Session Y20: Invited Session: Spin Accumulation: Experiment and Theory Behind the Controversy

Spin Accumulation and its Detection in Ferromagnet/III-V Semiconductor Devices

As the field of semiconductor spintronics has developed, there has been increasing interest in quantitative measurements of spin accumulation. Successful demonstrations of spin transport have now been reported in various heterostructures combining semiconductors and transition metal ferromagnets. In this talk, I will address the interpretation of spin transport experiments in transition metal ferromagnet/III-V semiconductor heterostructures. These are the most well-developed class of ferromagnet-semiconductor devices, and it is therefore possible to compare different types of measurements made on a single sample. These include the classic non-local spin valve measurement, the local, or ``three-terminal'' measurement, spin Hall and inverse spin Hall experiments, and a new type of microwave measurement suitable for measuring spin accumulations when the spin lifetime is short. The high quality interfaces that can be achieved in epitaxially grown Fe/GaAs as well as Heusler alloy Co$_2$Mn$_{1-x}$Fe$_x$Si/GaAs heterostructures allow for spin accumulations of the order of 50\% to be achieved. I will focus on two essential observations. First, the basic drift-diffusion model on which our understanding of spin transport is based provides an excellent description of the physics in the bulk of the semiconductor. In particular, the non-local spin valve and spin-Hall measurements establish the existence of bulk spin currents, which propagate as expected given the relative weights of drift and diffusion and ordinary mechanisms of spin relaxation. Second, there are significant complications that must be considered when the spin accumulation is large and/or a charge current is present, as is the case for a three-terminal measurement. These types of measurements are a double-edged sword, allowing for new methods of detecting spin accumulation even when a simple quantitative determination of a spin accumulation is not always available. I will highlight how these new approaches have enabled the measurement of spin accumulation with non-magnetic electrodes as well as the successful demonstration of Heusler alloy/III-V semiconductor spin valves operating at room temperature. \\[4pt] This work has been performed in collaboration with Chad Geppert, Kevin Christie, Sahil Patel, Changjiang Liu, Gordon Stecklein, Tim Peterson, and Chris Palmstr{\o}m.   

Spin transport across ferromagnetic tunnel contacts to semiconductors $-$ questions and answers

Ferromagnetic tunnel contacts to semiconductors are key building blocks of spintronic semiconductor devices. Such contacts allow the transport of spins from the ferromagnetic source into the semiconductor, driven either electrically or thermally, but also provide a means to detect spins in the semiconductor and convert the spin information into an electrical signal. Reproducible results have been obtained using either a local (3-terminal) or non-local (4-terminal) measurement geometry, and the electrical spin signals in both cases exhibit all the characteristic features of a current-induced non-equilibrium spin population. Nevertheless, the quantitative analysis has revealed surprising discrepancies with the existing theory. I will address several relevant questions about the nature of the spin transport in ferromagnetic tunnel contacts on semiconductors, and discuss the answers using the experimental observations that have been obtained over the last years and their comparison with available theories, including those involving localized states in the contact. The aim is to clarify what is established and understood and what is not, the latter pointing to interesting new physics yet to be uncovered.\\[4pt] [1] R. Jansen, Silicon spintronics, Nature Materials 11, 400-408 (2012) - review.\\[0pt] [2] R. Jansen, S.P. Dash, S. Sharma and B.C. Min, Silicon spintronics with ferromagnetic tunnel devices, Semicond. Sci. and Technol. 27, 083001 (2012) - review.\\[0pt] [3] S. Sharma, A. Spiesser, S.P. Dash, S. Iba, S. Watanabe, B.J. van Wees, H. Saito, S. Yuasa and R. Jansen, Anomalous scaling of spin accumulation in ferromagnetic tunnel devices with silicon and germanium, Phys. Rev. B 89, 075301 (2014).   

3-terminal Hanle measuements in metals: spin accumulation or novel magnetoresistance effect?

A simple device to study spin injection and transport in semiconductors uses a 3-terminal (3T) geometry, in which spin accumulation is induced and probed by a single magnetic tunnel contact, through the Hanle effect [1]. Since this geometry does not require submicron-sized fabrication, 3T-measurements have become very popular [1-3]. However, many of the reported results disagree with the standard theory of spin injection and have put these measurements into question [4-6]. Our recent works shine some light to this controversy. First, we fabricated ferromagnetic-insulator-nonmagnetic (FIN) 3T devices with metallic electrodes to avoid the complications brought by the Fermi-level pinning when using a semiconductor, and demonstrate that measured Hanle- and inverted Hanle-like features are not compatible with spin injection in these metals [5]. Subsequently, we detect this effect in nonmagnetic-insulator-nonmagnetic (NIN) tunnel junctions for the first time and we demonstrate experimentally beyond any doubt that the measured Hanle-like signals are due to impurities in the oxide layer [7]. We support these results with a theory for impurity-assisted tunneling which takes into account spin interactions and Coulomb correlations. We show that this is actually a novel magnetoresistance effect, which is general to any impurity-assisted tunneling process regardless of the oxide thickness or materials used. The presented work will thus be used as a benchmark to spin injection experiments to any nonmagnetic material, and specially will redirect research of semiconductor spintronics, with all the implications in such a technologically relevant area. [1] S. P. Dash et al., Nature 462, 491 (2009); [2] C. H. Li et al., Nature Commun. 2, 245 (2011); [3] A. Jain et al., Phys. Rev. Lett. 109, 106603 (2012); [4] Y. Aoki et al., Phys. Rev. B 86, 081201(R) (2012); [5] O. Txoperena et al., Appl. Phys. Lett. 102, 192406 (2013); [6] H. N. Tinkey al., Appl. Phys. Lett. 104, 232410 (2014); [7] O. Txoperena et al., Phys. Rev. Lett. 113, 146601 (2014).   

New interpretation for recent spin injection experiments

We elucidate the large discrepancy between known spin relaxation theory and the findings of recent spin injection experiments that make use of a single ferromagnetic-insulator-nonmagnetic junction for both injection and detection of spin-polarized currents. This local setup scheme gained popularity since 2009 when Dash et al. claimed to achieve room temperature spin injection in silicon [1], followed by avalanche of similar experiments in silicon and other materials that resort to this measurement technique. We show that those enhanced signals and their dependence on temperature are set by impurities embedded in the tunnel barrier with large on-site Coulomb repulsion compared with the voltage bias [2]. Depending on the electron occupation of the resonance level, the magnetoresistance effect is established by the interplay between the Zeeman energy and the impurity coupling to the ferromagnetic material. Considering molecular fields due to hyperfine and exchange interactions, we capture the shape and sign dependence of the signal on magnetic field orientation. The findings are used to explain both conventional spin injection [1], and cases where the bias voltage is distributed across the junction while the net charge current is zero (the so-called local Seebeck spin tunneling [3]). Finally, we extend the theory to impurity-rich tunnel junctions, showing that a similar magnetoresistance effect can persist in completely nonmagnetic junctions [4]. The extension beyond electrical spin injection from ferromagnetic electrodes paves the way for a new class of 1D nanometer-size memory cells which represents the ultimate scaling of memories (leaving no room in the bottom). \\[4pt] [1] S. P. Dash et al., Nature 462, 491 (2009).\\[0pt] [2] Y. Song and H. Dery, Phys. Rev. Lett. 113, 047205 (2014).\\[0pt] [3] J.-C. Le Breton et al., Nature 475, 82 (2011).\\[0pt] [4] O. Txoperena et al., Phys. Rev. Lett. 113, 146601 (2014).   

Crossover from Spin Accumulation into Interface States to Spin Injection in the Germanium Conduction Band

Electrical spin injection from ferromagnetic metals to silicon (Si) and germanium (Ge) is the first and basic requirement for the development of spintronic devices and their integration with mainstream semiconductor (SC) technology. The main obstacle to efficient spin injection is the conductivity mismatch between the ferromagnetic metal and Si, Ge and requires tunneling spin injection through an oxide barrier (Ox). However, tunneling spin injection raises other important issues in the interpretation of spin signals obtained in three-terminal geometry. In particular, the possible presence of localized states within the Ox or at the Ox/SC interface may lead to wrong conclusions. To study the exact origin of the spin signals measured in three-terminal geometry, we have grown Ta/CoFeB/MgO/SOI and GeOI samples with n and p type doping using variable MgO thicknesses. The use of SOI and GeOI substrates allows us to apply back gate voltages to the SC channel to vary its resistivity. Moreover we have used three different techniques to grow the MgO tunnel barrier: by sputtering of MgO or Mg followed by a plasma oxidation and e-beam evaporation of MgO. Using the Mg and plasma oxidation growth of the tunnel barrier, though less flexible than the other techniques, allowed us to show the temperature transition from the spin accumulation into interface states to the spin accumulation into the conduction band of n-Ge. Above 150 K, the magnitude of the spin RA product agrees well with the spin diffusion theory predictions and is proportional to the injected current and to the channel resistivity as expected. Temperature dependent spin pumping measurements showed the same transition. Using the same spin injector, we also found radically different spin signals using p-Ge supporting the fact that spin accumulation occurs into the SC channel. In this presentation, we will extend our spin signal analysis using MgO tunnel barriers of different thicknesses and grown by different methods.