Fabrication and Studies of Magnetic Nanodots
Magnetization stabilization, vortex state, exchange bias, and neutron studies
Igor V. Roshchin, C.-P. Li, Xavier Batlle, Ivan K. Schuller
Physics Department, UCSD
Present collaborators on this project:
- Fabrication (during sabbatical at UCSD)
Xavier Batlle, University of Barcelona, Catalonia, Spain.
- First Order Reversal Curve (FORC) measurements
Randy Dumas, Kai Liu, Physics Department, UC Davis, Davis, CA,
- Neutron Reflectivity
S. Roy, S. K. Sinha, Physics Department, UCSD and Los Alamos National Laboratory,
Los Alamos, NM,
S. Park, R. Pynn, M. R. Fitzsimmons, Los Alamos National Laboratory, Los Alamos, NM,
- Monte Carlo simulations
A. H. Romero, CINVESTAV, Universidad de Queretaro, Qro, Mexico,
J. Mejia Lopez, PUC, Chile,
D. Altbir Drullinsky, USACH, Chile,
Former collaborators on this and related projects:
- Neutron Reflectivity
M. Viret, F. Ott, C.E.A. Saclay-France,
- Ferromagnetic Resonance (FMR)
Chengtao Yu, M. Pechan - Ferromagnetic Resonance (FMR),
Physics Department, Miami University, Oxford, OH,
and many others.
Magnetism at nanoscale, when the size of the
structures is comparable to or smaller than both the ferromagnetic (FM)
and antiferromagnetic (AF) domain size, offers a great potential for new
physics. Moreover, modern technology demands techniques capable of
producing nanometer-sized structures over large areas.
Using self-assembled nanopores in anodized alumina
as a shadow mask [1-3], sub-100 nm magnetic dots covering over 1 cm2
area are deposited by electron-beam evaporation. This method provides
a good control over dot size and separation.
Single layer FM and bilayer FM/AF nanodots or FM
nanodots on a continuous layer of AF as well as continuous FM and FM/AF
films with 15-20 nm-thick FM (FM=Fe, permalloy), and 20-90 nm-thick AF
(AF=FeF2) are simultaneously grown on the same substrate, and
then capped in-situ with a 5-8 nm-thick Ag or Al layer to minimize oxidation
of the FM. Comparative studies of magnetic properties of these samples
in a wide range of temperatures and fields are presented.
Exchange bias (EB) for Fe nanodots due to the native
oxide increases with decreasing dot size, reaching values of 170 Oe at
10 K for 31 nm dots. Improved squareness (MH=0/Ms)
of the magnetization loops for Fe nanodots is caused by EB due to the AF
layer or a native oxide.
Fig. 1 a) Typical magnetization loop for a ferromagnetic. b) Adjacent layer
of an antiferromagnetic exchange biases the ferromagnetic, causing a shift
of the magnetization loop by He.
This condition determines how long the magnetization of a ferromagnetic
structure of volume V will survive thermal fluctuations at temperature
Anodization of Aluminum
Fig.2. Anodization of aluminum using electropolisher setup.
Anodization of aluminum is done using a Buehler electropolisher, controlled
by a computer via a house-built relay box. The same computer is used for
acquisition and real-time plotting of the anodization parameters (voltage,
current, temperature). Variety of anodization conditions are used for various
pore sizes: various acids (oxalic, phosphoric, sulfuric), voltages,
Fig.3 Pore formation in anodic alumina (from Ref.4).
In the electrochemical process of anodization, aluminum turns into aluminum
oxide at pore bottoms, forming columnar pores. Due to the competition of
electrical and mechanical (stress) forces, these pores tend to form hexagonal,
close-packed pattern (see Ref. 3,4).
Nanolithography with porous alumina
Fig.4 Fabrication of nanodot arrays using nanoporous alumina.
Using porous alumina we fabricate arrays of magnetic nanodots that cover
macroscopic area (over 1 cm2)[1,2,5].
Self-supporting masks (membrane)
Fig.5 Pores and dots for membrane (self-supporting) nanoporous alumina
Thin film masks on Si substrate
Fig.6 Pores and Dots for thin film nanoporous alumina mask grown directly
on Si substrate.
Magnetization for nanodots of various sizes
Magnetic properties of the nanodot arrays are studied using SQUID
and MOKE magnetometry.
Fig.7 Magnetization loops for dots of various sizes.
Magnetization: vortex state vs. single domain state
Results of SQUID magnetometry for the Fe dots with diameter
ranging from 30 nm to 100 nm are consistent with a transition from a single
domain to a vortex state. Micromagnetic simulations support the experimental
observations. Both dot surface roughness and dot ellipticity tend
to suppress vortex state. We also investigate effect of interdot interaction
on dot magnetization.
Fig.8 Comparison of magnetization loops for nanodots in a) vortex
state, b) single domain state. Right hand side images show a zoomed-in
virgin curve for each of the samples.
Other measurements and characterizations.
Fig.9 World first polarized SANS measurements on sub-100 nm magnetic dots
a) Raw data at an incident angle of 0.35° on the 2-D detector
b) Cross section through the dashed line and comparison with sample without
dots (log scale) after 5 hours of counting.
In addition to SQUID and MOKE magnetometry we
study magnetic nanodots by a variety of methods with our collaborators.
We were the first to perform neutron reflectivity measurements of sub-100
nm magnetic dots in collaboration with M. Viret and F. Ott (SEA, Saclay,
France). Measurements performed using a Small Angle Neutron Scattering
setup with polarized neutrons (diffractometer “PAPOL” in Saclay) in reflectivity
geometry, allow to determine both structural and magnetic form factors.
The neutron studies are planned to continue using the beam line
in the USA (LANL).
Other current and planned measurements include,
High resolution AFM/MFM, and VSM (FORC).
New results are coming soon (Winter 2006).
1. Chang-Peng Li, Igor V. Roshchin, Xavier Batlle, M. Viret, F. Ott, and
Ivan K. Schuller, "Fabrication and structural characterization of highly
ordered sub-100 nm planar magnetic nanodot arrays over 1 cm2 coverage
area." Journal of Applied Physics, 100(7), October 1 (2006).
2. Kai Liu, J. Nogues, C. Leighton, H. Masuda, K. Nishio, I. V. Roshchin,
and Ivan K. Schuller "Fabrication and Thermal Stability of Arrays of Fe
Nanodots", Applied Physics Letters 81,
3. H. Masuda and K. Fukuda, Science 268, 1466 (1995).
4. O. Jessensky, F. Muller, U. Gosele, Appl. Phys. Lett. 72,
5. MRS Bulletin 28 (7),
Back to the "SCIENCE" site (presently under construction)