in any in-plane direction. Hence, the Co magnetization direction is well-defined in these measurements. Figure 5(a) shows a contour plot of the inplane TAMR as a function of both injected current (I) and angle (θ). The crystallographic indices shown in Figure 5(a) correspond to the in-plane crystallographic axes of the epitaxial fcc-Co thin film at certain values of θ within the 360o in-plane rotation. The [110] direction, i.e. the easy axis of the 8 nm Co thin film, was chosen as the reference axis. The color-scale indicates the sign and the magnitude of the TAMR ratio, which was calculated using Equation 3. The plot manifestly reveals that the junction resistance is strongly influenced by the magnetization direction (θ) and also the bias current (I). The TAMR ratio, of up to about 0.7%, has mostly a positive sign. A clear two-fold symmetry can be observed in these in-plane TAMR measurements. The same symmetry has been observed previously in TAMR measurements of Co(8 nm)/AlOx(3.3 nm)/Al(35 nm) junctions and has been attributed to the interplay between in-plane uniaxial strain in the Co layer and Bychkov-Rashba SOC.18,19 In addition to this similarity, there are also considerable differences in the angle- and bias dependence of the TAMR (see discussion below). This is consistent with the different interfacial electronic structure, characterized by a hybridized Co/C60 interfacial DOS. Figure 5(b) and (c) depict the in-plane TAMR as a function of bias current, measured at some selected angles from 0o to 360o with respect to the reference crystallographic axis [110] of the fcc-Co thin film. Overall, the different curves display a similar trend, in that the TAMR ratios mostly decrease with increasing bias current. This might be primarily attributed to the larger number of energy states that are sampled at higher bias. The largest TAMR ratios appear at low bias (i.e., close to zero bias), while some peaks can also be seen at finite bias currents such as I = -0.3 µA, 0.25 µA and 0.5 µA corresponding to -10.5 mV, 8.7 mV and 17.2 mV respectively. Resonant tunneling involving hybrid electronic states at Co/C60 interfaces may contribute to these phenomena. This is supported by the notion that such 11 ACS Paragon Plus Environment


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

features were not observed for the aforementioned Co/AlOx/Al and Co/AlOx/C60/Al junctions, and also not for Co/C60/AlOx junctions subjected to plasma oxidation (Figure S5, supplementary information). The TAMR ratios of those systems decrease monotonically with increasing bias current. The plots of Figure 5(b) and (c) also exhibit a marked asymmetry for forward (positive) versus reverse (negative) bias currents. Figure 5(d) and (e) show the inplane TAMR as a function of magnetization angles, measured at some selected bias current. The TAMR ratios show a clearly two-fold symmetry at both negative (Figure 5(d)) and positive (Figure 5(e)) bias current. In this case, the positive bias current which corresponds to spin-polarized electrons flow from occupied Co/C60 hybrid interfacial states to unoccupied states of Al produces the relatively large TAMR ratios. Inspired by previous reports of strong tunneling spin polarization at Cr/C60 interfaces measured with spin-polarized scanning tunneling spectroscopy (SP-STS) and large TMR ratios of 60% measured with Co/C60/Co/Ni single molecule MTJs,11,23 we now discuss the results of an alternative experiment, designed to study the effect of the Co/C60 interface on the tunnel spin polarization (e.g. spin filter effects), using the Tedrow-Meservey technique. For this scheme the critical requirement of obtaining a very thin superconducting Al layer (< 10 nm), which required a modified device structure, as is discussed in the experimental section. As a consequence, the Co/C60 interface is inverted (Co is deposited on top of C60). The measurement probes the tunnel spin polarization of electrons mainly originating from the hybrid Co/C60 interfacial states. For the majority and minority spin sub-bands, the tunneling probabilities are different, but the probability for each spin sub-band is approximately constant over the small energy region that is probed (i.e., within about ±1 meV). The 4 nm thickness of the C60 thin film was chosen such that it is sufficiently thin to probe the Co/C60 hybrid interface, but thick enough to firmly rule out contributions from Co in direct contact with the AlOx tunnel barrier. It is expected that the spin polarization is largely conserved during transport through the 4 nm C60 spacer. Most spin-polarized electrons are transferred 12 ACS Paragon Plus Environment

Page 12 of 27

Page 13 of 27

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


either by direct or two-step tunneling (tunneling processes involving a larger number of intermediate steps that are multi-step tunneling will not contribute significantly). For junctions containing two ferromagnetic electrodes, this leads to a suppression of the spin polarization of the current, as we have investigated previously for C60 based organic spin valve structures (i.e., sapphire(substrate)/Co/AlOx/C60/NiFe/Al).5 However, if one of the electrodes is fully spin polarized, as is the case for the thin Al superconductor under application of a strong in-plane magnetic field, two-step tunneling does not a priori lead to a loss of spin polarization. Figure 6 (a) shows a contour plot of the differential conductance (i.e., ⁄ ) as a function of both magnetic field (B) and bias voltage (V) measured at 260 mK. As expected the differential conductance is highly symmetric with respect to bias voltage at zero magnetic field (Figure 6(b)) while increasing magnetic field, it becomes increasingly asymmetric and the Al superconducting gap (∆ = 0.39 meV) is gradually suppressed. The gap finally disappears at relatively large magnetic field (B ≥ 4 T) where the superconductivity of Al is suppressed. Measurements performed at higher temperatures, confirming that the superconductivity is suppressed much more quickly, are shown in Figure S7 of the supplementary information. For positive bias voltages, spin polarized electrons tunnel from the occupied Co/C60 hybrid interfacial states to the unoccupied states of Al. For a certain magnetic field (provided that the superconductivity of the Al still persists), the Zeeman effect splits the quasi-particle states into two energy bands for spin up and spin down electrons, with a Zeeman splitting of 2µBB. Thus, the spin polarized electrons originating from the Co/C60 hybridized interfacial states selectively tunnel into the corresponding spin-split energy states of Al. Figure 6(b)–(f) depicts the differential conductance measured at different magnetic fields at 260 mK. The Zeeman splitting, corresponding to the splitting of the gap-edge features, increases with magnetic field as expected. At 3 T (Figure 6(e)) the Zeeman splitting 13 ACS Paragon Plus Environment


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

is large enough to clearly resolve the two spin channels in the conductance curve, and the tunnel spin polarization of the Co/C60 interface can be extracted. In this plot, the four shoulders labeled as, δ 1 , δ 2 , δ 3 and δ 4 respectively, correspond to the different spin tunneling channels. The net spin polarization of the hybridized energy states result in the asymmetric conductance for spin up versus spin down channels. The tunnel spin polarization can be calculated as: =

− , 2 − + − " = 2 − 1 , 3

where is a factor that is determined by the conductance of the different spin channels, and P is the resultant tunnel spin polarization. We find P = 43%, where the positive sign indicates that the majority spin (i.e., spin up) states dominate the spin transport through the Co/C60 hybrid interface. By comparing with the value of P = 45% found in the literature for Al/AlOx/Co junctions,24 it can be concluded that the Co/C60 hybrid interface does not lead to a significant reduction of the spin polarization. This is somewhat surprising, since the interfacial electronic and magnetic structure is considerably different. We point out that similar experiments have been reported previously by J. S. Moodera’s group for ultra-thin Alq3 (C27H18N3O3Al) and rubrene (C42H28) molecular films, where the tunnel spin polarization was found to be reduced as compared to Al/AlOx/Co.25,26

DISCUSSION The complex hysteretic magnetoresistance observed under application of a sweeping magnetic field as well as the distinctly different (compared to Co/AlOx/Al and Co/AlOx/C60/Al junctions) bias and angle dependent TAMR under application of a magnetic field of constant magnitude indicate that the Co/C60 interface plays an important role. We attribute this to a significant modification of the Co surface electronic structure due to the 14 ACS Paragon Plus Environment

Page 14 of 27

Page 15 of 27

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


adsorption of the C60 molecules. In particular, the resistance plateaus observed in TAMR measurements recorded with sweeping magnetic fields may be ascribed to metastable configurations of Co/C60 interfacial magnetic moments. The magnetic properties of Co/C60 and similar interfaces have been characterized recently with XAS, XMCD, and SP-STM.10,11 XMCD magnetic hysteresis loops measurements for Co/C60 heterostructures at 100 K showed that the coercivity of the Co thin film could be increased dramatically (by about a factor of ten) by interface formation with C60.10 That this is indeed due to the Co/C60 interface has been shown by comparison with a control sample, in which a 2 nm Cu layer was inserted between Co and C60 to eliminate any hybridization or proximity effects across the interface by decoupling the Co 3dz and C60 π orbitals. Significant hybridization effects have been shown for C60/Fe,12 and to a lesser extent for C60/Fe3O4 interfaces.27 For C60/Fe interfaces, joint theoretical and experimental work has shown that a sizeable magnetic moment resides on the C60 molecules, which is antiparallel to the Fe moments.13 SP-STM studies have shown that single C60 molecules adsorbed on a Cr(001) surface also exhibit strong orbital hybridization.11 For such single C60 molecules on Cr(001), bias-dependent SP-STM measurements demonstrated that the sign of the TMR changed from positive to negative as the probed energy was shifted away from the Fermi level, due to spin-splitting of the state derived from the lowest occupied molecular orbital. The TAMR shown in Figure 4(e) is attributed to the interplay between the Co bulk magnetic moments and interfacial moments at the Co/C60 hybrid interface. K. V. Raman and co-authors have demonstrated that strong IMR effects indeed occur in magnetotransport measurements of poly-Co/C14H10O2Zn (ZMP)/Cu structures. The IMR effects could be clearly discriminated from typical AMR signals of the Co layer and show similarities with our measurements of junctions comprising Co/C60 interfaces.22 The IMR was proposed to be due to a coupling of a "spin polarizer" and a "spin analyzer", which are the Co layer and the interfacial layer formed by hybridized Co atoms and ZMP dimers, correspondingly. The ZMP 15 ACS Paragon Plus Environment


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

dimers are a rather special case of molecular adsorbates. The two molecules forming the dimers interact differently with the substrate: the molecule in direct contact with the metallic surface hybridizes strongly, while the other molecule is essentially decoupled and interacts weakly. The particular electronic and magnetic structure resulting from this type of dimer/substrate interaction was proposed to be a key ingredient of the observed IMR, in which the weakly coupled molecule played a major role. Our results indicate that IMR effects may be much more generic for ferromagnetic/molecular interfaces.

CONCLUSIONS In this work, we have investigated spin-dependent electronic transport in tunnel junctions containing Co/C60 hybrid interfaces. The hybridization of electronic states at the Co/C60 interface strongly influences the TAMR. The interplay between Co/C60 hybrid interfacial magnetic moments and Co bulk moments produces distinct TAMR features, i.e. clear plateaus in TAMR measurements attributed to metastable configurations of interfacial moments. Temperature-dependent measurements show that TAMR can only be detected for T < 50 K. The TAMR ratio reaches up to 0.7 % at 5 K under the application of a constant magnet field B = 800 mT and exhibits two-fold symmetry upon in-plane rotation of the magnetization of the Co layer. Bias-dependent TAMR measurements showed peaks that might be attributed to resonant tunneling via hybrid Co/C60 interfacial states. TedrowMeservey measurements showed that the tunnel spin polarization of the Co/C60 interface was 43% at 260 mK and B = 3 T. Our results indicate that interfacial magnetic moments at ferromagnetic/molecular interfaces may be used to generate magnetoresistance in tunnel junctions. Tuning such interfacial magnetoresistance via substrate/adsorbate interactions is a very interesting avenue to explore further.


Page 16 of 27

Page 17 of 27

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


ASSOCIATED CONTENT Supporting Information S1–Temperature dependence of AMR measurements for an 8 nm Co. S2–Bias dependence of in-plane TAMR measurements for sapphire(substrate)/Co(8 nm)/C60(4 nm)/AlOx(3.3 nm)/Al(35

nm).

S3-Temperature

dependence

of

TAMR

measurements

for

sapphire(substrate)/Co/CoO/C60(4 nm)/AlOx(3.3 nm)/Al(35 nm). S4-Bias dependence of TAMR measurements for sapphire(substrate)/Co/CoO/C60(4 nm)/AlOx(3.3 nm)/Al(35 nm). S5-Contour plot of the TAMR ratio as a function of bias current and in-plane magnetization angle for sapphire(substrate)/Co/CoO/C60(4 nm)/AlOx(3.3 nm)/Al(35 nm). S6-Currentvoltage characteristics of the Si(substrate)/SiO2(300 nm)/Al(6.7 nm)/AlOx(3.3 nm)/C60(4 nm)/Co(30 nm), AMR of the Co electrode and the corresponding contour plot of the TAMR. S7-Contour plot and temperature dependence of Tedrow-Meservey measurements for Si(substrate)/SiO2(300 nm)/Al(6.7 nm)/AlOx(3.3 nm)/C60(4 nm)/Co(30 nm). These materials are available free of charge via the Internet at http://pubs.acs.org.

ACKNOWLEDGEMENTS We acknowledge financial support from the European Research Council (ERC Starting Grant No. 280020), and the research program of the Foundation for Fundamental Research on Matter (FOM, Grant No. 10PR2808), which is part of the Netherlands Organization for Scientific Research (NWO).

Notes The authors declare no competing financial interest.



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

REFERENCES (1) Ruden, P. Organic Spintronics Interfaces are Critical. Nat. Mater. 2011, 10, 8-9. (2) Schulz, L.; Nuccio, L.; Willis, M.; Desai, P.; Shakya, P.; Kreouzis, T.; Malik, V. K.; Bernhard, C.; Pratt, F. L.; Morley, N. A.; Suter, A.; Nieuwenhuys, G. J.; Prokscha, T.; Morenzoni, E.; Gillin, W. P.; Drew, A. J. Engineering Spin Propagation across a Hybrid Organic/Inorganic Interface Using a Polar Layer. Nat. Mater. 2011, 10, 39-44. (3) Schulz, L.; Nuccio, L.; Willis, M.; Desai, P.; Shakya, P.; Kreouzis, T.; Malik, V. K.; Bernhard, C.; Pratt, F. L.; Morley, N. A.; Suter, A.; Nieuwenhuys, G. J.; Prokscha, T.; Morenzoni, E.; Gillin, W. P.; Drew, A. J. Engineering Spin Propagation across a Hybrid Organic/Inorganic Interface Using a Polar Layer. Nat. Mater. 2011, 10, 252. (4) Sanvito, S. Molecular Spintronics The Rise of Spinterface Science. Nat. Phys. 2010, 6, 562-564. (5) Tran, T. L. A.; Le, T. Q.; Sanderink, J. G. M.; van der Wiel, W. G.; de Jong, M. P. The Multistep Tunneling Analogue of Conductivity Mismatch in Organic Spin Valves. Adv. Funct. Mater. 2012, 22, 1180-1189. (6) Gobbi, M.; Golmar, F.; Llopis, R.; Casanova, F.; Hueso, L. E. Room-Temperature Spin Transport in C60-Based Spin Valves. Adv. Mater. 2011, 23, 1609-1613. (7) Ma/'Mari, F. A.; Moorsom, T.; Teobaldi, G.; Deacon, W.; Prokscha, T.; Luetkens, H.; Lee, S.; Sterbinsky, G. E.; Arena, D. A.; MacLaren, D. A.; Flokstra, M.; Ali, M.; Wheeler, M. C.; Burnell, G.; Hickey, B. J.; Cespedes, O. Beating the Stoner Criterion Using Molecular Interfaces. Nature 2015, 524, 69-73. (8) Barraud, C.; Bouzehouane, K.; Deranlot, C.; Fusil, S.; Jabbar, H.; Arabski, J.; Rakshit, R.; Kim, D. J.; Kieber, C.; Boukari, S.; Bowen, M.; Beaurepaire, E.; Seneor, P.; Mattana, R.; Petroff, F. Unidirectional Spin-Dependent Molecule-Ferromagnet Hybridized States Anisotropy in Cobalt Phthalocyanine Based Magnetic Tunnel Junctions. Phys. Rev. Lett. 2015, 114, No. 206603. (9) Bairagi, K.; Bellec, A.; Repain, V.; Chacon, C.; Girard, Y.; Garreau, Y.; Lagoute, J.; Rousset, S.; Breitwieser, R.; Hu, Y. C.; Chao, Y. C.; Pai, W. W.; Li, D.; Smogunov, A.; Barreteau, C. Tuning the Magnetic Anisotropy at a Molecule-Metal Interface. Phys. Rev. Lett. 2015, 114, No. 247203. (10) Moorsom, T.; Wheeler, M.; Khan, T. M.; Al Ma'Mari, F.; Kinane, C.; Langridge, S.; Ciudad, D.; Bedoya-Pinto, A.; Hueso, L.; Teobaldi, G.; Lazarov, V. K.; Gilks, D.; Burnell, G.; Hickey, B. J.; Cespedes, O. Spin-Polarized Electron Transfer in Ferromagnet/C60 Interfaces. Phys. Rev. B 2014, 90, No. 125311. (11) Kawahara, S. L.; Lagoute, J.; Repain, V.; Chacon, C.; Girard, Y.; Rousset, S.; Smogunov, A.; Barreteau, C. Large Magnetoresistance through a Single Molecule due to a Spin-Split Hybridized Orbital. Nano. Lett. 2012, 12, 4558-4563. 18 ACS Paragon Plus Environment

Page 18 of 27

Page 19 of 27

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


(12) Tran, T. L. A.; Wong, P. K. J.; de Jong, M. P.; van der Wiel, W. G.; Zhan, Y. Q.; Fahlman, M. Hybridization-Induced Oscillatory Magnetic Polarization of C60 Orbitals at the C60/Fe(001) Interface. Appl. Phys. Lett. 2011, 98, No. 222505. (13) Tran, T. L. A.; Çakır, D.; Wong, P. K. J.; Preobrajenski, A. B.; Brocks, G.; van der Wiel, W. G.; de Jong, M. P. Magnetic Properties of BCC-Fe(001)/C60 Interfaces for Organic Spintronics. ACS Appl. Mater. Interfaces 2013, 5, 837-841. (14) Barraud, C.; Seneor, P.; Mattana, R.; Fusil, S.; Bouzehouane, K.; Deranlot, C.; Graziosi, P.; Hueso, L.; Bergenti, I.; Dediu, V.; Petroff, F.; Fert, A. Unravelling The Role of the Interface for Spin Injection into Organic Semiconductors. Nat. Phys. 2010, 6, 615-620. (15) Matos-Abiague, A.; Fabian, J. Anisotropic Tunneling Magnetoresistance and Tunneling Anisotropic Magnetoresistance: Spin-Orbit Coupling in Magnetic Tunnel Junctions. Phys. Rev. B 2009, 79, No. 155303. (16) Moser, J.; Matos-Abiague, A.; Schuh, D.; Wegscheider, W.; Fabian, J.; Weiss, D. Tunneling Anisotropic Magnetoresistance and Spin-Orbit Coupling in Fe/GaAs/Au Tunnel Junctions. Phys. Rev. Lett. 2007, 99, No. 056601. (17) Ruster, C.; Gould, C.; Jungwirth, T.; Sinova, J.; Schott, G. M.; Giraud, R.; Brunner, K.; Schmidt, G.; Molenkamp, L. W. Very Large Tunneling Anisotropic Magnetoresistance of a (Ga,Mn)As/GaAs/(Ga,Mn)As Stack. Phys. Rev. Lett. 2005, 94, No. 027203. (18) Wang, K.; Tran, T. L. A.; Brinks, P.; Sanderink, J. G. M.; Bolhuis, T.; van der Wiel, W. G.; de Jong, M. P. Tunneling Anisotropic Magnetoresistance in Co/AlOx/Al Tunnel Junctions with FCC Co (111) Electrodes. Phys. Rev. B 2013, 88, No. 054407. (19) Wang, K.; Sanderink, J. G. M.; Bolhuis, T.; van der Wiel, W. G.; de Jong, M. P. Tunneling Anisotropic Magnetoresistance in C60-Based Organic Spintronic Systems. Phys. Rev. B 2014, 89, No. 174419. (20) Tedrow, P. M.; Meservey, R. Spin Polarization of Electrons Tunneling from Films of Fe, Co, Ni, and Gd. Phys. Rev. B 1973, 7, 318-326. (21) QT Designed Instrumentation & Measurement http://qtwork.tudelft.nl/~schouten/index-list.htm (accessed: November, 2015).

Information.

(22) Raman, K. V.; Kamerbeek, A. M.; Mukherjee, A.; Atodiresei, N.; Sen, T. K.; Lazic, P.; Caciuc, V.; Michel, R.; Stalke, D.; Mandal, S. K.; Blugel, S.; Munzenberg, M.; Moodera, J. S. Interface-Engineered Templates for Molecular Spin Memory Devices. Nature 2013, 493, 509513. (23) Fei, X. M.; Wu, G. F.; Lopez, V.; Lu, G.; Gao, H. J.; Gao, L. Spin-Dependent Conductance in Co/C60/Co/Ni Single-Molecule Junctions in the Contact Regime. J Phys Chem C 2015, 119, 11975-11981. 19 ACS Paragon Plus Environment


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

(24) Meservey, R.; Tedrow, P. M. Spin-Polarized Electron Tunneling. Phys. Rep. 1994, 238, 173-243. (25) Santos, T. S.; Lee, J. S.; Migdal, P.; Lekshmi, I. C.; Satpati, B.; Moodera, J. S. RoomTemperature Tunnel Magnetoresistance and Spin-Polarized Tunneling Through an Organic Semiconductor Barrier. Phys. Rev. Lett. 2007, 98, No. 016601. (26) Shim, J. H.; Raman, K. V.; Park, Y. J.; Santos, T. S.; Miao, G. X.; Satpati, B.; Moodera, J. S. Large Spin Diffusion Length in an Amorphous Organic Semiconductor. Phys. Rev. Lett. 2008, 100, No. 226603. (27) Wong, P. K. J.; Zhang, W.; Wang, K.; van der Laan, G.; Xu, Y.; van der Wiel, W. G.; de Jong, M. P. Electronic and Magnetic Structure of C60/Fe3O4(001): a Hybrid Interface for Organic Spintronics. J. Mater. Chem. C 2013, 1, 1197-1202.


Page 20 of 27

Page 21 of 27

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. Schematic diagrams of (a) top view of the spintronic device with structure sapphire(substrate)/Co(8 nm)/C60(4 nm)/AlOx(3.3 nm)/Al(35 nm), with crystallographic directions shown of the single crystalline sapphire substrate and the epitaxial 8 nm Co thin film. M indicates any magnetization direction; (b) three-dimensional (3-D) view of the device structure.



0.006 0.004 0.002

I (mA)

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 22 of 27

0.000 -0.002

5K 100K 150K 200K 300K

-0.004 -0.006 -0.008 -100

-50

0

50

100

V (mV) Figure 2. Temperature dependence of I-V measurements for sapphire(substrate)/Co(8 nm)/C60(4 nm)/AlOx(3.3 nm)/Al(35 nm) junctions.


Page 23 of 27

1.426

1.425

1.424

(d)

50 K

(c)

20 K

(b)

10 K

(a)

5K

1.439

1.438

R (kohm)

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


1.441

1.440 1.442

1.441

1.440 -300

-200

-100

0

100

200

300

B (mT)

Figure 3. Temperature dependence of in-plane hysteretic TAMR measured at (a) 5 K, (b) 10 K, (c) 20 K and (d) 50 K. The red (black) curves correspond to magnetic field sweeps from positive (negative) to negative (positive) values.



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 4. Plots of (a) AMR of a 8 nm pristine Co thin film grown on sapphire substrate (0001); (b) TAMR of a sapphire(substrate)/Co(8 nm)/AlOx(3.3 nm)/Al(35 nm) junction; (c) TAMR of a sapphire(substrate)/Co(8 nm)/AlOx(3.3 nm)/C60(2 nm)/Al(35 nm) junction; (d) AMR of a Co(8 nm)/C60(4 nm) bi-layer stack; and (e) TAMR of a sapphire(substrate)/Co(8 nm)/C60(4 nm)/AlOx(3.3 nm)/Al(35 nm) junction. All measurements were performed at 5 K, and all the junction areas are 250 µm × 300 µm. The red (black) curves correspond to magnetic field sweeps from positive (negative) to negative (positive) values.


Page 24 of 27

Page 25 of 27

[110] [101] [011] [110] [101] [011] [110]

(b) 0.8

-1.5

I (µA)

0.6

-0.5

0.4 0.2

0.0

0.0

0.5

-0.2 -0.4

1.0

TAMR ratio (%)

0.6

0.8

-1.0

(c)

20 deg. 40 deg. 60 deg. 80 deg. 90 deg.

0.4 0.2 0.0 -0.2 -0.4

(d)

0.3

TAMR ratio (%)

0.2

200

0.4 0.2 0.0

-0.4 -1.0

100

100 deg. 120 deg. 140 deg. 160 deg. 180 deg.

-0.2

1.5 0

0.8 0.6

TAMR ratio (%)

(a)

TAMR (%)

0.0

300

1.0

-1.0

I (µA)

0.0

1.0

I (µA)

θ (deg.) -0.1 µA -0.5 µA -1.0 µA -1.5 µA

0.1 0.0

(e)

0.6

TAMR ratio (%)

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


0.4

0.1 µA 0.5 µA 1.0 µA 1.5 µA

0.2 0.0 -0.2

-0.1 0

100

200

300

400

500

0

100

200

300

400

500

θ (deg.)

θ (deg.)

Figure 5. (a) Contour plot of the TAMR ratio as a function of applied bias current and inplane magnetization angle, measured at 5 K under application of a constant magnetic field of 800 mT. The color in the contour plot represents the magnitude of the TAMR ratio in percent (see color bar). (b) and (c) show TAMR versus bias current for several different angles from 0 to 180o. (d) and (e) show the angle dependence of the TAMR effect for negative and positive bias current, respectively.



2.0

3

1.5 2

1.0 0.5

1

2.0

dI / dV (mS)

2.5

(c) 2.5

2.5 2.0 1.5 1.0 0.5

0

0.0 -1.0

-0.5

0.0

0.5

1.0

-1.0

(e)

dI / dV (mS)

dI / dV (mS)

2.0 1.5 1.0

-0.5

0.0

2.0

0.5

1.0

-1.0

0.0

1.6

(f)

δ3

1.2

δ2

0.8

0.0

0.5

1.0

0.0

0.5

1.0

1.52 1.50

B=3T 0.4

-0.5

-0.5

V (mV) δ1

δ4

B=2T -1.0

B=1T

V (mV)

2.5

0.5

1.0

0.0

V (mV)

(d)

1.5

0.5

B=0T

dI / dV (mS)

B (T)

4

(b) 3.0

dI / dV (mS)

(a) 5

dI / dV (mS)

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 26 of 27

1.48 1.46 1.44 1.42 1.40

B = 4.2 T

1.38 1.36

-1.0

-0.5

V (mV)

0.0

0.5

1.0

V (mV)

-1.0

-0.5

0.0

0.5

1.0

V (mV)

Figure 6. Tedrow-Meservey measurements of an organic spintronic device with structure Si(substrate)/SiO2(300 nm)/Al(6.7 nm)/AlOx(3.3 nm)/C60(4 nm)/Co(30 nm). (a) Contour plot of the Tedrow-Meservey measurements, showing the differential conductance (dI/dV) as a function of bias voltage (V) and magnetic field (B), measured at 260 mK. (b)–(f) are selected plots for measurements performed at five different magnetic fields, B = 0 T, 1 T, 2 T, 3 T, and 4.2 T respectively, δ represents differential conductance.


Page 27 of 27

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


TOC graphic


Effect of Orbital Hybridization on Spin-Polarized Tunneling across Co

Recommend Documents