WIR SCHAFFEN WISSEN – HEUTE FÜR MORGEN

:: Jonathan White
:: Laboratory for Neutron Scattering and Imaging (LNS)
:: Paul Scherrer Institute (PSI), Switzerland

Skyrmions in magnetic materials

Download slides from: https://www.psi.ch/lns/jonathan-white

19th March, 2017

I. What is a Skyrmion?

Page 2
 What is a Skyrmion?

•A Skyrmion is a local, stable, soliton-like solution of a nonlinear field theory

•Interpreted as particle-like objects (baryons in a spinless pion medium)

•These particles persist due to topological-protection, which is characterised by a

topological integer (winding number).

•Skyrmion excitation in CM: Liquid crystals, QHE, BEC, non-centrosymmetric magnets

In the original sense

µ 0H
In real magnets

http://www.christophschuette.com/physics/skyrmions.php P. Milde et al., Science 340, 1076 (2013) Page 3

 Classifying a magnetic skyrmion
Homotopy theory: allows a classification of magnetic structures according to topology.
Examine the mapping of magnetic spin vectors in a physical space Sn to its order parameter
space Sm.
The order is considered ‘topological’ when the mapping cannot be shrunk to a single point:
πn(Sm) > 0
A topological structure can be characterised according to a winding integer N, which counts the
number of times the physical space fully covers the order parameter space.

Example: Spins with fixed length on a circle (1-sphere). Homotopy group: π1(S1) =
e.g. magnetic vortex
Order parameter is a 2D magnetization vector: m
Order parameter space = circle (one angle) = S1

N=0 N = -1

N = +1 N = +2
Page 4
 Classifying a magnetic skyrmion
Magnetic Skyrmions belong the homotopy group: π2(S2) =
Spins with fixed length on a 2-dimensional plane
Order parameter is a 3D magnetization vector: m
Order parameter space = surface of a sphere (two angles) =S2
Real Space Depiction of mapping to order parameter space

Characterised by winding number - |N| = 1.

• Closed particle-like state (closed surface in order parameter space with physical stability)
• Topologically non-trivial (Skyrmions are countable objects)
• Topological protection (a continuous change of homotopy is forbidden introduces an
energy barrier between different topological states in a real physical system)
Page 5
II. Where are they found?

Page 6
 Theory for ‘vortices’ in magnets

A. Bogdanov and A. Hubert, JMMM 138, 255 (1994). “spin vortices” as solutions of a
A.N. Bogdanov and D.A. Yablonsky, Sov. Phys. JETP 95, 178 (1989).
A.N. Bogdanov, U.K. Rößler and C. Pfleiderer, Physica B 359, 1162 (2006).
continuum model condensed
U.K. Rößler, A.A. Leonov and A.N. Bogdanov, J. Phys: Conf. Series (2010) hcp skyrmion phases

Zeeman DMI energy

Exchange Cnv with n=2,3,4,6
Anisotropy q
cycloidal groundstate
• DMI couplings included as Lifshitz-type invariants DM
• They generally exist in non-centrosymmetric magnets
• The pattern of allowed DMIs determines the Skyrmion type

Néel-type skyrmion

Page 7
 Theory for ‘vortices’ in magnets

A. Bogdanov and A. Hubert, JMMM 138, 255 (1994). “spin vortices” as solutions of a
A.N. Bogdanov and D.A. Yablonsky, Sov. Phys. JETP 95, 178 (1989).
A.N. Bogdanov, U.K. Rößler and C. Pfleiderer, Physica B 359, 1162 (2006).
continuum model condensed
U.K. Rößler, A.A. Leonov and A.N. Bogdanov, J. Phys: Conf. Series (2010) hcp skyrmion phases

Zeeman DMI energy

Exchange T, O crystal classes
Anisotropy
helical groundstate DM || q
• DMI couplings included as Lifshitz-type invariants
• They generally exist in non-centrosymmetric magnets
• The pattern of allowed DMIs determines the Skyrmion type

Bloch-type chiral skyrmion

Page 8
 Chiral magnet MnSi
Simple Hamiltonian with a monochiral helix groundstate
0.46 nm

Mn Ratio of J/D determines helical pitch = 18 nm.

Crystal class T H-O anisotropy terms align helices with <111>.

Cubic (P213)
No inversion DMI Phase diagram

µ0 H

S. Mühlbauer et al., Science 323, 915 (2009) Page 9

 Observing chiral magnetic order by SANS
Incommensurate chiral magnetic order Bragg satellites around commensurate positions

In real space In reciprocal space SANS pattern

S. Mühlbauer, PhD Thesis 2009

Large period magnetic structures (~30 to 5000 Å) low q SANS

Velocity Selector
Collimator Sample

• Length of instrument: 10-40 m

• Scattering angle: ~1º (length-scales
Magnet Coils from ~30 to 5000 Å)
Monitor
Area Multidetector • Non-destructive bulk probe
Neutron Source
10 - 40 m Page 10
 Observing the Skyrmion lattice by SANS
Skyrmion Lattice
Real Space

Reciprocal space

S. Mühlbauer et al., Science 323, 915 (2009)

Skyrmion lattice in MnSi (and others)

- Hexagonal SANS scattering pattern
- Propagation vectors perpendicular to B
- Weak pinning (alignment) with crystal directions
- First order phase boundary with conical + PM phases

SANS does not directly ‘see’ the particle-like skyrmion property

Page 11
 Real-space imaging of the Skyrmion particle
SANS - bulk MnSi Spin-polarised TEM – thin samples MnSi

A. Tonomura et al., Nano Lett., 12, 1673 (2012)

Lorentz force TEM – thin samples Magnetic tip AFM (MFM) – surfaces
Fe0.5Co0.5Si and FeGe Fe0.5Co0.5Si GaV4S8

Z.X. Yu et al., Nature 465, 901 (2010), Nat. Mater. 10, 106 (2011) P. Milde et al., Science 340, 1076 (2013)
I. Kézsmárki et al., Nat. Mater. 14, 116 (2015)
• Spin-polarised STM • Resonant soft x-ray scattering
• Ultrasonic measurements • PEEM
• Microwave spectroscopy • Classical measurements, M(H), χac
• Muon spin rotation • Transport (Hall effect) Page 12
 Table of DMI Skyrmion host materials
Material Crys. SG Skyrm. Tc λ (nm) Trans. References
Class Type
MnSi T P213 Bloch 30 K 18 Metal S. Mühlbauer et al., Science 323, 915
(2009)

FeGe T P213 Bloch 279 K 70 Metal X.Z. Yu et al., Nat. Mater. 10, 106 (2010)

Fe1-xCoxSi T P213 Bloch < 36 K 40-230 Semi- W. Münzer et al., PRB 81, 041203(R)
(2010)
cond.
X.Z. Yu et al., Nature 465, 901 (2010)

Mn1-xFexSi T P213 Bloch < 17 K 10-12 Metal S.V. Grigoriev et al., PRB 79, 144417 (2009)

Mn1-xFexGe T P213 Bloch < 220 K 5–220 Metal K. Shibata et al., Nat. Nanotech. 8, 723–
728 (2013)

Cu2OSeO3 T P213 Bloch 58 K 60 Insulator S. Seki et al., Science 336, 198 (2012)
T. Adams et al., PRL 108, 237204 (2012)

CoxZnyMnz O P4132 Bloch 150 K – 120 – Metal Y. Tokunaga et al., Nat. Commun. 6, 7638
(2015)
/P4332 500 K 200
(Fe,Co)2Mo3N O P4132 Bloch < 36 K 110 Metal W. Li et al., Phys. Rev. B 93, 060409(R)
(2016)
/P4332
GaV4S8 C3v R3m Néel 13 K 17 Semicond/ I. Kézsmárki et al., Nat. Mater. 14, 1116
Insulator (2015)

GaV4Se8 C3v R3m Néel 18 K 24 Semicond/ Previous talk!

Insulator

Page 13
 Nature is generous!
Non-centrosymmetric crystal classes

More Skyrmion host systems with novel functionalities are waiting to be discovered!
P. S. Halasyamani and K. R. Poeppelmeier, Chem. Mater. 10, 2753 (1998) Page 14
III. Getting out of a tight spot
Understanding Skyrmion phase stability

Page 15
 Important factors beyond DMI
MnSi Cu2OSeO3 Co-Zn-Mn alloys

S. Mühlbauer et al., Science (2009) S. Seki et al., Science (2012) Y. Tokunaga et al., Nat. Commun. (2015)

Skyrmion state stabilised by Gaussian fluctuations? Longitudinal magnetisation fluctuations

U.K. Rössler et al., Nature 442, 797 (2006).
A.N. Bogdanov and D.A. Yablonskii, Sov. Phys. JETP 68, 101 (1989).

Uniaxial anisotropy
(with no conical phase)
GaV4S8

I. Kézsmárki et al., Nat. Mater.

14, 116 (2015)
S. Mühlbauer et al., Science 323, 915 (2009)
Page 16
 Thin samples of chiral magnets
Fabrication of thin films (bottom-up) / thin plates (top-down)
MnSi (50nm thick) FeGe (15nm thick)

A. Tonomura et al., Nano Lett., 12, 1673 (2012) X.Z. Yu et al., Nat. Mater. 10, 106 (2010)

Uniaxial anisotropy is enhanced by reducing the sample thickness, d

competing conical phase suppressed when d ~ λ
A.B. Butenko et al., PRB 82, 052403 (2010)
S.D. Yi et al., Phys. Rev. B 80, 054416 (2009)

Going even thinner.. Page 17

 Ultrathin metallic films
Interfaces break inversion symmetry! PdFe bilayer on Ir(111) – 3 nm diameter Skyrmions

Ingredients to induce a DMI (spin twisting):

• An interface!
• Ultrathin magnetic bi- or monolayer, e.g. PdFe or Fe
• Bottom layer of heavy metal with SOC, e.g. Ir

Writing and erasing of individual Skyrmions

SP-STM imaging

four-spin interaction

N. Romming et al., Science 341, 636 (2013)

S. Heinze et al., Nat. Phys., 7, 713 (2011)
But at < 10 K .. Page 18
 Multilayers and room temperature Skyrmions
Ir/Co/Pt multilayer with additive interfacial chiral interactions strong DMI at RT
C. Moreau-Luchaire et al., Nat. Nanotech. 11, 444 (2016) Scanning X-ray transmission microscopy

Small < 100 nm 200 nm

wide tracks
diameter Néel-
type Skyrmions

300 nm
diameter T = 300 K
[Ir(1 nm)/Co(0.6 nm)/Pt(1 nm)]10 disks µ0H⊥ = 8 - 50 mT

Room temperature, current pulse-driven motion of skyrmions at high speeds

S. Woo et al., Nat. Mater. 15, 501 (2016)
400 nm diameter Néel- T = 300 K
type Skyrmions µ0H⊥ < 2 mT

[Pt(3 nm)/Co(0.9 nm)/Ta(4 nm)]15

[Pt(4.5 nm)/CoFeB(0.7 nm)/MgO(1.4nm)]15

A. Soumyanarayanan et al., Nature 539, 509 (2016)

Multilayers seem very promising for skyrmionics: see R. Wiesendanger, Nat. Rev. Mater. 1, 16044 (2016) Page 19
IV. Back to chiral magnets:
Consequences of the topology

Page 20
 Spin textures and emergent fields
In general… Development of an electron Non-coplanar spin structure
state vector in a distorted
parameter space

si sj sk

→Berry phase, gauge fields Berry phase:

Topological spin texture

-Electrons adiabatically traversing the Skyrmion spin texture adapt to
the local M and acquire a quantum-mechanical Berry phase.

-The Berry phase physics is associated with emergent EM fields:

‘Emergent flux’ B is
A. Rosch topologically quantised!
T. Schulz et al., Nat. Phys. 8, 301 (2012)
Page 21
 Experimental detection by Hall effect
Normal Hall effect Anomalous Hall effect Topological Hall effect MnSi

A. Neubauer et al., PRL 102, 186602 (2009)

Size of the topological Hall response is proportional to Bie

gives a direct measure of the flux quantum!

Quantised emergent flux:

Average Bie in Skyrmion phase of MnSi ≈ 11 T!

Page 22
 Detecting and driving skyrmion motion
Skyrmion motion is detectable by the THE MnSi
N. Nagaosa and Y. Tokura, Nat. Nanotech. 8, 899 (2013) T. Schulz et al., Nat. Phys. 8, 301 (2012)

The topological Hall effect in MnSi is

more pronounced above jc due to Eie

Current-driven Skyrmion motion is an

example of spin-transfer torques
Magnus force (Berry phase)

Page 23
 Spin transfer torques and SANS
Further indications of Skyrmion motion in MnSi F. Jonietz et al., Science 330, 1648 (2010)
+j -j

Compare: Current pulses drive the movement of FM domain

walls by STTs:

Why are very small current densities needed to

drive skyrmions compared with FM domains? Page 24
 Results from theory
• Efficient coupling between the conduction Numerical study using the LLG equation
electrons and emergent B-field (Berry phase)
that generates Magnus forces
F. Jonietz et al., Science 330, 1648 (2010)
K. Everschor et al., PRB 84, 064401 (2011), PRB 86, 054432 (2012)

• No Magnus forces in helical (trivial) phase

current–velocity relations similar to those of a FM
domain wall

• Weak pinning/alignment effects due to the large

length-scale of the Skyrmion lattice: λSkL >> a

• Skyrmions in motion are flexible: can deform

their shape and twist their trajectory to avoid Current density
pinning to defects/disorder. t=0 t = 65 ns
See: J. Iwasaki et al., Nat. Commun. 4, 1463 (2013)

STTs also generated by magnon currents Motion driven by currents of electrons or

deflected by emergent B-field of Skyrmions magnons relies on the emergent fields of the
M. Mochizuki et al., Nat. Mater. 13, 241 (2014) topological skyrmion spin structure
Page 25
 Electric field manipulation
Magnetoelectric insulator Cu2OSeO3 Electric field control of the Skyrmion lattice
S. Seki, et al., Science 336, 198, (2012) JSW et al., Phys. Rev. Lett. 113, 107203 (2014)
T. Adams et al., Phys. Rev. Lett. 108, 237204 (2012)

Class T (P2 13)

Magnetoelectric coupling in Skyrmion phase

Skyrmion lattice rotations achieved with

Symmetry of ME coupling: d-p hybridization model no STTs – i.e. no current or magnon flows
For more on the ME coupling, see also:
E. Ruff et al., Sci. Rep. 5, 15025 (2015) Microscopic explanation for rotation still missing!
P. Milde et al., Nano Lett. 16, 5612 (2016) Page 26
V. Some interesting topics

Page 27
 But certainly not limited to ..
Artificial Skyrmions Diverse topological spin textures
Mn1.4Pt0.9Pd0.1Sn - LTEM
A. K. Nayak et al., arXiv: 1703.01017

D2d symmetry

Implant a magnetic vortex into a

film with PMA
Skyrmions by design ‘Antiskyrmions’ spotted?

Skyrmions by magnetic frustration? Gigahertz oscillators + optical dichroism

A.O. Leonov and M. Mostovoy, Nat. Commun. 6, 8275 (2015)
Cu2OSeO3

Stability mechanism without DMIs Novel microwave functionalities Page 28

 Summary Slides: https://www.psi.ch/lns/jonathan-white
• Skyrmions are topologically non-trivial particle-like objects composed of magnetic spins.

• Identifiable using a variety of experimental techniques.

• SANS, LTEM, Hall effect, MFM, STM ..

• Skyrmions are well-established to be stabilised by Dzyaloshinskii-Moriya interactions in a

few bulk non-centrosymmetric materials like MnSi, GaV4S8.
• Other stability mechanisms I did not discuss in detail: dipolar fields, frustration, four-
spin interactions

• Skyrmions host quantised emergent electromagnetic fields in the presence of a flow of

conduction electrons.
• The conduction electrons exert forces via
STTs on skyrmions driving their motion.

• Electric fields can drive skyrmion motion in insulators.

• Open questions still to be addressed

• Ultrathin metallic films and multilayers host Néel-type skyrmions at room temperature
that can be controlled by electric currents.
• Most promising for Skyrmionics!

There are many talks on Skyrmions at this meeting!

Page 29
 Advertisement

SNF Sinergia project:

“Discovery and Nanoengineering of Novel Skyrmion-hosting Materials”
Start: Oct. 2017 for 4 years 2 post-docs and 6 PhDs

Dr. J.S. White 1 PhD

[email protected] 1 post-doc (2yr +)
- Application of large-scale facility techniques at PSI (SINQ + SLS)
- Real-space imaging at EPFL/PSI - LTEM, PEEM

large-scale facilities Prof. O. Yazyev

Prof. D. Grundler [email protected]
J.S. White
[email protected] (scattering)
- First-principles-
- New approaches to materials theory & chemical
science Room- synthesis driven materials
nanofabrication of temperature
D. Grundler O. Yazyev
(materials discovery
Skyrmion materials (nanostructures) discovery)
Skyrmionics - Synthesis
- Device fabrication & micro-
engineering

School of Engineering (STI) M. Poggio School of Basic

(sensing) Sciences (SB)
2 PhDs experimental 1 PhD
physics
1 post-doc (theory)
Prof. M. Poggio
[email protected]
Start date: 2 PhDs Contact us if
Summer/Fall 2017 - New techniques in nanosensing applied
interested!
to Skyrmion-hosting nanostructures
Page 30