2d materials - lqss.epfl.ch · 2 wte 2 10 mm mos 2 5 mm wse 2 5 mm 2μm bi 2 se 3 10 mm hbn...
TRANSCRIPT
2D Materialsan introduction starting from the discovery of graphene
• Developments in material control now enable perfect crystals only one or a few atom thick to be investigated
• The very broad variety of 2D materials availablegives access to unexplored physical phenomena
• Realization of artificial materials with properties engineeredby design at the atomic scale
Alberto Morpurgo
Philosophy: some idea is better than no idea
Natural Graphite Exfoliation
Transfer Contact
Novoselov/Geim2004/2005
Seeing one-atom layers one at a time
Two inequivalent C atoms † †
,i ii j iR RR R
i j
A AH Bt B
† †| | 0i i
i j
ik R ik R
k k kR Ri i
e BAe
Or†
†| | 0
i
j
k R
k
i k R
iik R
B
Ae
pseudo
spin
Relativistic electrons in graphene
| ( ) |H E k
𝑣𝐹(−𝑖ℏ𝛻) ∙ Ԧ𝜎𝛼𝑘𝑒
𝑖𝑘∙ Ԧ𝑟
𝛽𝑘𝑒𝑖𝑘∙ Ԧ𝑟
=
= E(k) 𝛼𝑘𝑒
𝑖𝑘∙ Ԧ𝑟
𝛽𝑘𝑒𝑖𝑘∙ Ԧ𝑟
Dirac EquationH | ۧ𝜓𝑘 = 𝐸(𝑘) | ۧ𝜓𝑘
Novoselov/Geim/Kim2005
Direct Experimental manifestationsGraphene field-effect transistor
Dirac “peak” 𝐺𝐻𝑎𝑙𝑙 = 4𝑒2
ℎ(𝑁 +
1
2)
Resistivity Quantum Hall effect
Quasiclassical orbits
Landau levels = quantization of cyclotron motion
Edge states
Transport through edge channels
Quantum Hall effect for dummies
𝐻 = 𝑣𝐹 −𝑖ℏ𝛻 ∙ Ԧ𝜎= ħ𝑣𝐹0 𝑘𝑥 − 𝑖𝑘𝑦
𝑘𝑥 + 𝑖𝑘𝑦 0
Band structure topology: basic idea
𝐵 𝑘 = 𝑘𝑥 , 𝑘𝑦 , 0
yk
?xk
Think of Hamiltonian as a k-dependent magnetic field
Bloch functions wind around in the Brillouin zone = topologically non-trivial
Graphene
𝐻 = 𝐴Δ/2 + 𝑏𝑘2 0
0 −Δ
2− 𝑏𝑘2
=
= 𝐴 𝐵𝑧𝜎𝑧
𝐵 𝑘 = 0,0, Δ/2 + 𝑏𝑘2
𝐻 = 𝛼 𝐵 𝑘 ∙ Ԧ𝜎Trivial insulator
yk
xk
No winding
Non-trivial topology leads to states at the edges
Different thickness = different electronic systems
𝐸 𝑘 ∝ 𝑘
𝐻1𝐿 ∝0 𝑘𝑥 − 𝑖𝑘𝑦
𝑘𝑥 + 𝑖𝑘𝑦 0
Monolayer Bilayer
𝐸 𝑘 ∝ 𝑘 2𝜓 =𝜓𝐴
𝜓𝐵𝜓 =
𝜓𝐴1
𝜓𝐵2
𝐻2𝐿 ∝0 𝑘𝑥 − 𝑖𝑘𝑦
2
𝑘𝑥 + 𝑖𝑘𝑦2
0
Low-energyband
Gate control of electronic bands
Double-gated devices
V= D/2
V= − D/2
Layer potential
𝐻2𝐿 ∝
Δ
2𝑘𝑥 − 𝑖𝑘𝑦
2
𝑘𝑥 + 𝑖𝑘𝑦2
−Δ
2
𝐸 𝑘 ∝ Δ + 𝑎𝑘 2Open band-gap
Turn bilayer grapheneinto an insulator
Manipulating atomic crystalsP. Kim’s group 2010
Gra
ph
ene
on
hB
N
Gra
ph
ene
on
WS 2
Kim
’s g
rou
pMoire superlattice for graphene on hBN
Contact with hBNgenerates periodic potential:
Graphene band structure modified
Satellite Dirac points
Contact mode AFM
removes residues of
nano-fabrication processes
Optical image Before cleaning During cleaning After cleaning
Cleaning 2D materials
5 mm
CrI3
Graphene
20 mm
Phosporene
NbSe2
WTe210 mm
MoS2
5 mm WSe2 5 mm
2μmBi2Se3
10 mmhBN
Advances in Material Control
. . .
Theory predicts that ~ 1000 different atomic crystals can be produced using similar techniques
A virtually infinite variety of systemsMore than 50 monolayers demonstrated, including: Insulators, semiconductors, semi-metals, topological insulators, superconductors, charge density waves, ferromagnets, antiferromagnets…
Exfoliate/transfer/encapsulate in glove box with controlled atmosphere
Can work with materials not stable in airExample: CrI3 --- the first ferromagnetic monolayer
Thin crystals (even 50 nm) dissolve in a few minutes
No
vose
lov
20
11
The Grand Vision – already becoming reality
Examples of interesting physics in 2D materials
Goal: to illustrate the breadth of scope --- no details
• Gate-induced superconductivity in semiconducting transition metal dichalcogenides (TMDs)
• Giant tunneling magnetoresistance through 2D magnetic semiconductor tunnel barriers
• Edge conduction in monolayer WTe2 (2D topological insulator)
Semiconducting transition metal dichalcogenides
A. S
ple
nd
ian
iet
al.
NL
10, 1
271
(201
0)
Bulk 3L 2L 1L
• Honeycomb (= graphene), • A/B atom different (=gapped), • very strong SOI
MoS2, WS2, MoSe2, WSe2, …
Ionic liquid gating
Position
Po
ten
tia
l
Ga
te
Dev
ice
Surf
ace
Electrostatics of ionic liquidssimilar to metals =
atomic-scale screening length
Potential drops at interfaces
Ex: WS2
Vsd = 100 mV
Gate-induced superconductivity in MoS2 ionic-liquid FETs
Evo
luti
on
of
the
resi
sta
nce
wit
hg
ate
volt
ag
e(i
on
icliq
uid
ga
tea
nd
ba
ckg
ate
)
Zero resistance state
Persists in monolayers
MoS2 bilayer
MoS2 monolayer
Gate-induced superconductivity in 1L MoS2 with reduced Tc and Hc
-40 -20 0 20 40
-8
-4
0
4
8
7K
V (
mV
)
I (mA)
1.5K
Tunneling spectroscopy
Multilayer graphene as tunneling electrode; MoS2 as tunnel barrier
Tunneling into the gate-induced electron accumulation layer
An idea of fabrication steps
Bottom G multilayer Define tunneling contacts Transfer MoS2 multilayer
Att
ach
met
alli
c co
nta
cts
& c
lea
n t
he
Mo
S 2su
rfa
ce
Tunneling spectroscopy of gate-induced superconductivity
Tunneling spectroscopy for
1014 cm-2 < n < 5 1014 cm-2
Incomplete DOS suppression & “V-shaped” low-energy DOSUnconventional superconductivity
2D magnetic materials
Does magnetism persists in 2D magnetic materials?How can we probe it? Is there new physics?
Examples
We will discuss experiments on one compound: CrI3
Hysteretic Kerr rotation angleas expected for ferromagnets
Magneto-optical Kerr effect
But note: Kerr effect probes broken symmetries not magnetization
Extremely unstable in air: Thin crystals (even 50 nm) dissolve in a few minutes
Exfoliating/processing/encapsulating in Glove box
CrI3 tunnel barriers
Thin CrI3 as tunnel barrierbetween graphene contacts
Optical microscope
Tunneling
T = 51 K
Tunneling Magnetoresistance
Giant tunneling magneto-resistance: 100 x at B = 2T
For comparison: common tunneling spin-valves = 2 x
Extracting magneticphase diagram from
tunneling magnetoresistance
Monolayer WTe2: topological insulator with edge transport
WS2 = Semiconductor = trivial insulatorConductance vanishes
(exponentially small ∝ 𝑒−∆
2𝑘𝑇)
WTe2 = topological insulator
Conductance at low T:
𝐺~𝑒2
ℎTopological insulator = Edge conductance
-40 -20 0 20 40
-8
-4
0
4
8
7K
V (
mV
)
I (mA)
1.5K
yk
xk
The exploration of atomically thin crystals has only just started
This is a vast and largely unexplored field of research