Full text loading...
(a) AFM image of a square with an edge length of , irradiated with an ion dose of . The topographic step between square and environment is below experimental resolution. The visibility of the square is mainly due to a minor change in the roughness in the course of the irradiation process. (b) MFM image of the same square. The Landau domain pattern is clearly visible; the distortions are a consequence of the magnetic tip moving the domain walls. This proves the ferromagnetic nature of the interlayer coupling in the irradiated region.
MFM images of a array of irradiated squares with an edge length of . The spacing between two squares is given by (a) , (b) 500 nm, (c) 200 nm, and (d) 100 nm. For all distances down to 200 nm, the squares show independent Landau domain patterns. Only for the smallest distance of 100 nm, a partly collective Landau pattern can be observed. This, however, may also result in the precision limitations of the employed FIB machine during the writing process. Image (c) also shows that some of the squares exhibit a uniform magnetization instead of a Landau pattern. This is probably a consequence of the tip interfering with the magnetization of the squares. The effect is enhanced if the spacing between the elements is very small.
(a) Sample of several irradiated squares with different sizes (1000, 500, 300, 200, 100 nm, from left to right). The MFM image was taken at zero external field. Apparently, the squares with an edge length of 100 nm show a uniform magnetization instead of a Landau domain pattern. The drawing depicts the magnetic configuration of such a square. The black and white areas at the borders are a result of the non-vanishing divergence . Switching the magnetization direction also switches the black and white areas. (b) Situation at an applied field of 25 mT. Note that four of the six 100 nm squares have reversed their magnetization into the direction of the magnetic field. (c) Situation after switching off the external magnetic field. The 100 nm squares have stored the information.
(a) MFM image of a array of squares at remanence with an edge length of 100 nm and an inter-element spacing of 200 nm. The independent character of any square is visible very well. This array corresponds to a bit density of . The dark spot in the lower half of the image is generated by some dirt on the surface. (b) Schematic explanation of the MFM image. The left and right borders appear more contrasted than the rest due to the finite resolution of the tip, which leads to an averaging of the magnetic signal over an area.
OOMMF simulation of a array of 150 nm squares. (a) and (b) show the results with a lateral spacing of 20 nm for the upper and lower layer and a thickness of 1 nm of each layer. The squares show no major interaction, and the antiferromagnetic coupling between the squares is still intact, indicating that there is no significant influence of the exchange coupling. The size of the squares is in a range where a Landau pattern as well as a homogeneous magnetized state can appear at the end of the relaxation process, depending on the random starting orientations of the spins (red color: magnetization pointing to the right; blue color: magnetization pointing to the left).
OOMMF simulation of a array of squares with thicknesses of 65 nm, embedded into a Fe/vacuum multilayer with six 10 nm thick Fe layers and five 1 nm thick vacuum layers representing the Cr. The antiferromagnetic coupling is included by a bilinear and biquadratic coupling term between two neighboring Fe layers. (a) Idealized image of the whole result, (b) a slice of the actual numeric result (grey area marked in (a)). The perpendicular orientation of the magnetization of the elements is clearly visible, while the antiferromagnetic coupling in the area between the elements remains mainly intact (red color: magnetization pointing downwards; blue color: magnetization pointing upwards).
Article metrics loading...