Spin Currents
|
March 17-19, 2006, at the IBM Research Almaden Research Center |
Download the Spin Currents workshop announcement poster [760KB PDF file]
|
|
Yves Acremann |
Stanford University |
|
|
Stewart Barnes |
University of Miami |
|
|
Gerrit Bauer |
TU Delft |
|
|
Kristen Buchanan |
Argonne National Laboratory |
|
|
Mike Coey |
Trinity College, Dublin |
|
|
Paul Crowell |
University of Minnesota
| |
|
Sergei Demokritov |
Universität Münster |
|
|
William Gallagher |
IBM TJ Watson Research Center |
|
|
Yuichiro Kato |
University of California, Santa Barbara |
|
|
Mathias Klaeui |
Universität Konstanz |
|
|
Sadamichi Maekawa |
Tohoku University |
|
|
Yoshichika Otani |
University of Tokyo |
|
|
Jonathan Sun |
IBM TJ Watson Research Center |
|
|
Gen Tatara |
University of Osaka |
|
|
Stuart Wolf |
University of Virginia |
|
|
Shufeng Zhang |
University of Missouri |
|
|
|
- Stuart Parkin
- Sadamichi Maekawa
- Shoucheng Zhang
- Jim Harris
|
|
|
- Friday: Reception/Dinner and Poster Session
- Saturday / Sunday: Invited Talks and Discussion
|
|
There is no fee for participation in this discussion workshop.
For more details, registration, and to submit a poster (poster size 36"wx40"h)
for the Friday evening reception/poster session, please contact
Brenda Starling at (408) 927-1268 or at brendas@us.ibm.com
|
|
Download the most recent version of the Spin Currents Workshop Agenda and Abstracts [200KB PDF file]
|
|
|
|
5:30 pm
|
Registration and reception (with wine) and posters
(Almaden lobby and outside auditorium
|
|
7:00-8:00
|
Dinner in Almaden Cafeteria
|
|
8:00-8:10
|
Welcome by Stuart Parkin
|
|
|
|
|
8:10-8:55
|
Sadamichi Maekawa, University
of Tohoku
Perspective on spin momentum
transfer
|
|
8:55-9:40
|
Shoucheng Zhang, Stanford
University
Perspective on Spin Hall Effect
|
|
|
|
|
|
|
8:15-8:45
|
Breakfast: Coffee and bagels at Almaden
|
|
8:45-9:00
|
Opening remarks
|
|
|
|
|
|
|
|
9:00- 9:45 am
|
Stuart Barnes, University of Miami
Spin angular momentum transfer, modified
Gilbert relaxation and the Berry phase emf in the dynamics of magnetic
domains
|
|
9:45- 10:30
|
Gen Tatara,
Tokyo Metropolitan University
Current-driven domain wall dynamics based on collective
coordinates
|
|
|
|
|
10:30- 10:45
|
BREAK
|
|
|
|
|
|
|
|
10:45-11:30
|
Shufeng Zhang, University
of Missouri
Spin transfer torques: theory
and simulation
|
|
11:30-12:15
|
Jun’ichi Ieda,
University of Tohoku
Spin Accumulation
and Resistance Due To Domain Walls in Magnetic Nano-Wires
|
|
|
|
|
12:15-1:15pm
|
Lunch at Almaden
|
|
|
|
|
|
|
|
1:15-2:00
|
Mathias Kläui, Universitaet Konstanz
Interactions
between domain walls and spin polarized currents
|
|
2:00- 2:45
|
Yves Acremann, Stanford University
Magnetization Dynamics in a Spin Valve Imaged with X-rays
|
|
|
|
|
2:45-3:15
|
BREAK
|
|
|
|
|
|
|
|
3:15-4:00
|
Jonathan
Sun, IBM T.J. Watson Research Center
Spin-torque-induced magnetic reversal in magnetic tunnel
junctions with MgO barriers
|
|
4:00- 4:45
|
Bill Gallagher, IBM T.J. Watson
Research Center
Spin-Current MRAM in Context
|
|
4:45-5:30
|
Stu Wolf, University of Virginia
Spin Momentum
Transfer-Is it the future of MRAM?
|
|
|
|
|
6:30
|
Dinner/ party at Stuart’s house
|
|
|
|
|
8:15-9:00
|
Breakfast: Coffee and bagels at Almaden
|
|
|
|
|
|
|
|
9:00-9:45
|
Kristen Buchanan, Argonne
National Laboratory
Magnetic vortex
dynamics in confined geometries
|
|
9:45-10:30
|
Sergei Demokritov, University of Muenster
Spin
waves in confined structures: quantization, localization and tunneling
|
|
|
|
|
10:30-11:00
|
BREAK
|
|
|
|
|
|
|
11:00-11:45
|
Paul Crowell, University of Minnesota
Spin Injection and Accumulation in Ferromagnet-Semiconductor
Devices
|
|
11:45-12:30
|
Yuichiro Kato, University of California, Santa Barbara
Imaging the spin Hall effect and current-induced polarization in semiconductors and heterostructures
|
|
|
|
|
12:30-1:30pm
|
Lunch at Almaden
|
|
|
|
|
1:30-2:15
|
Gerrit Bauer, TU Delft
Synchronization of magnetization
motion in multilayers by spin currents
|
|
2:15-3:00
|
Yoshichika Otani, University of Tokyo
Spin-transfer
induced magnetization reversal in lateral devices
|
|
|
|
|
3:00-3:15
|
BREAK
|
|
|
|
|
3:15-4:00
|
Mike Coey, Trinity College, Dublin
Magnetic Detection of Current-Induced Electron Spin-Polarization in Aluminum
|
|
4:00-4:45
|
Ching-Yao Fong, University of California, Davis
Design Half Metals with Simple
Structures
|
|
5:00
|
Closing of meeting
|
|
|
|
|
6pm
|
Dinner for invited speakers
|

|
Perspective on spin momentum transfer
Sadamichi Maekawa
Institute for Materials Research, Tohoku University, Sendai 980-8577 Japan
Spin current in magnetic nanostructures provides a variety of
quantum phenomena and their applications to spin-electronics devices [1,2].
Here, I present the brief review of the basic concepts in spin current and hope
to discuss the perspective on the physics. Special emphasis will be placed on
the spin angular momentum conservation between spin current and various
magnetic structures such as magnetic domain walls and spin accumulation.
[1] Spin Dependent Transport in Magnetic Nanostructures, eds. S. Maekawa
and T. Shinjo (Taylor and Francis, 2002).
[2] Concepts
in Spin Electronics, ed. S. Maekawa (Oxford University Press, 2006)
|

|
Perspective
on Spin Hall Effect
Shoucheng Zhang, Stanford University
I shall review the theoretical and experimental status of the intrinsic spin Hall effect.
A recent theory predicts that dissipationless spin currents can be induced purely by an
electric field in conventional semiconductors. The dissipationless spin current is derived
from a novel topological structure in momentum space, is independent of the sample disorder
and leads to the intrinsic spin Hall effect. In hole doped semiconductors, with or without
inversion symmetry breaking, there are no vertex corrections due to impurities scattering.
I shall analyze a recent experiment on the spin Hall effect in the hole doped system, and
show that it is consistent with the intrinsic nature of the effect. I shall also show that
the spin Hall effect can be quantized in semiconductors with appropriate strain gradients,
but in the absence of any external magnetic fields or the associated time reversal symmetry breaking.
Certain band insulator with a non-vanishing topological invariant can also lead to the
quantum spin Hall effect without time reversal symmetry breaking. Stability of the edge states are discussed as well.
|

|
Spin angular momentum transfer, modified Gilbert
relaxation and the Berry phase emf in the dynamics of magnetic domains
S. E. Barnes
Physics Department, University
of Miami, Coral Gables, FL 33124, USA
The transfer of angular momentum between
a spin polarised current and a magnetic domain has potentially wide practical
applications. The Landau-Lifhshitz-Gilbert (LLG) equations are often used to
describe the dynamics of such domains. In order to be consistent with the
second law of thermodynamics, the Gilbert relaxation term must be modified to
involve a 'particle derivative'. Faraday’s law
e = −df/dt equates the
induced electromotive-force (emf) to the derivative of the flux. A Berry phase g is accumulated by a spin 1/2 particle e.g.,
an electron (or even a neutron), during its adiabatic motion in the effective
magnetic field it sees, e.g., in a ferromagnet. For a closed path this defines
a flux Fs. As was pointed
out by O. Stern in the context of mesoscopic rings, this can result in a
motive-force −dFs/dt.
This arises from the coupling to fields due to the spin and not the charge and should
therefore be called a spinmotive-force (smf). O. Stern, followed by others,
suggested that such an smf arises when the external fields are time dependent,
i.e., when the system is pumped by an external energy source. Here it is shown
that such a smf (and an associated emf) is commonplace in the spin angular
transfer process, directly reflecting its non-conservative nature, even when
the applied fields are stationary. A moving domain wall generates such an smf
which results in current flow even in the absence of any external emf/smf. Due
to the coupling with the environment, it is notoriously difficult to put a
macroscopic system in a state other than that which corresponds to a solution
of the classical problem defined by an appropriate Lagrangian. Here, including
dissipation effects, this classical problem reduces to LLG equations modified
as described above. The implicit 'measurement' made by the environment has
experimental consequences for the angular dependence of the torque transfer
process for both valves and partially or wholly pinned domain walls.
|

|
Current-driven domain wall dynamics based on collective
coordinates
Gen Tatara
Tokyo Metropolitan University
Threshold
current of domain wall motion under spin-polarized electric current in ferromagnets
is theoretically studied. The wall is described by use of equations of motion
in terms of two collective coordinates, wall position and spin polarization
out-of easy-plane. Effects of non-adiabaticity and spin relaxation, both
represented by so-called a beta-term in Landau-Lifshits equation, are taken
account in addition to extrinsic pinning. It is demonstrated that there are
four different regime where threshold current is characterized by different
dependence on extrinsic pinning, shape-anistoropy and beta. The results are
compared with experimental results.
|

|
Spin transfer torques: theory and simulation
Shufeng
Zhang
University of Missouri
We present our
recent theoretical and computational efforts in understanding the spin transfer
torque. Various spin torque models in ferromagnets will be reviewed and
compared. Three current-driven effects will be discussed: 1) thermal effect as
a subject in statistical physics, 2) domain wall dynamics as a challenging
problem in computational micromagnetics, and 3) driven chaos and
synchronization as an emerging topic in non-linear physics.
|

|
Spin Accumulation and Resistance Due To Domain Walls in
Magnetic Nano-Wires
Jun’ichi Ieda
Institute for Materials Research, Tohoku University, 980-8577 Japan;
CREST, Japan Science and Technology Agency, 332-0012, Japan
Recent technological advances in material
fabrication have led to growing interest in spin-polarized transport in ferromagnet
materials [1]. Here we examine the coupling between the spin current and a domain
wall (DW). While conservation of the spin angular momentum leads to the spin
transfer torque driving DW [2], direct observation shows that the spin transfer
is about 10% efficiency [3]. This suggests that the rest of the spins dissipate
elsewhere, which is not yet resolved. On the other hand, transport measurements
on ferromagnetic wires with electric currents traversing 'pinned' DW have
revealed that the DW acts as a source of extra resistance [4]. Motivated by these
experiments, we study the spin accumulation around DW under a finite current.
Due to the asymmetry of the spin-dependent conductivities the spin accumulation
gives rise to local shift of the spin-dependent electrochemical potentials that
amounts to be detected as extra resistance. We solve the diffusion equation for
the spin accumulation around the DW of width w and obtain the excess
resistance. This decreases as a function of the ratio x = w/lF where lF is the spin diffusion length, and in the
abrupt wall limit (x -> 0)
reproduces the well-known result for interfacial resistance [5]. This work has
been done in collaboration with S. Takahashi, M. Ichimura, H. Imamura and S. Maekawa.
[1] Concepts in Spin Electronics, ed. by S. Maekawa (Oxford Univ. Press,
2006).
[2] S. E. Barnes and S. Maekawa,
Phys. Rev. Lett. 95, 107204 (2005).
[3] A.
Yamaguchi et al., Phys. Rev. Lett. 92, 077205 (2004).
[4] U. Ebels et
al., Phys. Rev. Lett. 84, 983 (2000).
[5] T. Valet
and A. Fert, Phys. Rev. B48, 7099 (1993).
|

|
Interactions
between domain walls and spin polarized currents
Mathias Kläui, Universitaet Konstanz
A promising
novel approach for switching magnetic nanostructures is current-induced domain
wall propagation (CIDP) where due to a spin torque effect, electrons transfer angular
momentum to a head-to-head domain wall and thereby push it in the direction of
the electron flow without any externally applied fields. We use
magnetoresistance measurements, spin polarized scanning electron microscopy and
photoemission electron microscopy to directly observe domain wall propagation
in-situ in ferromagnetic nanostructures induced by current pulses [1]. In
addition to wall propagation, we observe a range of wall transformations
including the theoretically predicted nucleation and annihilation of vortices
[2] and find that the wall velocity is directly correlated with the wall spin
structure. Studying different geometries we conclude that the fastest and most
reproducible wall propagation is found in wires, where no transformation
occurs. The observed wall velocities and critical current densities, where
wall motion sets in at room temperature, do not agree well with theoretical 0K
calculations [2]. We have therefore measured the critical current densities as
a function of the sample temperature. We find that the spin torque effect
becomes more efficient at low temperatures, which can be attributed to
thermally activated spin waves and which could account for some of the observed
discrepancies between the 300K experiment and the 0K simulations.
[1] M. Klaui et al., PRL 94, 106601 (2005), PRL 95, 26601 (2005); [2]
A. Thiaville et al., EPL 69, 990 (2005); G. Tatara et al., APL 86, 252509 (2005);
|

|
Magnetization
Dynamics in a Spin Valve Imaged with X-rays
Y. Acremann1, J.P. Strachan2, V. Chembrolu2,
S.D. Andrews3, T. Tyliszczak4, J.A. Katine5,
M.J. Carey5, B.M. Clemens3, H.C. Siegmann1, J.
Stöhr1
1 Stanford Synchrotron Radiation Laboratory, Stanford,
California 94309, USA
2 Department of Applied Physics, Stanford
University, Stanford, California 94305,
USA
3 Department of
Materials Science and Engineering, Stanford University,
Stanford, California 94305, USA
4 Advanced Light Source, Berkeley,
California 94720, USA
5 Hitachi Global Storage
Technologies San Jose Research Center, San Jose, California
95120, USA
Presented by
Yves Acremann
Advanced x-ray imaging techniques can image magnetic
structures much below 1µm at presently 200 ps time resolution. Magnetization
dynamics are excited in a CPP-spin valve by a pulsed spin polarized electron
current. The Oerstedt field of the associated charge current dominates the
dynamics, and magnetization switching takes place in less than 0.5 ns. As
opposed to the Néel or Stoner-Wohlfarth model, the switching does not occur by
uniform rotation of the magnetization, but rather by the motion of a magnetic
vortex across the sample. We discuss the use of spin currents to construct a
spin amplifier with ferromagnetic metals, similar to the one proposed by Dimtri
E. Nikonov and George I. Bourianoff (IEEE Trans. on Nanotechnology 4,
206, 2005) for the still rather elusive magnetic semiconductors.
|

|
Spin-torque-induced
magnetic reversal in magnetic tunnel junctions with MgO barriers
Jonathan Sun, IBM T.J. Watson Research Center
Spin-torque induced
magnetic reversal has been unambiguously demonstrated in magnetic tunnel
junctions with MgO barriers. For quasi-static measurements, the reversal is
dominated by events determined by spin-current amplified thermal activation,
resulting in a measured average switching current below that of the
zero-temperature dynamic threshold. Such sub-threshold switching current generally
shows stronger and non-linear magnetic field dependence, following a shape
determined by the magnetic field dependence of the thermal barrier height.
Time-resolved measurements are required for adequately assessing the dynamic
switching threshold current for fast (nano-second-level) deterministic
switching, and for revealing the magnetic field dependence of the threshold
current. The later would give direct experimental verification of the role a
large easy-plane demagnetization field plays as it determines the value of the
dynamic switching current threshold.
|

|
Spin-Current
MRAM in Context
W.J. Gallagher
IBM Thomas J. Watson Research Center, Yorktown Heights, NY 10598
glgr@us.ibm.com
In recent years there has been an explosion of research on
memory alternatives to the dominant three chip memory families: SRAM, DRAM,
and FLASH. Two factors driving this are (1) scaling limitations faced by each
of these technologies and (2) the desirability for system simplification that
one 'universal' memory could bring, especially in an era of very high
integration approaching a billion transistors per chip. Among the most
promising new-memory candidates for near-term commercialization are
field-switched MRAM, which has some potential as a universal memory, and
Phase-Change Memory, which has potential as a high-density nonvolatile memory
that is scalable beyond the limits of some FLASH technologies. This talk will
look at spin-MRAM in the context of these developments. Field-switched MRAM
is being developed primarily in the 90nm to 180nm nodes, and appears extendable
to perhaps the 45 node. Spin-MRAM appears attractive at and beyond this node.
The reasons for this will be discussed as well the device attributes required
for spin MRAM to be competitive in the overall memory context.
|

|
Spin Momentum
Transfer-Is it the future of MRAM?
Stuart Wolf
University of Virginia
In this brief
talk I will discuss spin momentum transfer as providing the pathway for
continual scaling of MRAM beyond the 32 nm node. This will be based on novel
new materials for the barriers, electrodes and perhaps even the overall
structure of the memory cell. The potential use of oxides, magnetic
semiconductors, multierroics and a hybrid memory cell structure will be some of
things discussed.
|

|
Magnetic vortex dynamics in confined geometries
Kristen S. Buchanan, 1 Pierre E. Roy, 2 Marcos
Grimsditch,1 Frank Y. Fradin, 1 Konstantin Yu. Guslienko, 1 Sam D. Bader, 1
and Valentyn Novosad1
- Materials Science Division and Center
for Nanoscale Materials, Argonne National Laboratory, ,
IL 60439
- Department of Engineering Science, Uppsala
University, Uppsala, Sweden
The magnetic vortex
state is the ground state for a wide variety of magnetically-soft structures
with dimensions on the order of a micrometer or smaller. A magnetic vortex in a ferromagnet with restricted geometry possesses
a characteristic dynamic mode called the translational mode that corresponds to
spiral-like motion of the vortex core around its equilibrium position. Elliptical
nanodots can take on a single vortex or vortex pair magnetization state,
providing a convenient model system for investigating the effects of geometric
confinement and dynamic vortex interactions. Experimentally
we have measured the eigenmodes of lithographically-defined Permalloy
ellipses (2 x 1 mm2 and 3 by 1.5 mm2,
both of 40-nm thick) using a microwave reflection technique. The single vortex magnetization state exhibits a single,
low-frequency excitation (77 MHz for the larger dots, 117 MHz for the smaller
for small applied fields). The frequency of this mode
changes very little for a field applied along the ellipse long axis, however,
it more than doubles in frequency for a field along the short axis, which can
be understood in terms of the shape of the vortex energy profile extracted from
micromagnetic simulations. A much richer excitation
spectrum was measured for vortex pairs confined
in the same elliptical dots [1]. By comparing with micromagnetic
simulations, the detected resonances were assigned to the translational
modes of vortex pairs with parallel or antiparallel core polarizations.
Although the vortex core polarizations play a negligible role in the static
interaction between two vortices, their effect dominates the dynamics.
[1] K. S. Buchanan, P. E. Roy, M. Grimsditch,
F. Y. Fradin, K. Yu. Guslienko, S. D. Bader, and V. Novosad, Nature Physics
1, 172-176 (2005).
|

|
Spin waves in confined structures: quantization,
localization and tunneling
S.O. Demokritov
University
of Muenster, Germany.
The non-local nature of magnetic-dipole interaction and
lateral confinement are responsible for a series of sophisticated physical
effects determining spin wave properties and dynamics of magnetic
nanostructures. I will pay a special attention to the recently observed spin
wave quantization and the spin wave well effect in patterned magnetic films and
consider in detail magnetic dynamics of non-ellipsoidal elements which are mostly
important for applications. Spin wave tunneling via potential barrier created
by an inhomogeneous magnetic field and via vacuum will be also addressed.
|

|
Spin Injection and Accumulation
in Ferromagnet-Semiconductor Devices
Paul Crowell
University of Minnesota
I will discuss recent experiments on ferromagnet-semiconductor
heterostructures.1,2 In the first, we image the accumulation of
spin-polarized electrons near both the source and drain of a lateral Fe/GaAs/Fe
device. Both Fe contacts are epitaxial Schottky tunnel barriers. The spin
accumulation near the source is due to spin injection through the barrier. The
accumulation near the drain is due to the spin-dependent reflectivity of the
barrier under forward bias. For transport through the drain contact, we can
also invert the imaging experiment so that we inject spin-polarized electrons
optically and detect them electrically. All of these experiments are conducted
in a geometry sensitive to spin precession and can be analyzed in detail using
a variation of the drift-diffusion model applied previously to spin transport
experiments in all-metallic structures. The success of this approach leads
one to ask about the possibility of an all-electronic measurement. I will
discuss the status of this effort, focusing on measurements of the spin accumulation
at the drain due to the forward-bias current. This work was supported by the
DARPA SpinS Program, ONR, the NSF MRSEC program under DMR 02-12032, the NSF
NNIN program, and the LANL LDRD program.
In collaboration with X. Lou, M. Furis, C. Adelmann, S.A.
Crooker, and C.J. Palmstrøm
1. S.A. Crooker et al., Science 309,
2191 (2005).
2. X. Lou et al., cond-mat/0602096.
|

|
Imaging the spin Hall effect and current-induced polarization in semiconductors and heterostructures
Yuichiro Kato
Center for Spintronics and Quantum Computation
University of California
Santa Barbara, CA 93106
Spin-orbit coupling in semiconductors relates the spin of an electron to its momentum, and provides a pathway for electrically initializing and manipulating electron spins in zero magnetic field for applications in spintronics and spin-based quantum information processing. This coupling can be regulated with strain in bulk semiconductors and quantum confinement in semiconductor heterostructures.
Using Faraday and Kerr rotation spectroscopies with temporal and spatial resolution, current-induced spin polarization [1] and the spin Hall effect [2] have been observed in bulk semiconductors. More recently, we have investigated the spin Hall effect and current-induced spin polarization in a two-dimensional electron gas confined in (110) AlGaAs quantum wells using Kerr rotation microscopy [3]. In contrast to previous measurements, the spin Hall profile shows complex structure and the current-induced spin polarization is out-of-plane. The experiments map the strong dependence of the current-induced spin polarization to the crystal axis along which the electric field is applied, reflecting the anisotropy of the spin-orbit interaction. These results reveal opportunities for tuning a spin source using quantum confinement, strain and device engineering in non-magnetic materials. This work was supported by ARO, DARPA, NSF and ONR.
1. Y. K. Kato, R. C. Myers, A. C. Gossard, D. D. Awschalom, Phys. Rev. Lett. 93, 176601 (2004).
2. Y. K. Kato, R. C. Myers, A. C. Gossard, D. D. Awschalom, Science 306, 1910 (2004).
3. V. Sih, R. C. Myers, Y. K. Kato, W. H. Lau, A. C. Gossard and D. D. Awschalom, Nature Physics 1, 31 (2005).
|

|
Synchronization of magnetization motion in multilayers by
spin currents
Gerrit
Bauer, TU Delft
The combination
of the spin currents emitted by moving ferromagnets and the spin transfer
torque that excites magnetization motion leads to a cross-talk of magnetic
films in hybrid metallic multilayers. This dynamical coupling through the
normal metals can lead to a synchronized magnetization motion that is favored
by a smaller global damping and observed in ferromagnetic resonance. Similar
phenomena are expected in granular ferromagnets.
*This
work has been carried out in collaboration with Yaroslav Tserkovnyak, Arne Brataas,
Bret Heinrich, and Babak Hosseinkhani.
|

|
Spin-transfer
induced magnetization reversal in lateral devices
Yoshichika Otani and Takashi Kimura
ISSP University of Tokyo
& FRS-The Institute of Physical and Chemical
Research
The spin torque
responsible for the current induced magnetization switching is known to be
proportional to the spin current density. Therefore not the charge current
causing the Joule heat but the spin current is essential to realize efficient
magnetization reversal due to the spin torque. Thus non-local spin injection
technique is employed to realize such a reversal. Here we discuss the
magnetization reversal due to non-local spin injection into a nano-scale
ferromagnetic particle in a lateral ferromagnetic/nonmagnetic hybrid device.
|

|
Magnetic Detection of Current-Induced Electron Spin-Polarization in Aluminum
P. Stamenov and J. M. D. Coey
School of Physics and CRANN, Trinity College, Dublin 2, Ireland
Spin-polarized electric currents have captured the imagination of magnetic theoreticians and
experimentalists for over 30 years. While room temperature applications of spin currents are
limited to length scales of the order of 10 nm, it is expected that spin diffusion lengths become
macroscopically large (of the order of 1 cm) at liquid helium temperatures in sufficiently clean
and light metals, which exhibit little spin-orbit scattering. While spin injection in all-metallic systems
has been detected electrically in various ways, the injection into semiconductors from ferromagnetic metals
is problematic in view of the conductance mismatch problem, and no great injection efficiency has been
achieved so far. Apart from the electrical detection methods, various optical studies have
established the generation and detection of non-equilibrium spin-density in non-magnetic semiconductors.
Here we describe experiments aimed at detecting injected spin density directly.
Taking a strip of aluminium, with Fe, Co, Ni or Zn electrodes, and using spatially-resolved
DC and AC SQUID magnetometry, we have measured the flux created by the non-equilibrium
magnetisation density. Below about 5 K, the induced magnetization at the centre of an aluminium
strip with iron electrodes is ≈ ! kA m-1, for current densities ≈ 108 A m-2.
The diffusion length at 1.8 K is about 3 mm. The spatially-asymmetric magnetization signal due
to the spin injection is superposed on a symmetric signal, linear in field.
The symmetric signal, which is the only effect present for non-magnetic, Zn electrodes,
arises from a reduction of order 10% in the Pauli susceptibility of aluminium due to adiabatic heating of
the electron system by the injected carriers. The observed sign of the induced spin polarization
is negative for iron, and positive for cobalt and nickel.
|

|
Design Half Metals with Simple
Structures*
C. Y. Fong, University of California, Davis
Transition metal
pnictides (group V), a carbide (group IV), their related superlattices with the
zincblende structure, and several carbon quantum wires doped with transition
metal elements have been designed to exhibit half metallic properties.
First-principles VASP code based on density functional theory (DFT) and the
generalized gradient approximation (GGA) has been used. MnAs [1] serves as an
example for the pnictides to explain the features of a half metal, the value of
the magnetic moment/unit-cell and the interactions causing the half metallicity.
MnC [2], a carbide, shows different half metallic properties from the pnictides.
The physical implication of this difference will be presented. We will discuss
half metallic superlattices made of pnictides [3] and the related ballistic
properties [4]. Finally, we will present results on doped carbon quantum wires
exhibiting robust half metallic properties, using the Cr case as a prototype
[5].
[1] J.
E. Pask, L. H. Yang, C. Y. Fong, W. E. Pickett, and S. Dag, Phys. Rev. B 67,
104417 (2003).
[2] M.
C. Qian, C. Y. Fong, and L. H. Yang, Phys. Rev. B 70, 052404 (2004).
[3] C.
Y. Fong, M. C. Qian, J. E. Pask, L. H. Yang, and S. Dag, Appl. Phys. Lett. 84,
239 (2004).
[4] M.
C. Qian, C. Y. Fong, W. E. Pickett, J. E. Pask, L. H. Yang, and S. Dag, Phys.
Rev. B 71, 012414 (2005).
[5] S.
Dag, S. Tongay, T. Yildirim, E. Durgun, R. T. Senger, C. Y. Fong, and S. Ciraci,
Phys. Rev. B 72, 155444 (2005).
*
This work is supported by National Science Foundation.
|
|
|
|
|