APS March Meeting 2010
Volume 55, Number 2
Monday–Friday, March 15–19, 2010;
Portland, Oregon
Session D8: Spin Transport in Carbon-based Materials
2:30 PM–5:30 PM,
Monday, March 15, 2010
Room: Portland Ballroom 255
Sponsoring
Unit:
GMAG
Chair: Jing Shi, University of California, Riverside
Abstract ID: BAPS.2010.MAR.D8.1
Abstract: D8.00001 : Recipes for lateral spin transport between magnetic contacts, advantage of carbon-based materials.
2:30 PM–3:06 PM
Preview Abstract
Abstract
Author:
Albert Fert
(CNRS/Thales Laboratory)
After the presentation of magneto-transport results [1] on
metallic carbon
nanotubes (CNT) between LSMO electrodes (MR $\approx $ 60-70{\%},
[V$_{AP}$
-- V$_{P}$] $\approx $ 60mV), I will discuss the general problem
of spin
transport in a nonmagnetic lateral channel between spin-polarized
contacts
in both the diffusive and ballistic regimes.
In the diffusive regime, a treatment by the classical
drift-diffusion
equations applied to a multi-terminal structure is used to
calculate what
can be expected for the output signal with local or non-local
voltage
probes. A general result is that the output signal ($\Delta $R =
$\Delta
$V/I where $\Delta $V is the local or non-local output voltage),
directly
related to the spin accumulation splitting in the channel, scales
with the
smallest of the relevant spin and interfaces resistances. In the
best
situation, that is with only tunnel contacts having the same (large)
resistance R$_{T}$ and separated by less than the spin diffusion
length
($\lambda )$ in a lateral channel limited to the zone of the
contacts, the
signal $\Delta $R increases in proportion of R$_{T}$ as long as
the dwell
time is smaller than the spin lifetime. $\Delta $R can be thus
much larger
than the spin resistance of the channel (product of its
resistivity by the
ratio $\lambda $/section ). This explain why, in the experiments
of Ref.[1]
on CNT, $\Delta $R can be as large as 90 M$\Omega $, that is of
the order of
the tunnel contact resistances and much larger than the spin
resistance of
the CNT (smaller signals in experiments with CNT or graphene are
often due
the leak of spin accumulation in lateral channels extending too
far outside
the contacts). The relative disadvantage for semiconductors comes
from the
too long dwell time due to much smaller electron velocities than
in metallic CNTs (and graphene).
We will conclude by a similar analysis of the ballistic regime
and a discussion of experiments with graphene.
\\[4pt]
[1] Hueso et al, Nature 445, 410 (2007).
To cite this abstract, use the following reference: http://meetings.aps.org/link/BAPS.2010.MAR.D8.1