For self-assembled nanodots, the ultimate dream is to simultaneously achieve tunable uniformity in size, spatial distribution, chemical composition, and crystallographic orientation. By utilizing the Volmer–Weber growth mode in thin film epitaxy, we have grown self-assembled two-dimensional arrays of FePt alloy nanodots that are uniform in size, chemical composition, and are all crystallgraphically aligned. These dot assemblies are ferromagnetic at room temperature and can be easily transferred onto other templates without destroying the size and orientation uniformity.

1.
2.
P.
Moriarty
,
Rep. Prog. Phys.
64
,
297
(
2001
).
3.
V. A.
Shchukin
and
D.
Bimberg
,
Rev. Mod. Phys.
71
,
1125
(
1999
), and references therein.
4.
S.
Sun
,
C. B.
Murray
,
D.
Weller
,
L.
Folks
, and
A.
Moser
,
Science
287
,
1989
(
2000
).
5.

The iron and platinum were evaporated from Fe and Pt wires (5N purity) heated by electron beam bombardment. The NaCl single crystal substrates were cleaved in air, then were immediately loaded into the UHV chamber and were annealed to 530 K for 1 h to remove surface contamination prior to the experiments.

6.

TEM was carried out using a Hitachi HF2000 field emission TEM at 200 kV. The nanodots assemblies along with the thin carbon capping layer were detached from the substrate by immersing the samples into distilled water and suspended on 400-mesh TEM copper grid right before the TEM measurements. A plasma cleaning of 90 s was applied to remove the hydrocarbon and thin the carbon film.

7.
Z.
Gai
,
B.
Wu
,
J. P.
Pierce
,
G. A.
Farnan
,
D. J.
Shu
,
M.
Wang
,
Z.
Zhang
, and
J.
Shen
,
Phys. Rev. Lett.
89
,
235502
(
2002
).
8.

Micro-electron diffraction was performed using a 50 m aperture setting. The diffraction pattern was calibrated using a silicon 110 pattern.

9.

The 1 keV peak in the experimental spectra is the Cu L line. In order to get proper background around the Fe peak, a high concentration of copper is modeled. The measured Cu peaks are fluoresced from the TEM copper grid bars, therefore, the measured K to L ratios are ° off, which is quite different from the simulated value.

10.
K.
Reichelt
,
J. Cryst. Growth
35
,
55
(
1976
).
You do not currently have access to this content.