Magnetism and X-Rays:

1
Magnetism and X-Rays Past, Present, and A
Vision of the Future
Joachim Stöhr Stanford Synchrotron Radiation
Laboratory Stanford University
Static image
Femtosecond single shot image
100 picoseconds dynamics
1993
2003
200X
http//www-ssrl.slac.stanford.edu/stohr/index.htm
2
Past
Press release by the Royal Swedish Academy of
Sciences, Nobel Prize in Physics B. N.
Brockhouse and C. G. Shull 1994 Neutrons are
small magnets (that) can be used to study the
relative orientations of the small atomic
magnets. .. the X-ray method has been powerless
and in this field of application neutron
diffraction has since assumed an entirely
dominant position. It is hard to imagine modern
research into magnetism without this aid."
Present
2004 It is hard to imagine modern research into
magnetism without the aid of x-rays!
3
Some Magnetic Devices in Computers
Present Size gt 100 nm, Speed gt 1 nsec Future
Size lt 100 nm, Speed lt 1 nsec
Ultrafast Nanoscale Dynamics
4
(No Transcript)
5
Experimental X-Ray Methods
6
Non-resonant magnetic x-ray scattering is weak
Relative intensity of spin scattering 10 - 4
Relative intensity of charge scattering 1
First experiment F. de Bergevin, M. Brunel
Phys. Lett. A 39, 141 (1972)
7
Development of X-Ray Techniques for Magnetism
Theory J.L. Erskine, E.A. Stern Phys. Rev. B
12, 5016 (1975) M. Blume J. Appl. Phys. 57, 3615
(1985) B.T. Thole, G. van der Laan, G.A.
Sawatzky Phys. Rev. Lett. 55, 2086 (1985)
Experiments X-Ray Magnetic Resonant
Scattering K. Namikawa, M. Ando, T. Nakajima, H.
Kawata J. Phys. Soc. Jpn 54, 4099 (1985) X-Ray
Magnetic Linear Dichroism G. van der Laan, B.T.
Thole, G.A. Sawatzky, J.B. Goedkoop, J.C. Fuggle,
J.M. Esteva, R. Karnatak, J.P. Remeika, H.A.
Dabkowska Phys. Rev. B 34, 6529 (1986) X-Ray
Magnetic Circular Dichroism G. Schütz, W.
Wagner, W. Wilhelm, P. Kienle, R. Zeller, R.
Frahm, G. Materlik Phys. Rev. Lett. 58, 737
(1987) X-Ray Magnetic Imaging J. Stöhr, Y. Wu,
B. D. Hermsmeier, M. G. Samant, G. R. Harp, S.
Koranda, D. Dunham, B. P. Tonner Science 259,
658 (1993)
8
Valence Shell Properties and X-Ray Magnetic
Circular Dichroism (XMCD)
Thole et al., PRL 68, 1943 (1992) Carra, et
al., PRL 70, 694 (1993) Stöhr and König, PRL
75, 3748 (1995)
9
Fe metal L edge
Kortright and Kim, Phys. Rev. B 62, 12216 (2000)
10
Magnetic Spectroscopy and Microscopy
x-ray "spin"
Soft X-Rays are best for magnetism!
11
bulk
surface
12
(No Transcript)
13
PEEM-2 at ALS

• Full Field Imaging
• Electrostatic (30 kV)
• 20 - 50 nm Resolution
• Linear and circular polarization

14
(No Transcript)
15
Element Specific Magnetic Imaging Ferromagnetic
Domains in Magnetite Magnetic Fe and Oxygen
Magnetite Fe3O4

12 mm
16
Spectro-Microscopy of Ferromagnets on
Antiferromagnets
Tune to Co edge use circular polarization
ferromagnetic domains

Tune to Ni edge use linear polarization
antiferromagnetic domains

H. Ohldag et al., PRL 86, 2878 (2001).
17

• Experimental Results
• Exchange bias
• Time resolved imaging of magnetic structures

18
Exchange bias a 50 year puzzle
A ferromagnet has a preference direction when in
contact with an antiferromagnet
The spin-valve sensor
FM 1
FM 2
AFM
Blue layer direction is fixed by exchange
bias Red layer direction determines resistance
Conventional techniques cannot study the magnetic
FM-AFM interface
19
The Basic Model Meiklejohn ( 1960)
Exchange coupling
Bulk FM spins S1
E12 J12 S1 S2
Uncompensated spins S2
E22 J22 S2 S2 anisotropy of AFM EK
Bulk AFM spins S2 S2
Observed loop shift (bias) is 100 times smaller
than expected from model !

• 40 years of theoretical models - reduce bias by
• new effective number of spins S2
• twist of AFM spins domain wall with energy

20
50 years of modelsneed experimental tests
Reduce bias through effective SAFM
SAFM uncompensated spins near AFM
surface Origin ? Number ? Size ? Parallel or
perpendicular ?
Malozemoff model
Koon model
Reduce bias through domain wall
Domain wall energy
Mauri-Siegmann model
Domain wall formation ?
21
Co on NiO(001)
s
s
s
010
2mm
2nm Co on NiO(001)
NiO after deposition
Bare NiO(001)
Co causes Ni spins at NiO surface to rotate into
plane AFM and FM spins couple parallel
22
X-Rays-in / Electrons-out - A way to study
Interfaces
FM Co tune to Co edge circular
polarization AFM NiO tune to Ni edge linear
polarization FM Ni(O) tune to Ni edge
circular polarization
23
Interface Microscopy
Co
Co
Interfacial spins
Nirich NiO
NiO
NiO
FM Ni-rich NiO
AFM NiO
FM Co
Circular pol. Ni edge
Circular pol. Co edge
Linear pol. Ni edge
Chemically induced interfacial Ni spins provide
the magnetic link
24
X-Ray Picture of Exchange Bias
The role of interfacial spins SAFM
Co/IrMn
Co/NiO
Co
NiO
pinned spins
Imaging
Element specific FM loops

• AFM axis is rotated at interface
• The interface is not sharp - SAFM
• SAFM SFM

• Free spins 96 of ML coercivity
• Pinned spins SAFM 4 of ML
• Small number determines bias size

25
Nanaoscale Magnetization Dynamics - Smaller and
Faster
26
(No Transcript)
27
Time resolved x-ray microscopy
PEEM2
50 nm / 100 ps resolution
Laser pump x-ray probe synchronization
lt 1 ps
excitation laser pulse
lt 100 ps
observation x-ray pulse
?t
328 ns
28
Production of Magnetic Field Pulses
Photoconductive switch
H 200 Oe
Conducting wire
50 ? gt I 200 mA, 10 V bias
Magnetic Cells
Current
29
Magnetic Patterns in 20 nm Co90Fe10 films on
waveguide
M
3mm
Field pulse
x-ray "spin"
30
Two pattern with same static structure, but ..
Field response
Field response
Opposite rotation is caused by direction of
vortex core magnetization, i.e. chirality
31
Response to a fast field pulse
Instanteneous precession determined by torque
T H x m
H
slow
"damping"
fast (lt1ns)
T
m
"precession"
H
Tiny vortex core determines fast dynamics of the
whole domain structure!
32

• A Vision of the Future..
• Improved microscopes toward atomic resolution
• X-ray lasers - ultrafast single shot imaging
• ..

33
Tomorrow 5 nm spatial resolution with PEEM3
Lenses
CCD
Deflector
Separator
Lenses
Manipulator
High voltage feedthroughs
CCD -alignment
Electron mirror
34
Spatial Resolution of PEEM3
4-5 nm
35
In 2007 The first x-ray laser - LINAC COHERENT
LIGHT SOURCE (LCLS)
0 Km
2 Km
3 Km
36

• SASE gives 106 intensity gain
• over spontaneous emission
• FELs can produce ultrafast
• pulses (of order 100 fs)

l
37
Growth of X-Ray Brightness and Magnetic Storage
Density
each pulse 1012 photons lt 100 fs coherent
38
Lensless Imaging by Coherent X-Ray Scattering
Eisebitt et al. (BESSY)
Challenge Inversion from reciprocal to real
space image
39

• A Glimpse of the Future..
• Ultrafast magnetic processes

40
(No Transcript)
41
(No Transcript)
42
Experimental Principle of Ultrafast Field Pulses
100 fs 10 ps

• Relativity allows 1010 electrons in short bunch
  of lt 1 ps length
• High field pulses up to 5 T 50,000 Oe

C. H. Back, R. Allenspach, W. Weber, S. S. P.
Parkin, D. Weller, E. L. Garwin, H. C. Siegmann,
Science 285, 864 (1999)
43
Torques on Magnetization by Beam Field
Maximum torque
Minimum Torque
44
The Ultimate Speed of Magnetic Switching
tpulse 3 ps
tpulse 100 fs
90 mm
90 mm
Deterministic switching
Chaotic switching
Under ultrafast excitation the magnetization
fractures !
45
Magnetization fracture under ultrafast field
pulse excitation
Uniform precession
chaotic excitation
46
The magnetism "team" Stanford (SSRL) - Berkeley
(ALS)
Funded by DOE-BES and NSF
Squaw Valley, April 2003
Missing Hans Christoph Siegmann
47
Conclusions

• X-rays have become an important probe of magnetic
  materials and phenomena
• X-rays offer elemental, chemical and magnetic
  specificity with nanoscale spatial resolution
• Transmission experiments probe bulk, electron
  yield experiments probe surfaces and interfaces
• X-rays allow time-dependent studies, paving the
  way for picosecond nanoscale technology
• Future x-ray sources, new techniques and
  instrumentation will allow the complete
  exploration of magnetic phenomena in space and
  time

For more, see http//www-ssrl.slac.stanford.edu/
stohr/index.htm
H. C. Siegmann and J. Stöhr Magnetism From
Fundamentals to Nanoscale Dynamics Springer 2004
(to be published)
48
The end