Imaging and magnetometry of switching in nanometer-scale iron particles
S. Gider,a) J. Shi, and D. D. Awschalomb)
Department of Physics, University of California, Santa Barbara, California 93106
P. F. Hopkins, K. L. Campman, and A. C. Gossard
Department of Materials, University of California, Santa Barbara, California 93106
A. D. Kent
Department of Physics, New York University, New York, New York 10003
´
S. von Molnar
Center for Materials Research and Technology, Florida State University, Tallahassee, Florida 32306
͑Received 29 August 1996; accepted for publication 18 September 1996͒
The reversal mechanisms in arrays of nanometer-scale ͑Ͻ40 nm diameter͒ iron particles are studied
by low-temperature Hall magnetometry and room-temperature magnetic force microscopy. Rotation
of the net array magnetization at low temperatures ͑20 K͒ occurs by both reversible and irreversible
modes, the latter revealed by Barkhausen jumps. Spatially resolved measurements at room
temperature show the particles to be single domain with remanence and coercivity indicating they
are not superparamagnetic. Individual particles are observed to switch irreversibly over a small field
range ͑Ͻ10 Oe͒ between preferred magnetic directions parallel to the growth direction of the
particles. Scaling of the arrays offers the possibility of magnetic storage at the 45 Gbit/in.2 level,
nearly 50 times greater than current technology. © 1996 American Institute of Physics.
͓S0003-6951͑96͒Q4547-0͔
The control of anisotropy in magnetic systems is essen-
tial for technological applications requiring hysteresis curves
with particular qualities. Shape anisotropy generally domi-
nates the hysteresis of single domain particles, but the rema-
nence and coercivity are reduced with a distribution of par-
ticle shapes and orientations.1 Furthermore, hysteresis loops
may collapse as the magnetic volume decreases and the an-
isotropy energy barrier becomes comparable to the available
thermal energy ͑superparamagnetism͒. Work on defining the
hysteresis of nanometer-scale particles has included surface
chemistry modification2 to exploit the large surface-to-
volume ratio and lithographic patterning of thin films to de-
termine the shape anisotropy.3–7 Shape may also be con-
trolled during growth by electrochemical deposition7–9 and
by scanning tunneling microscopy ͑STM͒ deposition. The
latter techniques allow exploration down to the 10 nm scale,
which is currently inaccessible by lithography. The present
work employs local organometallic deposition with a STM
to produce nanometer-scale iron particles with control of the
shape and orientation. Previously, the magnetic properties of
the STM particles were studied in ensembles at low
temperature.10 By complementary low-temperature Hall
magnetometer and room-temperature magnetic force micro-
scope ͑MFM͒ measurements, the average magnetic proper-
ties of an array of particles are now compared with the prop-
erties of individual particles. The limitations of each
measurement technique are also considered.
High aspect ratio particles are produced by maintaining a
constant emission current as the tip is withdrawn. The
samples are characterized in situ by imaging with a lower
current in tunneling mode or ex situ by transmission electron
͑TEM͒, scanning electron ͑SEM͒, scanning Auger, and
atomic force ͑AFM͒ microscopies.12 Particles with aspect ra-
tios in the SEM ranging from one to three are studied in the
present work. Only upper limits of 40 nm in diameter and
40–120 nm in height can be established with confidence be-
cause of a carbon coating that develops during growth and/or
during observation by electron probes in the presence of hy-
drocarbons. The existence of the carbon coating is confined
by TEM and Auger analysis12 and is consistent with subse-
quent magnetic measurements.10
The magnetic properties of the arrays of iron particles
are studied at low temperatures ͑Ͻ100 K͒ with a Hall
magnetometer10 fabricated from a Be modulation doped
GaAs/Al0.3Ga0.7As heterostructure grown by molecular-
beam epitaxy. At 4.5 K the structure has a carrier concentra-
tion of 3.1ϫ1015 mϪ2, yielding a sensitivity of 0.2 ⍀/G and a
mobility of 17 m2/V s. An array is grown 100 nm above the
2
͑2.5 m͒ square active area of a Hall cross. The sensitivity
of the Hall magnetometer depends on the placement of the
magnets with respect to the active area of the Hall cross. To
a first approximation, the Hall voltage corresponds to the
average over the active area of the perpendicular field, i.e.,
the average magnetic flux. Magnets at the corners of the
active area ͑the lowest points of symmetry͒ couple the least
amount of return flux and therefore give a larger contribution
to the Hall voltage ͑Fig. 1͒. For an array of magnets, the
maximum flux coupling is achieved for complete coverage
of the sensing area. Tilting of the magnetic moments can
lead to negative flux coupling from dipoles at an edge of the
active area and hysteresis loops which are difficult to inter-
pret. Therefore, all measurements are performed with the ex-
ternal field applied perpendicular to the active area of the
Hall magnetometer ͑henceforth known as the vertical direc-
To produce the arrays of iron particles, iron pentacarbo-
nyl ͓Fe͑CO͒ ͔ is introduced into the ultra high-vacuum
5
͑ϳ2ϫ10Ϫ10 Torr͒ chamber of the STM. When the tip is
negatively biased with respect to the substrate, electrons field
emit from the tip to the substrate, dissociate the carbonyl
ligands from the organometallic, and deposit iron locally.11,12
a͒
Present address: IBM Almaden Research Center, D1, 650 Harry Road, San
Jose, CA 95120.
b͒
Electronic mail: awsch@lotemp.physics.ucsb.edu
Appl. Phys. Lett. 69 (21), 18 November 1996 0003-6951/96/69(21)/3269/3/$10.00 © 1996 American Institute of Physics 3269
128.114.34.22 On: Sat, 22 Nov 2014 04:38:48