Full text loading...
SEM images of an array of Ta (3 nm)/Cu (3 nm)/Co (3 nm)/IrMn (7 nm)/Pt (2 nm) dots on a substrate. The images are performed at (a) 45° and (b) vertical incidence with respect to the sample plane. In (b), the direction of the magnetic field during FC is indicated.
Hysteresis loops measured by SQUID at 4 K along the FC direction, for different annealing temperatures , for (a) a continuous film and (b) an array of dots. The loops are subsequent to specific cooling procedure. The samples were initially field cooled under a positive saturating field of 10 kOe from 450 down to 4 K. T was then raised up to and the samples were field cooled under −10 kOe from down to 4 K. (c) and (d) corresponding dependences of the loop shift on . The scales on the right hand side correspond to , where the minimum, and maximum, values for are symmetric in regards to the axis and are indicated by dashed lines on the graphs.
(a) Dependence on of the derivative as deduced from the full lines in Figs. 2(c) and 2(d) , both for a continuous film and an array of dots. vs represent the blocking temperature distributions. The graphs are zoomed on the low T distributions. The normalization is dictated by the experimental results of Figs. 2(c) and 2(d) : the integral of the distribution from 4 to 200 K equals 50% for the continuous film and 75% for the dots. (b) Difference between the distributions for the continuous film and for the dots as plotted from (a). This corresponds to the additional contribution from the edges of the dots. For comparison the distribution for the continuous film, which corresponds to the contribution of the F–AF interface, is also reminded in (b).
Article metrics loading...