Chemistry Reference
In-Depth Information
fundamental interactions, observations at the atomic scale are needed. This can be
accomplished using direct STM imaging of surfaces of a current-carrying conductor,
with extreme care to address the issues of temperature and drift that arise with a large
current density.
Single-crystal thin-film Ag samples with (111) surface orientation were prepared
by thermal evaporation of 100-350 nm of Ag onto a freshly cleaved mica substrate
[
40
-
42
]. The samples were then transferred into a treatment chamber. After sev-
eral Ar
+
sputtering and annealing cycles, atomically clean Ag(111) surfaces were
obtained. To modify the surface morphology for experiments on Ag island motion,
gentle sputtering without annealing was used. The types of surface structures that
can be controllably achieved are illustrated in Fig.
5.2
. After these procedures, the
samples were loaded onto an STM stage designed to heat-sink the mica substrates.
Following application of direct current, 30-40min is needed for thermal stabiliza-
tion. To measure the structural evolution or motion of defects, time-lapse imaging is
used. To assess step displacement, repeated one-dimensional scans are made across
the step at a scan rate of
∼
9 pixels/ms. To assess island motion, the fastest scan
×
×
speed is 60 s per 256
256 pixel image and 120 s per 512
512 pixel image.
∼
∼
.
The typical tunneling parameters are
6 V. Under these
tunneling conditions the tip-sample interactions can be negligible for Ag surfaces
[
40
,
43
].
Two methods were used to determine the temperature of the sample under applied
current. The temperature was measured using an alumel-chromel thermocouple,
which was spot-welded to a tiny Ta tab (strip) and brought into direct contact with
the film surface in a calibration performed after the electromigration measurements
[
40
]. In addition, the sample temperature was calibrated via a thermocouple in a
separate chamber (base pressure
50 pA at sample bias
1
10
−
8
torr) using the same heater system as
in the UHV system. The thermocouple wire diameter was 0.001
, and a simple
analysis shows that the heat-sinking effect of the thermocouple results in less than
1 K temperature change.
∼
4
×
800 nm
2
STM
Fig. 5.2
STM images of clean Ag(111) films grown on mica in UHV.
Left
: 800
×
image of structure achieved by annealing during sputtering.
Right
: 500 nm
×
500 nm STM image
of structure achieved by sputtering at room temperature [
79
]
