21.12.2012 Views

Ab initio investigations of magnetic properties of ultrathin transition ...

Ab initio investigations of magnetic properties of ultrathin transition ...

Ab initio investigations of magnetic properties of ultrathin transition ...

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

4.2 Results <strong>of</strong> 3d-Monolayers on Rh(001) Substrate 55<br />

In our study, we determined the structural, electronic and <strong>magnetic</strong> <strong>properties</strong> <strong>of</strong> 3d<br />

TM monolayers on Rh(001) by performing the first principles calculations using the full<br />

potential linearized augmented plane wave (FLAPW) method in film geometry as implemented<br />

in the (FLEUR) code (see chapter 2). The generalized-gradient approximation <strong>of</strong><br />

Perdew, Burke and Ernzerh<strong>of</strong> was applied (see section 1.5) leading to a Rh bulk lattice<br />

constant <strong>of</strong> 3.819 ˚A which is only 0.4% larger than the experimental lattice constant <strong>of</strong><br />

3.804 ˚A[120]. The film was modeled by a symmetric seven layer Rh(001) slab covered by<br />

a single 3d monolayer (see figure 2.2), using the calculated in-plane Rh lattice constant.<br />

Relaxations were considered for the topmost two layers, i.e., the 3d ML and the interface<br />

ML Rh(I). Both, the FM and the c(2×2) AFM configuration were relaxed. We used about<br />

120 LAPW basis functions per atom with a muffin-tin radius <strong>of</strong> 1.22 ˚A for the 3d monolayer<br />

atoms and 1.28 ˚A for the Rh atoms. The irreducible part <strong>of</strong> the two-dimensional<br />

Brillouin zone (I2DBZ) was sampled with 78 k� points for the FM (AFM) configuration.<br />

4.2.1 Relaxations and <strong>magnetic</strong> moments:<br />

Here we present relaxations results <strong>of</strong> 1 ML 3d/Rh(001). The formula we used to calculate<br />

the relaxation, Δdxy, between the layer i and j is:<br />

Δdij = dij − do<br />

do<br />

(4.1)<br />

where, dij is the interlayer distance and do is the ideal bulk interlayer distance <strong>of</strong> the<br />

substrate. First we relaxed the top most monolayers <strong>of</strong> the clean Rh(001) surface. The<br />

non<strong>magnetic</strong> relaxations <strong>of</strong> the interlayer spacing between the topmost monolayer and the<br />

second layer, Δd12, and the second Rh interlayer spacing, Δd23, were −0.3% and +0.2%<br />

respectively. Our prediction <strong>of</strong> top most two monolayers’ Rh(001) relaxations are in a very<br />

good agreement compared to the experimental relaxations:Δd12 = +0.5±1.0 (−1.6±1.6)%<br />

and Δd23 =0±1.5% (0±1.6)%[121]([122]).<br />

The relaxations <strong>of</strong> the interlayer spacing between the 3d monolayer and the topmost<br />

substrate layer, Δd12, and the first Rh interlayer spacing, Δd23, are presented in Figure 4.4<br />

for both FM and AFM configurations.<br />

For the FM configuration, we notice that the smallest inward relaxation <strong>of</strong> the 3d<br />

monolayer occurs for Mn and Fe on Rh(001), where the magneto-volume effect (the atomic<br />

volume dependence <strong>of</strong> the <strong>magnetic</strong> susceptibility) is strongest, i.e. the large <strong>magnetic</strong><br />

moments (see Fig. 4.5) <strong>of</strong> these TM compensate the strong inward relaxation – caused by<br />

the larger Rh lattice constant – most efficiently. Of course there are also other factors<br />

controlling the Δd12 that is, e.g., for V smaller than for Cr although the <strong>magnetic</strong> moment<br />

<strong>of</strong> the latter is much larger that <strong>of</strong> vanadium. Here, also the fact that the bulk lattice<br />

constant <strong>of</strong> V is 5% larger than that <strong>of</strong> Cr has to be considered. With the exception<br />

<strong>of</strong> Ni, in all cases the equilibrium distance between the interface Rh layer and the bulk<br />

Rh underneath increases with respect to the substrates bulk interlayer spacing. Moving<br />

from left to right through the periodic table, we find that Δd23 decreases, supporting the<br />

interpretation that the d-band filling controls these relaxations[123, 124]. This highlights

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!