Home Search Collections Journals About Contact us My IOPscience

Relaxation and edge reconstruction in integer quantum Hall systems

This content has been downloaded from IOPscience. Please scroll down to see the full text.

2012 New J. Phys. 14 105009

(http://iopscience.iop.org/1367-2630/14/10/105009)

View the table of contents for this issue, or go to the journal homepage for more

Download details:

IP Address: 84.25.233.197
This content was downloaded on 31/08/2015 at 09:16

Please note that terms and conditions apply.

 New Journal of Physics The open–access journal for physics

Relaxation and edge reconstruction in integer

quantum Hall systems
Torsten Karzig1,4 , Alex Levchenko2 , Leonid I Glazman3
and Felix von Oppen1
1
Dahlem Center for Complex Quantum Systems and Fachbereich Physik,
Freie Universität Berlin, 14195 Berlin, Germany
2
Department of Physics and Astronomy, Michigan State University,
East Lansing, MI 48824, USA
3
Department of Physics, Yale University, 217 Prospect Street, New Haven,
CT 06520, USA
E-mail: [email protected]

New Journal of Physics 14 (2012) 105009 (14pp)

Received 1 June 2012
Published 10 October 2012
Online at http://www.njp.org/
doi:10.1088/1367-2630/14/10/105009

Abstract. The interplay between the confinement potential and the

electron–electron interactions causes reconstructions of quantum Hall edges.
We study the consequences of this edge reconstruction for the relaxation of
hot electrons injected into integer quantum Hall edge states. In translationally
invariant edges, the relaxation of hot electrons is governed by three-body colli-
sions, which are sensitive to the electron dispersion and thus to reconstruction
effects. We show that the relaxation rates are significantly altered in different
reconstruction scenarios.

4
Author to whom any correspondence should be addressed.

Content from this work may be used under the terms of the Creative Commons Attribution-NonCommercial-
ShareAlike 3.0 licence. Any further distribution of this work must maintain attribution to the author(s) and the title
of the work, journal citation and DOI.
New Journal of Physics 14 (2012) 105009
1367-2630/12/105009+14$33.00 © IOP Publishing Ltd and Deutsche Physikalische Gesellschaft
2

Contents

1. Introduction 2
2. Unreconstructed edge 3
2.1. Three-body scattering formalism . . . . . . . . . . . . . . . . . . . . . . . . . 5
2.2. Results for the unreconstructed edge . . . . . . . . . . . . . . . . . . . . . . . 6
3. Spin reconstruction 7
4. Charge reconstruction 8
5. Conclusions 10
Acknowledgments 11
Appendix A. Calculation of the Coulomb matrix element Vq (k1 − k2 ) 11
Appendix B. Calculation of the three-body matrix element T123
10 20 30 12
References 13

1. Introduction

The kinetic properties of one-dimensional (1D) quantum systems are an active area of current
research [1, 2]. What makes the field exciting is that many-particle physics is drastically
different in one spatial dimension. This difference is already evidenced in basic nonequilibrium
properties such as the microscopic mechanisms of relaxation. Within the scope of Fermi-liquid
theory, relaxation processes in higher dimensions proceed by pair collisions of electrons, which
provide an efficient mechanism for the relaxation of initial nonequilibrium states. In contrast,
conservation of energy and momentum strongly restricts scattering in one spatial dimension
so that pair collisions necessarily result in zero-momentum exchange or an interchange of
the momenta of the colliding particles. Neither process causes relaxation. This poses the
fundamental question of the microscopic origin of relaxation in 1D systems. Notably, the
absence of relaxation by pair collisions, which holds true regardless of the strength of
interaction, has received experimental support [3].
The question of equilibration emerges in a diverse set of 1D many-body systems. These
include energy and momentum-resolved tunneling experiments with nanoscale quantum wires
[3, 4], quench dynamics of cold atomic gases [5, 6] as well as energy-spectroscopy experiments
on quantum Hall edge states driven out of equilibrium [7–10]. This paper is motivated by the
latter experiments, which were carried out in a high-mobility two-dimensional electron system
at Landau-level filling factor ν = 2. This system hosts two co-propagating edge states, which can
be driven out of equilibrium by inter-edge tunneling in the vicinity of quantum point contacts.
This generates a nonequilibrium distribution of electron energy (in the sense of electronic edge
transport) downstream from the contact, which is monitored as a function of the propagation
distance by means of a quantum-dot-based energy spectrometer. The experiments show that
the initial nonequilibrium distribution relaxes to a stationary form that is close to the thermal
distribution but with an effective temperature and chemical potential.
Edge-state equilibration was also probed in experiments at Landau-level filling factor
ν = 1 [11]. Heat is carried unidirectionally by the single chiral edge mode as confirmed by
thermopower measurements along the edge. These experiments found that hot electrons injected
locally into the edge cool down while propagating along the edge. It is worth emphasizing that

New Journal of Physics 14 (2012) 105009 (http://www.njp.org/)

3
the standard chiral-Luttinger-liquid model for quantum Hall edge states [12] does not account
for equilibration effects. Indeed, this model is exactly solvable and, as usual, its integrability
is an obstacle to thermalization. In early works [13] this apparent difficulty was overcome
by assuming a disordered edge where impurity-mediated scattering allows for interchannel
equilibration.
These experimental discoveries led to a flurry of theoretical activity. We briefly summarize
these contributions and place our work in their context. Two early publications [14, 15]
used entirely different concepts. The work by Lunde et al [14] was based on a Boltzmann
kinetic equation for a disordered edge. Since translation invariance is broken, momentum
is no longer a good quantum number and relaxation becomes possible even by two-particle
collisions. Degiovanni et al [15] adopted a bosonization approach and combined it with a
phenomenological model for the plasmon distribution generated at the quantum point contact.
Within this model, thermalization was interpreted to be a consequence of plasmon dispersion,
which causes the electron wave packets to broaden as they propagate with different group
velocities. This picture was elegantly elaborated and extended in [16–18]. A third mechanism
was proposed in the context of electronic Mach–Zehnder interferometers [19, 20] based on
electron–plasmon scattering [21]. This mechanism relies on the scattering of high-energy
electrons by low-energy plasmons enabled by the curvature of the fermionic spectrum.
Despite the insight provided by these theories, important issues need to be sorted out.
First, these works do not give a definitive answer to the question whether relaxation is
possible in translationally invariant clean edges. Specifically, the dispersion of plasmon modes
may lead to a steady state but does not constitute true relaxation as the energy in each
plasmon mode is conserved. Second, the edge of quantum Hall systems can be reconstructed
due to Coulomb interactions. The precise nature of reconstruction depends on the steepness
of the confinement potential, ranging from no reconstruction for very sharp confinement
potentials [22] to alternating compressible and incompressible stripes for very smooth edges
[23]. Indeed, experiments [24, 25] point to an important role of reconstruction effects in energy
transfer along the edge.
The purpose of this paper is to address these issues within minimal models of
unreconstructed and reconstructed edges. Specifically, we consider energy relaxation of a hot
particle injected into translationally invariant quantum Hall edges at Landau-level filling factors
ν = 1, 2. With the assumption that the velocity v1 of the injected particle differs sufficiently
from the Fermi velocity vF , we treat the Coulomb interaction perturbatively [26, 27]. In this
limit, relaxation processes are dominated by three-body collisions which depend sensitively on
the electron dispersion and hence on the edge reconstruction. We begin with a discussion of
energy relaxation for the unreconstructed edge in section 2. We then discuss two simple models
of reconstructed quantum Hall edges. In section 3, we discuss relaxation processes for a spin-
reconstructed edge for filling factor ν = 2. In section 4, we turn to a minimal model of charge
reconstruction of a ν = 1 edge, which provides the simplest realization of counter-propagating
edge modes. We conclude in section 5.

2. Unreconstructed edge

A confinement potential Vc (x) that is sharp on the scale of the Coulomb interaction
(i.e. Vc0  e2 /(κl 2B ), where κ is the dielectric constant and l B denotes the magnetic length)
remains stable against interaction-induced reconstructions and the electron dispersion ε(k)

New Journal of Physics 14 (2012) 105009 (http://www.njp.org/)

4

Figure 1. Edge without reconstruction effects. (a) Single-particle dispersions and

typical relaxation processes in the the lowest Landau level for a spin degenerate
system. Vertical arrows label temperature T and cyclotron energy ωc = eB/m.
Panel (b) shows the corresponding sharp (T = 0) occupation number in terms of
the guiding center position X .

can be obtained approximately from the noninteracting Schrödinger equation [22]. A generic
electronic dispersion of an unreconstructed edge is sketched in figure 1(a), exhibiting a
confinement-induced bending of the Landau levels near the edge of the sample.
In the limit of high magnetic fields (Vc0  ωc /l B ), the electron states near the edge can be
described by the lowest Landau-level wave functions
√
ψ X (x, y) = (Ll B π )−1/2 e−iky e−(x−X ) /2l B
2 2
(1)
in the Landau gauge. Here, k = X/l 2B and L denotes the length of the sample edge (taken along
the y direction). The defining feature of the unreconstructed edge is the sharp zero-temperature
occupation function νσ (X ) = 2(−X ) of Landau-level states with the guiding center X when
the Zeeman splitting εZ is negligible (see figure 1(b)).
The single-particle dispersion near the Fermi energy (corresponding to momentum kF ) is
controlled by the confinement potential and can be approximated as
ε(k) = vF (k − kF ) + (k − kF )2 /2m c . (2)
The dispersion is parametrized through the edge velocity vF = Vc0l 2B at the Fermi energy and
the curvature 1/m c = Vc00l 4B . Note that these parameters become maximal for an infinitely sharp
edge for which vF ∼ ωcl B and 1/m c ∼ 1/m [22]. Note, however, that a description in terms of
the wave functions in equation (1) is no longer valid in this extreme limit.
The finite curvature of the dispersion implies that at least three particles are required
for an energy and momentum conserving relaxation process. Relaxation of a high-energy
electron (labeled by i = 1 in figure 1(a)) is possible by scattering two electrons (labeled i = 2, 3
in figure 1(a)) near the Fermi energy. Indeed, due to the curvature of the dispersion near
the Fermi energy, exciting electron i = 2 from the Fermi energy requires more energy than
scattering electron i = 3 deeper into the Fermi sea. Clearly, this relaxation process relies on
finite temperature, and the typical energy transfers for electrons i = 2, 3 at the Fermi energy
are of the order of T . Quantitatively, this process can relax the hot particle with excess
energy ε ≈ vF (k1 − kF ) = vF 1k by q1 = q3 (k20 − k3 )/1k, where qi = ki 0 − ki is the momentum
transferred to particle i in the collision. Note that q1  q3 so that relaxation occurs in many
small steps vF q1 ∼ T 2 /ε.
For Landau-level filling factor ν = 2, these considerations apply when the Zeeman splitting
is small compared to the temperature. In the opposite limit εZ  T , the curvature of the
New Journal of Physics 14 (2012) 105009 (http://www.njp.org/)
5
dispersion implies that the Fermi momenta and hence the Fermi velocities differ for the two
spin directions. In this case, relaxation is dominated by processes in which the electrons i = 1
and 2 have opposite spins and thus different Fermi momenta kF j and Fermi velocities v j with
j = 1, 2. To include a finite Zeeman splitting at Landau-level filling factor ν = 2 as well as for
later convenience, it is thus beneficial to consider a modified dispersion
v (k − kF j ), k ≈ kF j ,

ε(k) = j (3)
v1 (k − k1 ) + ε, k ≈ k1 ,
which is linearized in the vicinity of each of the three particles, including the hot particle with
velocity v1 and momentum k1 . This captures the behavior in the regime of strong Zeeman
splitting εZ  T , which we will focus on in the following. Nevertheless, we can also recover the
results for the quadratic dispersion and weak Zeeman splitting εZ  T by identifying v2 − v3
with the typical velocity difference T /(vF m c ) due to the curvature of the dispersion.
Using the dispersion in equation (3), energy and momentum conservation leads to
v2 − v3
q1 = q3 . (4)
v1 − v2
The velocity difference v2 − v3 = εZ /(v2 m c ) is controlled by the Zeeman splitting, which we
assume to be small compared to the excitation energy ε such that (v2 − v3 )  (v1 − v2 ).

2.1. Three-body scattering formalism

Energy relaxation by processes of the kind shown in figure 1(a) has already been discussed in
the context of quantum wires in [26]. While our calculation here follows the same outline, there
are characteristic differences related to the nature of the interaction matrix elements. The energy
relaxation rate via three-body collisions is again given by
1 X −v1 q1
= 0 20 30 n 2 n 3 (1 − n 10 )(1 − n 20 )(1 − n 30 ),
W1123 (5)
τE 2310 20 30 ε
where n i is the Fermi–Dirac distribution function at ki . The factor involving q1 weights the
out-scattering rate with the relative relaxed energy, accounting for the fact that the hot particle
relaxes only a fraction of its energy in a single collision. The three-body matrix element can be
evaluated by using the generalized golden rule
0 20 30 = 2π |h1 2 3 |V G 0 V |123ic | δ(E − E ).
W1123 0 0 0 2 0
(6)
Here, G 0 is the free Green’s function, V = (1/2L) k1 k2 qσ1 σ2 Vq (k1 − k2 )ak†1 +qσ1 ak†2 −qσ2 ak2 σ2 ak1 σ1
P
is the generic two-body interaction potential and the subscript c emphasizes that only connected
processes contribute that involve all three particles.
The calculation for quantum Hall edges differs from that for quantum wires in the form
of the Coulomb matrix element Vq (k1 − k2 ), which now has to be evaluated using the Landau-
level wave functions in equation (1). For quantum Hall systems, the Coulomb matrix element
is exponentially suppressed by a factor of exp(−q 2l 2B /2) for large momentum transfers. This
is particularly relevant because large momentum transfers yield the leading contribution to
relaxation in quantum wires [26]. Moreover, Vq (k1 − k2 ) depends not only on the momentum
transfer but also on the initial momentum difference that controls the distance between the

New Journal of Physics 14 (2012) 105009 (http://www.njp.org/)

6
guiding centers of the interacting electrons. Focusing on the remaining small momentum
transfer processes (q  1/l B ), one obtains (see appendix A for details)
2
(
− 2eκ ln |ql B |, B ,
k1 − k2  l −1
Vq (k1 − k2 ) ' 2 (7)
B ,
− 2eκ ln |q(k1 − k2 + q)l 2B |, k1 − k2  l −1
with the understanding that at small q, the matrix elements will be eventually cut off by a large
length scale λ  l B which is given by the distance to a screening gate. For k1 − k2  1/l B ,
the Coulomb matrix element is that of a quantum wire of width l B . For k1 − k2  1/l B , the
interaction is that of electrons in two quantum wires separated by a distance of (k1 − k2 + q)l 2B ,
which is equal to the average of the guiding center distances of the electrons before and after
the collision.
With the absence of large momentum transfer processes the three-body scattering is
dominated by the direct matrix element. The importance of the (k1 − k2 ) dependence of the
Coulomb matrix element can be seen from the fact that the linearized dispersion of equation (3)
leads to a vanishing direct matrix element for a quantum-wire-like Coulomb interaction Vq (0)
(see appendix B). In contrast, when reiterating the derivation [28] of the direct matrix element
0 20 30 including the dependence of Vq (k 1 − k 2 ) on the initial momenta, the result does not vanish
T1123
and takes the form
2e2 k2 − k3 Vq3 (k2 − k3 ) 2Vq3 (1k) v2 − v3 Vq1 (1k)
 
123
T10 20 30 = − 2 − + . (8)
L κ v2 − v3 (1k)2 (v1 − v2 ) (1k) k2 − k3 (v1 − v2 )2
Here we used k1 − k2 ≈ k1 − kF2 = 1k. This expression is applicable under the assumption that
all initial momentum differences are large compared to 1/l B to also suit the reconstruction
effects that will be discussed later. For the unreconstructed edge, it is however more reasonable
to assume k2 − k3  1/l B (for typical εZ , T  e2 /κl B ), in which case the last term of
equation (8), involving Vq1 , does not show up (see appendix B).

2.2. Results for the unreconstructed edge

For the unreconstructed edge, the momentum and velocity differences are linked by the
curvature of the confinement potential via v2 − v3 = (k2 − k3 )/m c and v1 − v2 = 1k/m c . The
direct matrix element then takes the form
2e2 m c 
T1123 Vq3 (k2 − k3 ) − 2Vq3 (1k) .

0 20 30 = − (9)
L κ(1k)
2 2

Since for large Zeeman energy the particles at k2 and k3 have opposite spins, there is no
exchange contribution (remember that exchange is appreciable for small momentum transfers
only) and equation (9) fully determines the three-body matrix element. The corresponding
energy relaxation rate can then be obtained by power counting, which yields
 2 4
1 e (m c v12 )4 T 4
∼ m c (v2 − v3 )2
. (10)
τE κv2 ε8
Here 1/m c = Vc00l 4B , (v2 − v3 ) = εZ /(v2 m c ) and we also used 1k = m c (v1 − v2 ) = ε/v1 . In
obtaining equation (10), a factor of L/(v1 − v2 ) emerges from eliminating the energy δ-function
in equation (5), each summation over the remaining k2 , k3 , q3 contributes a phase space factor
of ∼ T /v2 and the weighting factor v1 q1 /ε takes the form v1 (v2 − v3 )T /[v2 (v1 − v2 )ε]. Finally,
we have to account for the competition between excitation (q1 > 0) and relaxation (q1 < 0) of

New Journal of Physics 14 (2012) 105009 (http://www.njp.org/)

7
the hot particle. The latter is slightly favored because the momentum transfer working against
the Fermi distribution is reduced by a fraction v2 q1 /T ∼ (v2 − v3 )/(v1 − v2 ).
Equation (10), valid at ε  m c v1 e2 /κ and εZ  T , implies that the relaxation rate is
strongly temperature dependent and can be enhanced by increasing the magnetic field.
As mentioned above, the relaxation rate in the opposite limit of weak Zeeman splitting
εZ  T can be obtained up to prefactors by replacing (v2 − v3 ) ∼ T /(v2 m c ). Note that this
regime allows for a small momentum transfer exchange term T1123 0 30 20 because particles 2 and 3

are no longer necessarily of opposite spin. T1123 0 30 20 can then be obtained from equation (9) by

replacing q3 → k20 − k3 , which does not change the power counting argument.
It is therefore possible to combine both cases by setting v2 − v3 = max{εZ , T }/(v2 m c ). In
the case of v1 ≈ v2 = Vc0l 2B , it is then possible to rewrite equation (10) as
4 4
max{εZ , T }2  Vc0 2 e2

1 T
∼ , (11)
τE Vc00l B2
Vc00l B κl B ε8
which applies for the regime Vc0l 2B  ε(Vc00l 2B /Vc0 )  e2 /κ.
For a later comparison of the relaxation rates before and after edge reconstruction, it will
be useful to consider the unreconstructed case as the v1  v2 ∼ e2 /κ limit of equation (10)
(which can be applied for ε  (e2 /κl B )2 /(Vc00l 2B )). Formally, this regime leaves the condition for
applicability of the Taylor expansion of the confinement potential that defines 1/m c = Vc00 (µ)l 4B
and would lead to another inverse mass Vc00 (µ + ε)l 4B for curvature effects at energies of the order
of ε. Distinguishing these different masses does not, however, lead to qualitative changes in the
results, and for brevity of the presentation we assume a quadratic confinement potential over the
energy interval [µ, µ + ε]. We can then rewrite equation (10) as
max{εZ , T } 2 00 2
T 4
   
1
∼ Vc l B , (12)
τE(u) e2 /(κl B ) ε
where we used that in this regime, m c v12 ∼ . The crossover between equations (11) and (12)
can be obtained at their limits of applicability by setting ε = e2 Vc0 /(κl 2B Vc00 ) and Vc0 = e2 /(κl 2B ).
Note that for a spin polarized edge, equations (10)–(12) apply only if the Coulomb
interaction is not screened for momenta of the order of T /v2 . For a screened short-range
interaction (T /v2  1/λ), the Pauli principle then leads to a suppression of the energy relaxation
rate by an additional factor of (T k/v2 )4  1 [27].

3. Spin reconstruction

The edge reconstruction in quantum Hall systems results from competition between the
Coulomb interaction and the confinement potential. Spin reconstruction at ν = 2 takes place
when the confinement potential Vc varies sufficiently slowly so that Vc0 < e2 /κl 2B and can be
understood at the level of the Hartree–Fock approximation [29–32]. Once the slope of the
confinement potential becomes weaker than that of the repulsive Hartree potential VH , it is
favorable to deposit charges outside the edge. This can be done without paying exchange energy
by a relative shift of the Fermi momenta of spin-up and spin-down particles, as depicted in
figure 2. In the absence of Zeeman splitting, εZ = 0, this is a second-order phase transition with
spontaneous breaking of the spin symmetry. Then, the distance between the two Fermi momenta
varies as kF2 − kF3 ∝ (|VH0 | − Vc0 )1/2 , eventually saturating at ∼ 1/l B [29]. For finite Zeeman

New Journal of Physics 14 (2012) 105009 (http://www.njp.org/)

8

Figure 2. Spin reconstructed edge for εZ = 0. (a) Depicts the Hartree–Fock

single-particle dispersion in the reconstructed region. Note that we set
the curvature of the confinement potential to zero such that the velocity
difference v2 − v3 > 0 is not obvious from the figure. (b) Shows the T = 0
occupation numbers of the different spin species in terms of the guiding center
coordinate X .

splitting εZ , the spin symmetry is lifted by the Zeeman field and the transition is smeared on the
scale of kF2 − kF3 ∼ εZ /v2 .
Spin reconstruction leads to characteristic changes in the single-particle dispersion that
develops an ‘eye structure’ (see figure 2(a)). Important for the relaxation dynamics is the
increase of v2 − v3 = (kF2 − kF3 )/m c , which enhances the typical energy transferred per step
of relaxation (see equation (4)).
For truly long-range interactions, the Hartree–Fock approximation predicts a logarithmic
singularity ∼ e2 /j ln(|k − kF |l B ) of the particle velocity at the Fermi energy, which is however
cut off in the presence of screening, say by a nearby gate electrode. The Fermi velocity is thus
still of the order of v2 , v3 ∼ e2 /κ for typical choices of the screening length.
Even with spin reconstruction, the relaxation of hot particles can be described within the
model dispersion of equation (3). We consider the case when the hot particle (not shown in
figure 2) is injected well outside the energy window e2 /(κl B ) of the reconstructed region. This is
compatible with the condition for validity of a perturbative expansion, which reduces to v1  v2
for the case of the Fermi velocity determined by the interaction.
The energy relaxation rate 1/τE(s) can now be derived in the same way as for the
unreconstructed edge, and consequently, equation (10) applies also to spin reconstructed edges.
The crucial difference is that the velocity difference v2 − v3 is now strongly enhanced by
the spin reconstruction, taking values up to v2 − v3 ∼ 1/(m cl B ). Comparing the rates before
[v2 − v3 ∼ max{εZ , T }/(m c v2 )] and well after spin reconstruction, we find an enhancement of
the relaxation rates given by
2
e /(κl B )
 2
1 1
(s)
∼ . (13)
τE max{εZ , T } τE(u)

4. Charge reconstruction

For confinement potentials that vary even more smoothly, changing by e2 /κl B over a region
w > l B , charge reconstruction may occur such that some of the electrons at the edge are pushed
away from the bulk by a length of the order of l B [31, 32]. It leads to nonmonotonic behavior

New Journal of Physics 14 (2012) 105009 (http://www.njp.org/)

9

Figure 3. Charge reconstructed edge. (a) Depicts the Hartree–Fock single-

particle dispersion in the reconstructed region of a spin polarized sample at
ν = 1. (b) Shows the occupation number near the edge of the sample.

of the dispersion with momentum and the creation of two additional counter-propagating5 edge
modes, as depicted in figure 3.
A minimal model for charge reconstruction considers filling factor ν = 1 within the
Hartree–Fock approximation [31]. It is convenient to formally model the confinement potential
by a positive background charge which is distributed spatially as if it was occupying the lowest
Landau-level wave functions ψ X with occupation numbers νc (X ) = 2(−X ). The advantage
of this model is that such a confinement potential exactly cancels the Hartree potential
of the electrons for an unreconstructed edge. In this case, the electron occupation of the
unreconstructed edge is stabilized by the (attractive) exchange potential.
The reconstruction transition can then be modeled by changing the abrupt drop of νc (X )
into a linear decrease over a length w. For the unreconstructed electron occupations, this leads
to negative (at X < 0) and positive (at X > 0) excess charges, causing a dipole field that favors
separating electrons from the bulk. Once this dipole field overcomes the exchange potential,
a charge reconstruction transition takes place. Within the Hartree–Fock approximation, this
happens for w ∼ 8l B . Due to the particle–hole symmetric choice of the confinement potential
around X = 0, the width and distance of the additional stripe from the bulk electron droplet
both take the same value b. Moreover, the transition is of first order in the sense that b changes
abruptly at the transition from zero to a value of the order of l B .
Note that the same mechanism induces new (weaker) effective dipole fields at each of the
three Fermi points as the edge becomes yet smoother. Thus, increasing w even further causes
additional stripes to appear, eventually approaching the limit of a compressible edge which is
expected for w  l B [23]. In the following we will focus on w & l B , remaining well outside the
compressible limit.
Energy relaxation in the charge reconstructed case can also be captured by the dispersion
(3) when setting v3 < 0 and choosing the particle i = 2 to lie in one of the co-propagating
branches6 . The three Fermi velocities of the charge reconstructed edge are essentially
determined by the variation of the exchange potential, which is short range such that b & l B
already approximates the bulk edge (b → ∞) behavior. Consequently, the magnitudes of the
5
Note that the conductance may stay at the ν = 1 level due to impurity-induced elastic backscattering of electrons
at the Fermi level. At higher energies, inelastic scattering leads to a shorter mean free path and we can therefore
omit the disorder-induced backscattering.
6
Processes where the particle i = 2 is on the inner or outer co-propagating branch contribute at the same order
in the power counting argument. The choice depicted in figure 3 is slightly favored because of a larger Coulomb
matrix element due to the shorter distance to the hot particle.

New Journal of Physics 14 (2012) 105009 (http://www.njp.org/)

10
Fermi velocities are equal to that of the unreconstructed edge and ∼ e2 /κ. In line with the
discussions above, we consider the relaxation of a hot particle injected well outside the
reconstructed region with v1  e2 /κ.
The nonmonotonic behavior of the dispersion introduces a new relaxation process which
relaxes the hot particle by exciting two counter-propagating electron–hole pairs (see figure 3).
This eliminates the restriction that the energy transfers at the Fermi energy cannot exceed the
temperature, and makes the relaxation process similar to that for nonchiral quantum wires [26].
Unlike for quantum wires, however, the momentum transfers at the Fermi energy of the co- and
counter-propagating branches are of the same order.
The three-body matrix element of equation (8) still applies in the presence of charge
reconstruction because its derivation did not require a specific sign of v3 . Note, however,
that for the charge reconstructed case v2 − v3 ∼ v2 ∼ e2 /κ and is therefore not connected to
kF2 − kF3 ∼ 1/l B by the curvature of the confinement potential. Assuming that there are no
substantial curvature effects on the scale of the reconstructed region Vc00l 2B  e2 /(κl B ) the last
term of equation (8) dominates the three-body matrix element and equation (9) modifies to
2e2 e2 /(κl B )
T1123
0 20 30 = − Vq (1k). (14)
L 2 κ(1k)2 (Vc00l 2B )2l 2B 1
The crucial difference for the energy relaxation rates compared to the unreconstructed
case arises from the large allowed momentum q3 ∼ 1/l B , which is limited only by the size
of the reconstructed region for which the linearized dispersion applies. This increases both
the momentum phase space to (L/l B )3 and the typical relaxed momentum to (v2 − v3 )/[(v1 −
v2 )l B ]. Moreover, excitation and relaxation processes no longer need to be balanced when
e2 /κl B  T , and we find
3 q 4
e /(κl B ) e /(κl B )
 2  2
1
∼ εVc l B
00 2
, (15)
τE(c) Vc00l 2B ε
which applies for ε  (e2 /κl B )2 /(V 00l 2B ) and allows for relaxation even at T = 0. Equation (15)
implies that the increased phase space and the energy relaxation step size lead to a dramatic
enhancement of the relaxation rate compared to the unreconstructed case (see equation (12)) as
3  2 6
ε e /(κl B ) e /(κl B )
 2
1 1
r
∼ , (16)
τE(c) V 00l 2B Vc00l 2B T τE(u)
where we used the limit T  εZ .
00
To obtain an estimate of the relaxation rates, we assume that Vc l 2B is of the order of the
Coulomb interaction scale e2 /(κl B ), say, e2 /(κl B ) ∼ 30 K for B = 5 T. Then, we find relaxation
lengths of the order of micrometers for excitation energies of the order of ωc ∼ 100 K. The
experiments in [7–9] are conducted at significantly lower excitation energies < 1 K, making real
collisions conserving both energy and momentum an unlikely origin of the observed relaxation.

5. Conclusions

We studied three-body processes as an intrinsic mechanism for relaxation of hot electrons

in clean integer quantum Hall edges at Landau-level filling factors ν = 1 and 2. These
processes rely crucially on the form of the electron dispersion and are thus susceptible to
edge reconstruction effects. For an unreconstructed edge, energy relaxation requires a finite

New Journal of Physics 14 (2012) 105009 (http://www.njp.org/)

11
temperature which determines the phase space for the relaxation processes. The energy given
up by the hot electron in a single three-body collision is controlled by curvature effects on the
scale of temperature or Zeeman energy so that the relaxation rate can be tuned by a magnetic
field once εZ  T .
While unreconstructed edges are expected for steep confinement potentials, smoother
confinement potentials with Vc0 . e2 /(κl 2B ) may lead to an interaction-induced spin
reconstruction, which causes a relative shift of the Fermi momenta of the two spin species
by ∼ 1/l B . The three-body processes are then controlled by curvature effects on the scale
of the interaction energy e2 /(κl B ), which causes a strong increase of the relaxation rate (see
equation (13)).
Even softer confinement may cause charge reconstruction, which introduces additional co-
and counter-propagating edge modes. The presence of counter-propagating modes allows for
relaxation even at T = 0. Consequently, the phase space for three-body collisions is no longer
controlled by temperature but by the size of the reconstructed region ∼ e2 /(κl B ), which ensures
an additional dramatic enhancement of the relaxation rate (see equation (16)).
Experimental studies of interaction-induced reconstruction transitions in high magnetic
fields have been performed [24]. Our study suggests that it would be rewarding to
experimentally investigate relaxation processes in such systems.

Acknowledgments

We acknowledge financial support from the Deutsche Forschungsgemeinschaft under SPP 1538
(FUB) and by DOE under contract no. DEFG02-08ER46482 (Yale) and by the US–Israel BSF
under grant no. 2010366 (Yale). AL acknowledges support from Michigan State University.

Appendix A. Calculation of the Coulomb matrix element Vq (k1 − k2 )

Within this section we provide all essential details needed for the derivation of equation (7)
presented in the main text. We assume that the edge is smooth enough that we can approximate
the electron wave functions by those of the bulk. We start from the interaction matrix element
in real space
 e2 /κ  (1) (2)
   
(1) (2)
VdX (X, X ) = ψ X +dX , ψ X 0 −dX  (1)
0   ψ , ψX 0 . (A.1)
|r − r(2) |  X
In the following, we will measure all length scales in units of magnetic length l B . In these units
the guiding center coordinate directly translates to momenta. With the lowest Landau-level wave
functions of equation (1), we then find that
√
2 Z − 1x 2 +1y 2 /λ
e e
e−i1y dX e− 2 (x−X ) e− 2 (x−X −dX )
1 2 1 2
VdX (X, X 0 ) = dx dy d1x d1y p
πκ L 2
1x 2 + 1y 2
× e− 2 (x+1x−X ) e− 2 (x+1x−X +dX ) ,
1 0 2 1 0 2
(A.2)
√
where we used the screened Coulomb potential which carries an extra factor e− 1x +1y /λ
2 2

with λ being the distance to a screening gate. The integration over 1y gives 2K 0 (|1x dX |)
in the case when dX  1/k, where K 0 is the Bessel function of imaginary argument. If,
however, dX  1/λ the integral is cut off and the result changes to 2K 0 (|1x/λ|). We will derive
New Journal of Physics 14 (2012) 105009 (http://www.njp.org/)
12
results for the dX  1/λ case and keep in mind appropriate changes for the other limit. After
y integration, which gives a factor of L, we obtain the intermediate step
2e2 − 1 dX 2
Z
dx d1x K 0 (|1x dX |)e− 2 (2x−X −X +1x) e− 2 (X −X +dX +1x) .
1 0 2 1 0 2
VdX (X, X ) =
0
e 2 (A.3)
πκ L
√
Performing now the Gaussian integral over x, which gives a factor of π/2, followed by using
the Landau gauge to replace guiding center coordinates by momenta, one arrives at
r
2 e2 −q 2 /2
Z
dξ e− 2 (k+q+ξ ) K 0 (|ξ q|),
1 2
Vq (k) = e (A.4)
π κ
where we used the short-hand notation X − X 0 = kl 2B = k (with l B = 1). Note that Vq (k) =
Vq (−k − 2q) and is therefore not symmetric, which plays an important role. We see immediately
that scattering processes with large momentum transfer q  1 are exponentially suppressed. We
therefore concentrate on the opposite limit of q  1 when the exponential prefactor e−q /2 can
2

be set to unity.
Let us study limiting cases of equation (A.4). In the√
case when k  1, one can approximate
the exponential under the integral by the delta-function 2πδ(k + q + ξ ) and thus obtain
2e2
Vq (k) = K 0 (|q(k + q)|). (A.5)
κ
Using the asymptotic form of the Bessel function and restoring units of l B , one recovers the
second limit in equation (7).
In the other limiting case when k  1, one can approximate the exponential under the
integral of equation (A.4) by e−ξ /2 and then complete integration exactly with the result
2

e2
Vq (k) = K 0 (q 2 /4). (A.6)
κ
With the logarithmic accuracy at small q this translates into the first limit of equation (7).

Appendix B. Calculation of the three-body matrix element T123

10 20 30

In general, the three-particle scattering amplitude h10 20 30 |V G 0 V |123ic contains six terms: one
direct and five exchange contributions [28]. As explained in the text, we need only the former
one, which reads explicitly [28]
δ6,6 0 Vk30 −k3 (k3 − k2 )Vk10 −k1 (k20 − k10 ) Vk10 −k1 (k1 − k3 )Vk20 −k2 (k30 − k20 )

123
T10 20 30 = 2 +
L εk3 + εk2 − εk30 − εk2 +k3 −k30 εk1 + εk3 − εk10 − εk3 +k1 −k10
Vk20 −k2 (k2 − k1 )Vk30 −k3 (k10 − k30 ) Vk20 −k2 (k2 − k3 )Vk10 −k1 (k30 − k10 )
+ +
εk2 + εk1 − εk20 − εk1 +k2 −k20 εk2 + εk3 − εk20 − εk3 +k2 −k20
Vk10 −k1 (k1 − k2 )Vk30 −k3 (k20 − k30 ) Vk30 −k3 (k3 − k1 )Vk20 −k2 (k10 − k20 )

+ + , (B.1)
εk1 + εk2 − εk10 − εk2 +k1 −k10 εk3 + εk1 − εk30 − εk1 +k3 −k30
where the spin structure is δ6,6 0 = δσ1 ,σ10 δσ2 ,σ20 δσ3 ,σ30 and the Coulomb matrix element Vq (k) was
derived in appendix A. Now using the dispersion relation from equation (3) and the constraint

New Journal of Physics 14 (2012) 105009 (http://www.njp.org/)

13
on momentum transfers from equation (4) imposed by the conservation laws, one can simplify
T1123
0 20 30 to

δ6,6 0 Vq3 (k3 − k2 )Vq1 (k1 − k2 + q3 ) − Vq1 (k1 − k2 )Vq3 (k3 − k2 + q1 )


123
T10 20 30 ≈ 2
L q3 (v2 − v3 )
Vq (k2 − k3 )Vq1 (k1 − k3 + q2 ) − Vq1 (k1 − k3 )Vq2 (k2 − k3 + q1 )
+ 2
q2 (v3 − v2 )
Vq3 (k3 − k1 )Vq2 (k2 − k1 + q3 ) − Vq2 (k2 − k1 )Vq3 (k3 − k1 + q2 )

+ , (B.2)
q3 (v1 − v3 )
where we used the property Vq (k) = Vq (−k − 2q). It is important to stress that the above
expression would vanish by ignoring the dependence of the Coulomb matrix element on initial
momenta, namely for Vq (k) = Vq . To proceed further we make use of the assumption that the
injected particle is of high energy, such that v1  v2,3 and k1  k2,3 . In this case we expand
2
Vqi (1k + qi ) in qi . For the interaction Vq (k) = − 2eκ ln(|kq|l 2B ), we obtain after the expansion
2e2 δ6,6 0 Vq3 (k3 − k2 ) Vq1 (k1 − k2 ) Vq2 (k2 − k3 )

123
T10 20 30 ≈ − − +
κL 2 (v2 − v3 )(k1 − k2 ) (v1 − v2 )(k2 − k3 ) (v2 − v3 )(k1 − k3 )
Vq1 (k1 − k3 ) Vq3 (k3 − k1 ) Vq2 (k2 − k1 )

+ + + . (B.3)
(v1 − v3 )(k2 − k3 ) (v1 − v3 )(k1 − k2 ) (v1 − v2 )(k1 − k3 )
Note that if we are in the regime when k2 − k3  l −1 B we have to use the interaction po-
2e2
tential Vq (k) = − κ ln(|q|l B ), which has a vanishing derivative with respect to k. This
can be accounted for by removing the two terms with Vq1 (. . .) in the above formula for
0 20 30 . Finally, to leading logarithmic order we can set Vq1 (k 1 − k 2 ) = Vq1 (k 1 − k 3 ) as well
T1123
as Vq2 (k2 − k1 ) = Vq3 (k3 − k1 ) = Vq3 (k1 − k2 ) and Vq3 (k3 − k2 ) = Vq2 (k2 − k3 ) to obtain equa-
tion (8) since the spin summation is equal to unity.

References

[1] Deshpande V V, Bockrath M, Glazman L I and Yacoby A 2010 Nature 464 209
[2] Imambekov A, Schmidt T L and Glazman L I 2011 arXiv:1110.1374
Imambekov A, Schmidt T L and Glazman L I 2011 Rev. Mod. Phys. submitted
[3] Barak G, Steinberg H, Pfeiffer L N, West K W, Glazman L, von Oppen F and Yacoby A 2010 Nature Phys.
6 489
[4] Chen Y-F, Dirks T, Al-Zoubi G, Birge N O and Mason N 2009 Phys. Rev. Lett. 102 036804
[5] Kinoshita T, Wenger T and Weiss D S 2006 Nature 440 900
[6] Hofferberth S, Lesanovsky I, Fischer B, Schumm T and Schmiedmayer J 2007 Nature 449 324
[7] Altimiras C, le Sueur H, Gennser U, Cavanna A, Mailly D and Pierre F 2010 Nature Phys. 6 34
[8] le Sueur H, Altimiras C, Gennser U, Cavanna A, Mailly D and Pierre F 2010 Phys. Rev. Lett. 105 056803
[9] Altimiras C, le Sueur H, Gennser U, Cavanna A, Mailly D and Pierre F 2010 Phys. Rev. Lett. 105 226804
[10] Paradiso N, Heun S, Roddaro S, Sorba L, Beltram F and Biasiol G 2011 Phys. Rev. B 84 235318
[11] Granger G, Eisenstein J P and Reno J L 2009 Phys. Rev. Lett. 102 086803
[12] Wen X G 1990 Phys. Rev. Lett. 64 2206
Wen X G 1991 Phys. Rev. B 43 11025
[13] Kane C L and Fisher M P A 1994 Phys. Rev. B 51 13449
Kane C L and Fisher M P A 1995 Phys. Rev. B 52 17393
Kane C L and Fisher M P A 1997 Phys. Rev. B 55 15832

New Journal of Physics 14 (2012) 105009 (http://www.njp.org/)

14

[14] Lunde A M, Nigg S E and Büttiker M 2010 Phys. Rev. B 81 041311

[15] Degiovanni P, Grenier C, Fève G, Altimiras C, le Sueur H and Pierre F 2010 Phys. Rev. B 81 121302
[16] Kovrizhin D L and Chalker J T 2011 Phys. Rev. B 84 085105
[17] Kovrizhin D L and Chalker J T 2011 arXiv:1111.3914
[18] Levkivskyi I P and Sukhorukov E V 2012 Phys. Rev. B 85 075309
[19] Ji Y, Chung Y, Sprinzak D, Heiblum M, Mahalu D and Shtrikman H 2003 Nature 422 415
[20] Neder I, Heiblum M, Levinson Y, Mahalu D and Umansky V 2006 Phys. Rev. Lett. 96 016804
[21] Heyl M, Kehrein S, Marquardt F and Neuenhahn C 2010 Phys. Rev. B 82 033409
[22] Halperin B I 1982 Phys. Rev. B 25 2185
[23] Chklovskii D B, Shklovskii B I and Glazman L I 1992 Phys. Rev. B 46 4026
[24] Barak G, Pfeiffer L N, West K W, Halperin B I and Yacoby A 2010 arXiv:1012.1845
[25] Deviatov E V, Lorke A, Biasiol G and Sorba L 2011 Phys. Rev. Lett. 106 256802
[26] Karzig T, Glazman L I and von Oppen F 2010 Phys. Rev. Lett. 105 226407
[27] Micklitz T and Levchenko A 2011 Phys. Rev. Lett. 106 196402
[28] Lunde A, Flensberg K and Glazman L I 2007 Phys. Rev. B 75 245418
[29] Dempsey J, Gelfand B Y and Halperin B I 1993 Phys. Rev. Lett. 70 3639
[30] Stoof T H and Bauer G E W 1995 Phys. Rev. B 52 12143
[31] Chamon C de C and Wen X G 1994 Phys. Rev. B 49 8227
[32] Barlas Y, Joglekar Y N and Yang K 2011 Phys. Rev. B 83 205307

New Journal of Physics 14 (2012) 105009 (http://www.njp.org/)