Photoemission study of coupled magnetic layers

Z. Q. QiuDept. of Physics, University of California at Berkeley

Outline

• Motivation• Magnetic Phase Transition• Spin Reorientation Transition• Wish List

2D 0D1D

Motivation

GMR and Oscillatory Coupling• P. Grünberg, et. al., Phys. Rev. Lett. 57, 2442 (1986).• M. N. Baibich, et. al., Phys. Rev. Lett. 61, 2472 (1988).• S. S. P. Parkin et. al. Phys. Rev. Lett. 64, 2304 (1990).

5ML 3010 15 2520Spectroscopy of valence electrons

• How does the interlayer coupling change the magnetic properties of the individual layer?

Microscopy of core electrons (PEEM)• Element-specificity + imaging

What is effect of the interlayer coupling on the phase transition?•

Single or separate phase transitions?

•

How is the TC of the coupled sandwich compared to TC1 and TC2?

Magnetic Phase Transition

FM2

FM1FM1 FM2

TC1 TC2

?

Co

Ni

T=310K 380K 430K

Ni(3ML)/Fe(5ML)/Co(10ML)/Cu(100)

Ni

Co

dc

Ni(wedge)/Fe(5ML)/Co(10ML)/Cu(100)

Co(0.23ML)/Cu(2ML)/Ni/Cu(100)

Ni

0.23ML

5.3ML 5.8ML 6.2ML 7.1ML4.9ML

Co

dNi =

dCo= 0.23ML 0.23ML0.23ML 0.23ML

Co(2.3ML)/Cu(2ML)/Ni/Cu(100)

Ni

2.3ML

3.4ML 4.1ML 5.3ML 7.5ML

Co

dNi =2.5ML

dCo= 2.3ML 2.3ML 2.3ML 2.3ML

Co(1.3ML)/Cu(2ML)/Ni/Cu(100)

dCo=

Ni

3.7ML

Co

dNi =

1.3ML

4.5ML 5.0ML 5.4ML 6.1ML

1.3ML 1.3ML 1.3ML 1.3ML

0 1 20

1

2

3

4

5

6

7

IV

III

II

I

Ni thickness(ML)

Co thickness (ML)

Co Ni

0 1 20

1

2

3

4

5

6

7

IVII

IIINi thickness (ML)

Co thickness (ML)

Co Ni

I

Co/Cu(2ML)/Ni/Cu(100) Co/Fe(5ML)/Ni/Cu(100)

I II III IV

Ni PM FM PM FM

Co PM PM FM FM

• PM state of one layer has little effect on the Tc of the other film.

• FM state of one layer reduces the dc of the other layer.

dCo

dNi

I1: PM2: PM

II1: PM2: FM

IV1: FM2: FM

III1: FM2: PM

Jint=0 Jint0

dCo

dNi

I1: PM2: PM

II1: PM2: FM

IV1: FM2: FM

III1: FM2: PM

4.0 5.0 6.0 7.0 8.0 9.01.0

2.0

3.0

4.0 FCAFC

Cu/Co(3ML)/Fe/Ni/Cu(100) Cu/Co(1.6ML)/Fe/Ni/Cu(100) Cu/Co(1ML)/Fe/Ni/Cu(100)

Ni thickness (ML)

Fe thickness (ML)

Spin Reorientation Transition (SRT)

H = -J Si•Sj – K Siz2 What will happen at K 0 ?

K > 0 K < 0K 0

?

• Spin-orbit interaction gives -KSZ2 if z symmetry is broken.

• Dipolar interaction gives 2M2SZ2.

E = -KSZ2 = -(KS/d - 2M2)SZ

2

Changing d or T to tune K around zero.

• Electron Microscope can be used to study stripe domains as a function of “magnetic field”.

M1

M2

If M2/ r 0, E = -J M1•M2 = - M1 •H with H = J <M2>

H = 0

M1

H = J<M2> 0

equals

How do the stripe domains respond to H ?

• Electron microscope is used for domain imaging.

• It is difficult to apply H in an electron microscope.

• SRT of M1 in a coupled sandwich for H=0 is equivalent to the SRT of M1 single layer within a H=J< M2>.

• Sample is grown by MBE.

• Fe/Ni(5ML) film as the SRT layer.

• Cu/Co(10ML) provides the in-plane magnetic field.

Cu(001)

Fe/Ni

H = JMCo

Co

Cu

2.3 2.4 2.5 2.6 2.70

1

2

3

4

5

6

Domain width (

μ)m

( )Fe thickness ML

2.4ML

(c)

2.6ML 2.8ML

50μm //(a)

(b)10μm

Fe thickness (ML)

15μm

Co Fe

15μm

Co(10ML)/Cu/Fe/Ni(5ML)/Cu(001)

The Co magnetization (in-plane field) aligns the stripe domains of the Fe/Ni layer.

Co(10ML)/Cu/Fe/Ni(5ML)/Cu(001)

6.3ML

6.9ML

dFe(ML)

dCu(ML)

FC

AFC

2.84ML2.44ML

50μm

H

K

dFe=2.53ML 2.59ML 2.66ML 2.72ML

dCu=6.57ML 6.66ML 6.73ML 6.81ML

6.47ML 6.38ML 6.27MLdCu=6.57ML

(a)

(b)

(c)

w

w

L

z

y

x

μH=Jint

Ω−≈Ω

+Ω−≡ KJ

KKe 4

2

π

( )1/

1/exp

6

524

242

+Ω

+Ω⋅

Ω=

e

e

KJ

KJJL

π

ππ

Discussion of Jint=0.

2.3 2.4 2.5 2.6 2.70

1

2

3

4

5

6

Domain width (

μ)m

( )Fe thickness ML

2.4ML

(c)

2.6ML 2.8ML

50μm //(a)

(b)10μm

Fe thickness (ML)

⎪⎭

⎪⎬⎫

⎪⎩

⎪⎨⎧

⋅Ω

−=≈

+⎥⎦

⎤⎢⎣

⎡⎟⎠⎞

⎜⎝⎛ +−

Ω

⎪⎭

⎪⎬⎫

⎪⎩

⎪⎨⎧

+⎥⎦

⎤⎢⎣

⎡⎟⎠⎞

⎜⎝⎛ +−

Ω⋅

Ω= int2

4

int

int2

4

int2

4

2

88.0exp)0(

12

14

12

14exp

6

5J

K

JJL

JKJ

JKJ

JL

e

e

e

π

π

π

π

π

π

Discussion

6.2 6.4 6.6 6.8

1.0

2.0 dFe

=2.49ML

2.54ML 2.60ML

Domain width (

μ)m

( )Cu thickness ML0.00 0.01 0.02 0.03

1.0

2.0

Ke-(/4+1/2)Jint

dFe=2.49ML

2.54ML 2.60ML

(Domain width

μ)m

• Keep pushing the spatial resolution down to nm for laterally modulated structures (vicinal surfaces, wires, and dots …).

• Element-specific measurement of antiferromagnetic materials other than oxides (Mn, Cr, FeMn).

• Microscopy within magnetic field.

• Fast spin Dynamics (t<1-10ps).

• Improve in-situ sample preparation capability.

Wish List

1. Develop techniques that are unique for certain type of problems (e.g., element-specificity).

2. Attract users who are doing outstanding research on magnetic nanostructures, but not yet using synchrotrons.

• Department of Physics, University of California at Berkeley Y. Z. Wu, C. Won, and H. W. Zhao

• Advanced Light Source, Lawrence Berkeley National Laboratory A. Doran and A. Scholl

• Department of Physics, Fudan University X. F. Jin

Collaborators