11.03.2014 Views

Itinerant Spin Dynamics in Structures of ... - Jacobs University

Itinerant Spin Dynamics in Structures of ... - Jacobs University

Itinerant Spin Dynamics in Structures of ... - Jacobs University

SHOW MORE
SHOW LESS

Create successful ePaper yourself

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

54 Chapter 3: WL/WAL Crossover and <strong>Sp<strong>in</strong></strong> Relaxation <strong>in</strong> Conf<strong>in</strong>ed Systems<br />

Q SO Wc<br />

1.6<br />

1.4<br />

1.2<br />

1.0<br />

0.8<br />

0.6<br />

0.4<br />

0.00 0.02 0.04 0.06 0.08 0.10 0.12<br />

1/ ( )<br />

D e Q 2 SOτ ϕ<br />

Figure 3.13: Width <strong>of</strong> wire WQ SO at which there is a crossover from negative to positive<br />

magnetoconductivity as function <strong>of</strong> the lower cut<strong>of</strong>f 1/D e Q 2 SOτ ϕ .<br />

electron is scattered from such a smooth boundary.[GKD04] If this applies, the potential<br />

is adiabatic and the sp<strong>in</strong> <strong>of</strong> the scattered electron stays parallel to the field B SO as its<br />

momentum is changed. This leads to the boundary condition for the sp<strong>in</strong>-density[SDGR06]<br />

s x | y=±W/2 = 0, (3.85)<br />

s y | y=±W/2 = 0, (3.86)<br />

∂ y s z | y=±W/2 = 0. (3.87)<br />

We can transform this boundary condition to the one <strong>of</strong> the triplet Cooperon by us<strong>in</strong>g the<br />

unitary rotation between the sp<strong>in</strong> density <strong>in</strong> the s i representation and the triplet representation<br />

<strong>of</strong> the Cooperon, ˜s i , Eq.(3.62), which leads to the boundary condition<br />

1<br />

√<br />

2<br />

(−s x +is y )| y=±W/2 = ˜s ⇈ | y=±W/2 = 0, (3.88)<br />

∂ y s z | y=±W/2 = ∂ y˜s ⇉ | y=±W/2 = 0, (3.89)<br />

1<br />

√ (s x +is y )| y=±W/2 = ˜s | y=±W/2 = 0.<br />

2<br />

(3.90)<br />

Now, if werequirevanish<strong>in</strong>gmagnetization for the1Dcase, then thediagonalization is done<br />

straightforwardly, plotted <strong>in</strong> Fig.3.14 (see also Ref. [SDGR06]). We use a basis which satisfies<br />

the boundary conditions and therefore consists <strong>of</strong> ∼s<strong>in</strong>(qy)(1,0,0) T , ∼cos(qy)(0,1,0) T ,<br />

and ∼s<strong>in</strong>(qy)(0,0,1) T , with q = nπ/W, n ∈ N ∗ . However, look<strong>in</strong>g at the sp<strong>in</strong>-diffusion<br />

operator [Eq.(3.57)], we see immediately that if we set k to zero and use the fact that<br />

s x,y must vanish at the boundary and s z has to be constant for the chosen k, we receive

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

Saved successfully!

Ooh no, something went wrong!