AAWSAP DIRD, Metallic Spintronics, March 2010

UNCLASSIFIED/ /FOA OFFl&ilAk W&& 8PtLY
sample to RP. For lighter curves of back-sweeps (Figure 12b), the measured resistance
follows these changes in reverse order.
Focusing on the dark curves, we see that the switching field of the free layer Fl is
essentially independent of the magnitude of I and shows little broadening. In contrast,
the switching field of the pinned layer F2 broadens significantly as the magnitude of I
increases, and the midpoiint of the switching also shifts with I, increasing for - I and
decreasing for + I. Similar behaviors are seen also in the lighter curves. Opposite shifts
for + I and - I indicate these shifts cannot be due to Joule heating, which should cause
shifts in the same direction for both directions of I. But Joule heating might contribute
to the broadening of the switch ing transitions. The shifts of the dark curves are
specified more clearly in Figure 12c, which shows gr,ey -scale plots of the heights of the
curves in Figure 12b, with white representing the antiparallel state of maximum
resistance, and black the parallel state of minimum resistance. Data for three
representative contacts (out of 29) on three different samples show that the behavior
of interest is not limited to a single sample or contact, and that similar results are
generally obtained for straight line fits to 30 percent (white dashed lines), SO percent
(solid white lines), and 70 percent (solid black lines) of the maximum change in R. The
sample in Figure 12d is a contact with R = 1.6 n to an inverted version of the sample in
Figures 12b and c, so that the "directions" of currents are reversed. The sample in
Figure 12e has equal thickness Fl = F2 = 3-nm layers. All three samples show the
same features-that is, electrons passing through F2 into the AFM layer enhance
pinning, and electrons passing through the AFM layer into F2 reduce it. Similar to the
case of AFM=CoFe, a negative current density N10 12 A/m 2 injected through the F=CoFe
into an AFM=IrMn/CoFe interface (Reference 95, 96) was found to increase the
exchange bias, while a positive current decreased it.
Wei et al. (Reference 91) proposed the following qualitative explanation for these
asymmetric changes in switching (exchange bias) field with current. Near the switching
field, the metastability of F2's opposite to field orientation is due almost entirely to
exchange interactions witlh uncompensated moments in the surface layer of the
antiferromagnet. Some of these spins are pinned, thereby inducing an energy barrier
for ferromagnetic layer spin reversal (Reference 88, 89). Electrons flowing from F2 into
AFM induce torques on moments in the AFM matrix, altering its magnetic configuration
(Reference 84). These STT torques tend to favor parallel alignment of moments at the
F2/AFM interface and will therefore tend to increase the exchange bias field. Electrons
flowing in the opposite direction will tend to have the opposite effect. The observed
variations in exchange bias mediated by an electrical cu rrent thus can be taken as good
evidence of the STT effect in AFM. However, such transport measurements do not
distinguish between effects of the current on the bulk AFM and those on interfacial AFM
moments, and more elaborate techniques are needed to obtain a detailed
understanding of the phenomenon.
5. Summary and Conclusions
The semiconductor industry has distinguished itself by a long-term trend known as
Moore's law (Reference 97) that foresees an exponential increase in transistor density
on a chip, doubling approximately every 2 years. Continuing at this pace, the transistor
density will reach ~ 1013 cm·2 by 2035, which at clock speeds of ~10 GHz wou Id result
in ~40 MW/cm 2 of power dissipated on a chip . Unless the energy dissipation from
UNCLASSIFIED//FOA OFFICIOL Wili 0NL¥
16