emil j. bergholtz- one-dimensional theory of the quantum hall system
TRANSCRIPT
-
8/3/2019 Emil J. Bergholtz- One-dimensional theory of the quantum Hall system
1/29
One-dimensional theory ofthe quantum Hall system
Anders Karlhede (Stockholm)
Thors Hans Hansson (Stockholm)Janik Kailasvuori (Berlin)Emma Wikberg (Stockholm)Maria Hermanns (Stockholm)Eddy Ardonne (Caltech/Nordita)
Juha Suorsa (Helsinki/Oslo)Susanne Viefers (Oslo/Stockholm)
with:
Seminar atENS, ParisNovember 2008
Emil J. Bergholtz(MPI-PKS, Dresden)
-
8/3/2019 Emil J. Bergholtz- One-dimensional theory of the quantum Hall system
2/29
I will:
1. Introduce the quantum Hall effect (QHE)
2. Solve it exactlyin a simple well defined limit
3. Convince you that this limit is relevant(I hope.......)
a) Hierarchyb) Gapless states
c) Non-abelian states
-
8/3/2019 Emil J. Bergholtz- One-dimensional theory of the quantum Hall system
3/29
I
Vx
Vy
2 dimensional electron gas
x
y
B
T 0
The basic experiment
1. Sendcurrent I through sample
Vx Vyand2. Measure
Rxx = Vx/IRxy = Vy/I3.Plot and as functions ofB
Low temperature T - near 0KStrong magnetic field BClean sample
-
8/3/2019 Emil J. Bergholtz- One-dimensional theory of the quantum Hall system
4/29
(Willett et al1987)
Rxx
(h /e2)
Rxy
B (incompressiblequantum liquids)
Quantum Hall states
Rxy =1
s
h
e2Rxx = 0
h Plancks constanteelectron charge
Metallic states
(compressiblequantum liquids)
s =1
2,1
4, . . .s 1/B
1/21234s
s = 1, 2, 3,... Integer QHE
von Klitzing, Dorda, Pepper 1980
Fractional QHE
Tsui, Strmer, Gossard 1982
s =
1
3 ,
2
5 ,
3
7 . . .
Non-abelian
Fractional QHE?
Willet et al 1987
s =5
2, . . .
-
8/3/2019 Emil J. Bergholtz- One-dimensional theory of the quantum Hall system
5/29
Problem to understand:
What are they?
Groundstate and low energy excitationsdetermine everything.
Cold electrons moving on a surface in a perpendicularmagnetic field.
(+disorder)
Is the system gapped?
-
8/3/2019 Emil J. Bergholtz- One-dimensional theory of the quantum Hall system
6/29
ONE ELECTRON IN A MAGNETIC FIELD B(two dimensions)
En = (n + 12)hc n = 0, 1, 2, 3...
c =eB
mc
Kinetic energy quantized
Landau levels
0
3
2
1
E/hc
spin
= number of filled LL (filling factor)Define
s 1/B
Each Landau level (LL) is highly degenerate.
One state per flux quantum 0 = hc/e
# of states B
The crucial parameter!
(within a Landau level)
Quantized area of electron state 10 nm
c
eB =
257AB
1 teslaA = 2 ;
-
8/3/2019 Emil J. Bergholtz- One-dimensional theory of the quantum Hall system
7/29
= 1IQHE
unique state
empty one electron state
filled one electron state
Fixed area per state 0 = hc/e(one state per flux quantum )
?
incompressible liquid
!
= 1/3FQHE
673132974506580171230064
degenerate states
10 nm
(Real samples much bigger)
-
8/3/2019 Emil J. Bergholtz- One-dimensional theory of the quantum Hall system
8/29
Standard theory - brief review
e = e/(2m + 1)incompressible quantum liquid with particles = 1/(2m+ 1)
The fractionally charged qps condense and form an incompressible liquid just aselectrons condensed at 1/(2m+1). Iterating this gives all odd. = p q, qHierarchy:
(Haldane 83, Halperin 83)
(Jain 89)
Composite
fermions:
and/or
Electrons form new particles, composite fermions,by absorbing magnetic flux. FQHE is IQHE of these composite fermions.Gives directly. = p/(2mp + 1)
= p/qBut other fractions less clear. = p/q, q(All odd, experimentally similar.)
Many-body wave functions
(Moore and Read 91)Non-abelions Appear (?) in higher Landau levels (and/or in rotating condensates)Motivated by conformal field theory (CFT-FQHE correspondence)
...and it goes on.....
(Laughlin 83)
Gapless states (Halperin, Lee and Read 93)Half filled Landau level; free composite fermions
-
8/3/2019 Emil J. Bergholtz- One-dimensional theory of the quantum Hall system
9/29
= 1/3
(By Kwon Park)
composite fermion
Composite fermions at
Nice pictures, but (how) does it work?
(cf. ) = 1
incompressible liquid
!
Supported by:
Numerical tests of wavefunctions
Mean field theory and exp. near = 1/2
But:
Microscopic understanding
is lacking... (my opinion)
-
8/3/2019 Emil J. Bergholtz- One-dimensional theory of the quantum Hall system
10/29
L1
L2
Each electron state has fixed area
(any shape is OK)
choose
(Landau gauge)
Consider sample with lenghts L1, L2
L2
L1Two dim electron gas
Our approach: a one-dimensional view(and a solvable limit)
-
8/3/2019 Emil J. Bergholtz- One-dimensional theory of the quantum Hall system
11/29
L1
L2
One-dimensional lattice model
Each box is either empty 0 or filled 1
A possible state at = 1/31 0 0 1 1 0 0 0 1 0 0 0 0 1 0
1..0.....0..1 0..1.....1..0 (all ee-terms that preserve position of CM)
Hamiltonian (ee-interaction) (egCoulomb V(r)=e /r)2 No kinetic energy!
k+m k-m
Vk,m
(electrostatic repulsion)including
1......1 Vk,0k
x
y k eik 2L1
ye(xk 2
L1)2/2
(Landau gauge)
Exact mapping of a single Landau level!
-
8/3/2019 Emil J. Bergholtz- One-dimensional theory of the quantum Hall system
12/29
Idea: physics as varies L2 fixed and largeL1
L1
Experimental situation
Complicated 1D interaction
L1 L
2 large
L2
Smooth development (we claim)
2/L1(Lattice constant: )
Simple 1D interaction
Solvable limit
smallL1
-
8/3/2019 Emil J. Bergholtz- One-dimensional theory of the quantum Hall system
13/29
Exact solution
Hopping 1..0.....0..1 0..1.....1..0 makes ground state complicated.
L1 0But, when
hopping vanishes and only electrostatic repulsion remains: 1......1
This is a simple classical electrostatics problem!
States with electrons in fixed positions are the energy eigenstates -
groundstate obtained by separating the electrons as much as possible:
= 2/5
1 0 0 1 0 1 0 0 1 0 1 0 0 1 0 1 0 0 1 0
= 1/3
Unit cell
1 0 0 1 0 0 1 0 0 1 0 0 1 0 0 1 0 0
(Tao-Thouless (TT)states)
At ground state is TT-state with p electrons in unit cell of length q. = p q
gapped crystal
-
8/3/2019 Emil J. Bergholtz- One-dimensional theory of the quantum Hall system
14/29
We claim:
Motivation is to come.
* Same qualitative properties
The TT-state at any odd, is a QH-state!1. = p/q, q
The stability of the state decreases when q increases.2.
(More stable states are seen at higher disorder.)
L* Develops smoothly as no phase transition
The QH hierarchy
-
8/3/2019 Emil J. Bergholtz- One-dimensional theory of the quantum Hall system
15/29
100100100100100100100100100 groundstate
10010010100100100100100100 -e/3(domain wall separating degenerate groundstates)
(At , quasiparticles with charge ) = p/q e = e/q
Charge determined by Su-Schrieffers counting argument:
100100100100100100100100100100100100100100100100100100
100101001001001001010010010010010010100100100100100100
-2e-3e
Excited states: fractional charge as L 0
= 1/3
Quasielectrons with charge -e/3 are obtained by inserting a 10:
e = e/3
-
8/3/2019 Emil J. Bergholtz- One-dimensional theory of the quantum Hall system
16/29
Stability of states ~ 1/q
Disorder destroys a QH-state - the more energy it takes
to disturb a state, the more stable it is.
-e/q +e/q
The smallest energy disturbance is creation of a pair:e/q
Stability ~ gap ~ 1/q
(This argument is correct in spirit, not in detail.
However, the result is correct.)
-
8/3/2019 Emil J. Bergholtz- One-dimensional theory of the quantum Hall system
17/29
Parents and daughters - the hierarchy
Add many quasielectrons 10 to 1/3, lowest energy states:
100101001010010100101001010010100101001010010100101001010010.... 10210, = 2/5
10010010010100100100100101001001001001010010010010100100100100.... (102)310, = 4/11
100100100100100100100100100100100100100100100100100100.... 100 = 102, = 1/3
100100100100100101001001001001001001010010010010010010010.... (102)510, = 6/17
Condensation of quasielectrons in give states with
unit cells - the ground states at .(102)2m110 = 2m/(6m 1) = 1/3
1/q
p/q
Stability
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 10
0.1
0.2
0.3
0.4
0.5
0.6
!!p/q
"#q
Fractal structure
-
8/3/2019 Emil J. Bergholtz- One-dimensional theory of the quantum Hall system
18/29
Phase transitions:
Two assumptions:
1) Condensation of quasielectrons or quasiholes
gives phase transition from one QH state to
another
2) More stable states are seen at higher disorder
leads to......
-
8/3/2019 Emil J. Bergholtz- One-dimensional theory of the quantum Hall system
19/29
......the phase diagram
0.25 0.3 0.35 0.4 0.45 0.50
0.05
0.1
0.15
0.2
0.25
0.3
1/3
2/5
2/7 3/7
3/11
4/9
4/11
4/13 5/135/17 6/17
6/19
7/177/19
8/19
!=p/q
1/q
disorder
1/4 1/2Shown here for
- all regions look the same.
Experiment, varying B
.....at lower disorder
-
8/3/2019 Emil J. Bergholtz- One-dimensional theory of the quantum Hall system
20/29
The observed states:
Stability ~ 1/q agrees with experiment
0.2 0.4 0.6 0.8 1
p
q
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
1
q
range of experiment
stability
CF states
non-CF states
States observed by
Pan et al (2003)
-
8/3/2019 Emil J. Bergholtz- One-dimensional theory of the quantum Hall system
21/29
TT-states 100, ... develop smoothly into bulk QH-states Surprising - but true!
Same qualitative properties (gap, quasiparticles, ...)
New unique bulk wave function for any state that is obtained by
successive condensation of quasielectrons. (conformal field theory construction)
Gives TT-states as L 0
Gives Laughlin/Jain wave functions where these exist
Going to the bulk L the crucial question
Numerical studies (exact diagonalization and DMRG)
Supported by numerics for simplest non-L/J state = 4/11
Overall picture. More to come...
Same structure from a different perspective
Proven for Laughlin states and pseudopotential interaction
-
8/3/2019 Emil J. Bergholtz- One-dimensional theory of the quantum Hall system
22/29
Gapless states =1/2
But: metallic, ie gapless state as (experiment & theory)L ?!
This solution develops smoothly as L (numerical evidence)
TT-state 10101010.... ground state when (has a gap )L 0
Phase transition to gapless state at finite L
Interacting electrons in magnetic field free neutralparticles in 1 dim
composite fermions!?
(valid at small but finite L)- exact solution for gapless state
-
8/3/2019 Emil J. Bergholtz- One-dimensional theory of the quantum Hall system
23/29
Exact solution at = 1/2
Hamiltonian: V211001 0110 V10 = 2V20and (rest =0)
(Good approximation for small L)
Ground state is 1d Fermi sea of free neutral dipoles
composite fermions!?
Free particles not affected by magnetic field
The TT state melts at finite L
(Note the contrast to the1/3 state 100100100...)
101010... is annihilated by the shortest range hopping 1001 0110Competition between electrostatics and hopping
Luttinger liquid!Corrections
= V21
i
(s+is
i+1+H.c.)
10 , 01
Low energy sectorcontained in spin space
-
8/3/2019 Emil J. Bergholtz- One-dimensional theory of the quantum Hall system
24/29
Luttinger liquid(s) describe the bulk at half-filling!?
Towards the bulk...L( )
9.08.2 8.4 10.6
10101010....
0.993 0.996 0.995 0.998
L5.3
(5,2) (6,4) (8,3) (0,0)
1
Solvable!Luttinger liquid
Solvable!
Smooth transitions
Overlap1.000
Exact diagonalization:
Bulk state believed to be described by the Rezayi-Read wave function
free 2d composite fermions
-
8/3/2019 Emil J. Bergholtz- One-dimensional theory of the quantum Hall system
25/29
Non-trivial groundstate degeneracy
Non-abelian QH statesL 0(as )
Domain wall excitations with non-trivial chargeand degeneracy.
Crucial for non-abelian statistics!
(Examples to come....)
Consistent with CFT results where these exists (and
gives clues on how to construct new theories).
Appear (?) in higher Landau levels (2D electrons and/orin rotating condensates)
Interaction (matrixelements) changes
-
8/3/2019 Emil J. Bergholtz- One-dimensional theory of the quantum Hall system
26/29
The Moore-Read pfaffian(the example....)
Six ground states
101010101010101010101010
110011001100110011001100
Three-body interaction ---never three electrons on four sites
Half quasiholes as domain walls
101010101010101010101010101010101010101010101010
101010100110011001100101010101001100110011001010 e
Supposedly describes the state = 5/2 ( 1/2)
e/4 e/4 e/4 e/4
e
= e/4
-
8/3/2019 Emil J. Bergholtz- One-dimensional theory of the quantum Hall system
27/29
Quasihole degeneracy
10101010100110011001100
100110011001100or Two possibilities!1010101010
Create a pair of qhs
Leads to 3x2n/2 degenerate states for n qh s at fixed positions.
Generalizes to quasiparticles by particle hole symmetry.
-
8/3/2019 Emil J. Bergholtz- One-dimensional theory of the quantum Hall system
28/29
Total degeneracy for n quasiholes is given by the number of periodic paths...
This can be generalized..... (here for the k=4, M=0 bosonic Read-Rezayi state)
These domain walls encode the fusion rules of the pertinent CFT
[04]
[13]
[22]
[31]
[40]
ddd
ddd
ddd
ddd
ddd
. . .
. . .
ddd
ddd
ddd
. . .
. . .
n = 0 1 2 3 4 5
Bratteli diagram
-
8/3/2019 Emil J. Bergholtz- One-dimensional theory of the quantum Hall system
29/29
Conclusion
The quantum Hall problem is exactly solvable in a limitthat accommodate the rich structure of the system.
The solution is smoothly connected to the experimental
regime.
References:
Half-filling: Hierarchy: Non-abelian states:
Bergholtz and Karlhede,PRL 94, 026802, (2005)
Bergholtz and Karlhede,J. Stat. Mech. L04001, (2006)
Bergholtz, Hermanns, Hansson,and Karlhede,
PRL 99, 256803 (2007)
Bergholtz and Karlhede,PRB 77, 155308 (2008)
Bergholtz, Hermanns, Hansson,
Karlhede and Viefers
Bergholtz, Kailasvuori, Wikberg,Hansson, and Karlhede,
PRB 74, 081308(R) (2006)
Ardonne, Bergholtz, Kailasvuori, Wikberg,J. Stat. Mech. P04016, (2008)