Taking the Hall Effect for a Spin

Junichiro Inoue and Hideo Ohno

Science 309, 2004 (2005);
DOI: 10.1126/science.1113956

This copy is for your personal, non-commercial use only.

If you wish to distribute this article to others, you can order high-quality copies for your
colleagues, clients, or customers by clicking here.
Permission to republish or repurpose articles or portions of articles can be obtained by
following the guidelines here.

The following resources related to this article are available online at

www.sciencemag.org (this information is current as of April 26, 2014 ):

Downloaded from www.sciencemag.org on April 26, 2014

Updated information and services, including high-resolution figures, can be found in the online
version of this article at:
http://www.sciencemag.org/content/309/5743/2004.full.html
This article cites 19 articles, 5 of which can be accessed free:
http://www.sciencemag.org/content/309/5743/2004.full.html#ref-list-1
This article has been cited by 12 article(s) on the ISI Web of Science
This article appears in the following subject collections:
Physics
http://www.sciencemag.org/cgi/collection/physics

Science (print ISSN 0036-8075; online ISSN 1095-9203) is published weekly, except the last week in December, by the
American Association for the Advancement of Science, 1200 New York Avenue NW, Washington, DC 20005. Copyright
2005 by the American Association for the Advancement of Science; all rights reserved. The title Science is a
registered trademark of AAAS.
 PERSPECTIVES
PHYSICS spin polarization (that is, the magnetiza-
tion). This is the extrinsic AHE. Here, the
Hall voltage produced by movement of
Taking the Hall Effect for a Spin charge is accompanied by spin; thus, there
also exists, along with the charge accumu-
Junichiro Inoue and Hideo Ohno lation that produces the Hall voltage, spin
polarization with opposite polarity at the
n 1879, 18 years before the discovery of dependent resistivity ρ σ by means of the two ends. This accumulation of spins shows

I the electron and long before the discov- expression Jσ = (1/ρσ)E (where Jσ is the cur-
ery of its spin, Edwin Hall observed an rent density). The spin dependence of ρσ
effect that now bears his name; he meas- may be caused by spin-dependent elec-
that the spin Hall effect (SHE) exists, but in
this case it is extrinsic because it originates
from spin-dependent scattering. Spin polar-
ured a voltage (the Hall voltage) that arises tronic states or by spin-dependent scatter- ization is usually much more difficult to
from the deflected motion of charged car- ing attributable to imperfections and probe locally with high enough sensitivity.
riers (electrons and/or holes—the absence phonons in crystals. These effects manifest For nonmagnets, although the two charge
of electrons) in solids under an external themselves in the Hall voltage via the spin- Hall currents cancel and no Hall voltage
electric field E and a magnetic field H (1). orbit interaction that couples spin with the develops, spin-dependent scattering still
In the late 20th century, the Hall effect, orbital motion of carriers. Historically, produces the up and down spin currents
which by then had become a routine and these effects were thought to result from an (flow of spins) that flow in the opposite
perhaps unexciting characterization
technique, unearthed the unexpected H M
nature of two-dimensional charged
carriers in semiconductors. It hap-
pened twice, first in the discovery of
the quantized Hall effect and then
when the fractional Hall effect was
found (2). Because electrons have
spin in addition to charge, one may
wonder whether spin plays a role in
the Hall effect, whether a “spin” Hall
effect exists, and then how to observe Ordinary Hall effect Anomalous Hall effect (Pure) spin Hall effect
it and what are the details of its nature with magnetic field H with magnetization M no magnetic field necessary
(3). Recent research has begun to (carrier spin polarization)
Hall voltage but No Hall voltage but
answer all these questions. no spin accumulation Hall voltage and spin accumulation
In ferromagnets, the Hall voltage spin accumulation
consists of two contributions: the ordi-
nary Hall effect (OHE) that leads to Three Hall effects.(Left) The ordinary Hall effect is caused by deflection of carriers moving along an applied elec-
the effects originally discovered as tric field (electrons or holes) by an applied magnetic field. Charge accumulation results in a Hall voltage, but there
mentioned above, and an “anomalous” is no net spin accumulation because there are the same number of spin up carriers as spin down ones. (Middle) The
part that is proportional not to the anomalous Hall effect is the result of spin-dependent deflection of carrier motion, which produces a Hall voltage
external field H but to the magnetiza- and spin accumulation at the edges.(Right) The pure spin Hall effect is caused by spin-dependent deflection of car-
tion of the ferromagnet. This latter riers and produces no Hall voltage when the numbers of deflected spin up and spin down electrons are the same
phenomenon is called the anomalous but gives rise to spin accumulation. For simplicity, only the motion of a few carriers is shown in the figure panels.
Hall effect (AHE) (4). Although the
mechanism of the AHE has been a subject intrinsic effective magnetic f ield in the directions, as long as the spin-orbit interac-
of controversy, it was known to originate momentum space due to the phase called tion is nonvanishing, resulting in spin polar-
from the spin polarization of carriers (that the Berry phase acquired by the moving ization of opposite signs at the edges even
is, the imbalance in the population of carri- electron (7). Two extrinsic mechanisms, in the absence of applied magnetic fields.
ers with different spins). The charge current skew-scattering (8) and side-jump (9), were Thus, SHE may exist with no accompany-
in fer romagnets is dependent on spin then proposed. Most of the experiments ing Hall voltage (12, 13).
(denoted by σ, which can be either “up” or have been analyzed in terms of the extrinsic As in the case of AHE, one can conceive
“down”), and assuming Mott’s two-carrier mechanisms, but the intrinsic AHE was of an intrinsic SHE in nonmagnets on
approximation (5, 6), one can define spin- recently revisited (10) to give quantitative which no external magnetic f ield is
explanations of AHE in ferromagnetic applied. Murakami et al. (14) have pre-
semiconductors (11). dicted for p-type semiconductors that the
J. Inoue is in the Department of Applied Physics, When scattering is spin-dependent, up effective magnetic field originated from
Graduate School of Engineering, Nagoya University,
Chikusa-ku, Nagoya 464-8603, Japan. H. Ohno is at and down spin electrons are scattered into the Berry phase makes up and down spin
the Laboratory for Nanoelectronics and Spintronics, opposite directions, resulting in spin-up and electrons drift toward opposite directions
CREDIT: P. HUEY/SCIENCE

Research Institute of Electrical Communication, spin-down charge Hall currents along the and leads to SHE. The spin-orbit interac-
Tohoku University, Aoba-ku, Sendai 980-8577, Japan, direction perpendicular to E. In ferromag- tion that exists in any material may also
and the ERATO Semiconductor Spintronics Project,
Japan Science and Technology Agency, Kitame-machi
nets, the intrinsic spin imbalance makes the produce the intrinsic SHE even for n-type
1-18, Aoba-ku, Sendai 980-0023, Japan. E-mail: two charge Hall currents asymmetric and semiconductors. Sinova et al. (15) have
ohno@riec.tohoku.ac.jp produces a Hall voltage proportional to the predicted a constant spin Hall conductivity

2004 23 SEPTEMBER 2005 VOL 309 SCIENCE www.sciencemag.org

Published by AAAS

for two-dimensional electron gas with a
Rashba-type spin-orbit interaction pro-
duced by the asymmetry of the potential.
The intrinsic SHE is a result of the inherent
property of the material, as opposed to the
extrinsic SHE caused by scattering.
Elucidating the nature of the pure SHE is
now an emergent issue for experimentalists
as well as theorists. Despite the difficulties
associated with the absence of the Hall volt-
age in the pure SHE, two groups have suc-
ceeded in measuring the spin accumulation
in nonmagnetic semiconductors by opti-
cally detecting the spin accumulation at the
sample edge. Kato et al. (16) spatially
resolved the Kerr rotation of the reflected
light from n-type bulk GaAs and InGaAs
samples and found accumulation of oppo-
site sign at the two edges of the sample.
Subsequently, Wunderlich et al. (17) meas-
ured the polarization of light emitted from a
p-n junction placed at the edge of a struc-
ture. Kato et al. suggested that the observed
effect may be the extrinsic SHE, as the spin
Hall conductivity is low and independent of
the crystal orientation, whereas Wunderlich
et al. concluded that the effect is the intrin-
sic SHE, because the magnitude of the
polarization is consistent with the theoreti-
cal prediction. The interpretation of the
experimental results is complex, because
the current theories predict that the intrinsic
SHE is suppressed by disorder effects for
two-dimensional electron gas with a
Rashba type spin-orbit interaction (18),
whereas it can remain finite, depending on
the type of the spin-orbit interaction (19)
ECOLOGY
Making Sense of Evolution
in an Uncertain World
Vincent A. A. Jansen and Michael P. H. Stumpf
any organisms have adapted to a
M life with uncertainties. For instance,
some pathogenic bacteria have
genes that can be switched off to stop dis-
ease progression in a host organism or pre-
vent their recognition by an immune system.
Such strategies increase an organism’s
reproductive success and tend to be found in
environments in which the conditions are
strongly fluctuating. To understand the
development of such strategies, evolution-
V. A. A. Jansen is at the School of Biological Sciences,
Royal Holloway–University of London, Egham, Surrey
TW20 0EX, UK. M. P. H. Stumpf is at the Centre for
Bioinformatics, Division of Molecular Biosciences,
Imperial College London, London SW7 2AZ, UK.
E-mail: vincent.jansen@rhul.ac.uk
www.sciencemag.org
and the electronic states. The latter predic-
tion may explain the experimental results
within the framework of the intrinsic SHE.
On the other hand, a recent theory on the
extrinsic effect predicts the observed SHE
within experimental error with no adjusta-
ble parameters (20). Nonconservation of
spin in the presence of the spin-orbit inter-
action is also a source of difficulty associ-
ated with theoretical analysis.
Very recently, Sih et al. have imaged the
SHE in a series of two-dimensional electron
gases within (110) AlGaAs quantum wells
having the crystal orientation in which the
Dresselhaus and Rashba spin-orbit interac-
tions are separated (the former out-of-plane,
the latter in-plane) (21). This information
will aid us in establishing the microscopic
relation between the spin Hall current and
the observed quantities. On the theoretical
front, SHE in insulators and its quantized
version has been proposed (22–24); a search
for material systems that allow observation
of such an effect has been initiated.
The SHE has a practical relevance to the
field of spintronics, where spin polariza-
tion, manipulation, and detection are essen-
tial. Theoretical studies to link SHE with
measurable quantities such as spin accumu-
lation and an optical signature are highly
desired, because even if spin Hall current
itself is intrinsic, the stationary spin accu-
mulation is a result of a balance between
spin Hall current and intrinsic/extrinsic
effects of the spin relaxation at the edges of
the sample. Further systematic experiments
that use controllable parameters such as
ary biologists determine the long-run repro-
ductive success of organisms in fluctuating
environments by calculating the Lyapunov
exponent, a measure of the average expo-
nential growth rate in an unpredictable
environment. Often these calculations are
ferociously difficult and rarely lead to sim-
ple results. On page 2075 of this issue,
Kussell and Leibler (1) describe a new
method to approximate the long-term repro-
ductive success in fluctuating environments
and reveal remarkable insights into evolu-
tion in an uncertain world.
Consider again bacteria, which have
developed strategies to cope with a sea of
potential troubles in their rapidly changing
environments, such as the variable natures
SCIENCE VOL 309 23 SEPTEMBER 2005
Published by AAAS
PERSPECTIVES
doping or gate voltage certainly will pro-
vide us a clear view of SHE. A unified pic-
ture of the Hall effect is still being devel-
oped 126 years after its discovery.
References and Notes
1. E. H. Hall, Am. J. Math. 2, 287 (1879).
2. Nobel lectures by K. von Klitzing (1985) and by R. B.
Laughlin, H. L. Störmer, and D C . Tsui (1998)
(http://nobelprize.org/physics).
3. G. E. W. Bauer, Science 306, 1898 (2004).
4. E. H. Hall, Philos. Mag. 19, 301 (1880).
5. N. F. Mott, Proc. R. Soc. London Ser. A 153, 699 (1936).
6. N. F. Mott, Proc. R. Soc. London Ser. A 156, 368 (1936).
7. R. Karplus, J. M. Luttinger, Phys. Rev. 95, 1154 (1954).
8. J. Smit, Physica 24, 39 (1958).
9. L. Berger, Phys. Rev. B 2, 4559 (1970).
10. T. Jungwirth, Q. Niu, A. H. MacDonald, Phys. Rev. Lett.
88, 207208 (2002).
11. H. Ohno, J. Magn. Magn. Mater. 200, 110 (1999).
12. M. I. D’yakonov, V. I. Perel, Zh. Eksp. Teor. Fiz. Pis. Red.
13, 657 (1971).
13. J. Hirsch, Phys. Rev. Lett. 83, 1834 (1999).
14. S. Murakami, N. Nagaosa, S.-C. Zhang, Science 301,
1348 (2003); published online 7 August 2003
(10.1126/science.1087128).
15. J. Sinova et al., Phys. Rev. Lett. 92, 126603 (2004).
16. Y. K. Kato, R. C. Myers, A. C. Gossard, D. D. Awschalom,
Science 306, 1910 (2004); published online 11
November 2004 (10.1126/science.1105514).
17. J. Wunderlich, B. Kaestner, J. Sinova, T. Jungwirth, Phys.
Rev. Lett. 94, 047204 (2005).
18. J. Inoue, G. E. W. Bauer, L. W. Molenkamp, Phys. Rev.
B 70, 041303 (2004).
19. B. A. Bernevig, S.-C. Zhang, Phys. Rev. Lett. 95, 16801
(2005).
20. H.-A. Engel, B. I. Halperin, E. I. Rashba, http://
arxiv.org/abs/cond-mat/0505535 (2005).
21. V. Sih et al., http://arxiv.org/abs/cond-mat/0506704
(2005).
22. S. Murakami, N. Nagaosa, S.-C. Zhang, Phys. Rev. Lett.
93, 156804 (2004).
23. C . L. Kane, E. J. Mele, http://arxiv.org/abs/cond-
mat/0411737 (2004).
24. B. A. Bernevig, S.-C Zhang, http://arxiv.org/abs/cond-
mat/0504147 (2005).
10.1126/science.1113956
of their hosts or the changing number of
available hosts. Many pathogenic bacteria
have evolved phase variation, a process that
turns the expression of certain genes on and
off (2, 3). This “switch” works through
genetic reorganization, mutation, or modi-
f ication of the regions in the bacterial
genome that control gene expression. These
genetic changes are heritable, reversible,
and stochastic. The effect is that a single
bacterium within a population switches
independently of others, and the progeny of
a bacterial population is phenotypically
diverse. This phenotypic diversity serves as
a buffer against fluctuations in the environ-
ment and allows the population to adapt to
unpredictably changing environments.
This strategy of randomization of pheno-
type is known to ecologists as bet-hedging
(4). Bet-hedging does more than just pro-
duce variation that reduces the chances of
population extinction: In a fluctuating envi-
ronment, bet-hedging evolves and bet-
hedgers will in the long run replace equally
diverse populations whose members have
offspring that are all the same. Even if this
2005