51
ong-range correlations in driven system David Mukamel

Long-range c orrelations in driven systems David Mukamel

  • Upload
    coye

  • View
    50

  • Download
    0

Embed Size (px)

DESCRIPTION

Long-range c orrelations in driven systems David Mukamel. Outline Will discuss two examples where long-range correlations show up a nd consider some consequences. Example I : Effect of a local drive on the s teady state of a system - PowerPoint PPT Presentation

Citation preview

Page 1: Long-range  c orrelations in driven systems David  Mukamel

Long-range correlations in driven systems

David Mukamel

Page 2: Long-range  c orrelations in driven systems David  Mukamel

Outline

Will discuss two examples where long-range correlations show upand consider some consequences

Example I: Effect of a local drive on the steady state of a system

Example II: Linear drive in two dimensions: spontaneous symmetry breaking

Page 3: Long-range  c orrelations in driven systems David  Mukamel

Example I :Local drive perturbation

T. Sadhu, S. Majumdar, DM, Phys. Rev. E 84, 051136 (2011)

Page 4: Long-range  c orrelations in driven systems David  Mukamel

N particlesV sites

Particles diffusing (with exclusion) on a grid

𝑝 (𝑘)= 𝑁𝑉Prob. of finding a particle at site k

Local perturbation in equilibrium

occupation number

Page 5: Long-range  c orrelations in driven systems David  Mukamel

N particlesV sites

Add a local potential u at site 0

The density changes only locally.

0 1𝑒𝛽𝑢

1

1

1

1

1

Page 6: Long-range  c orrelations in driven systems David  Mukamel

Effect of a local drive: a single driving bond

Page 7: Long-range  c orrelations in driven systems David  Mukamel

In d ≥ 2 dimensions both the density corresponds to a potential ofa dipole in d dimensions, decaying as for large r.The current satisfies .

The same is true for local arrangements of driven bonds.The power law of the decay depends on the specificconfiguration.

The two-point correlation function corresponds to a quadrupole In 2d dimensions, decaying as for

The same is true at other densities to leading order in (order ).

Main Results

Page 8: Long-range  c orrelations in driven systems David  Mukamel

Density profile (with exclusion)

The density profile along the y axisin any other direction

Page 9: Long-range  c orrelations in driven systems David  Mukamel

• Time evolution of density:

𝛻2=𝜙 (𝑚+1 ,𝑛 )+𝜙 (𝑚−1 ,𝑛)+𝜙 (𝑚 ,𝑛+1 )+𝜙 (𝑚 ,𝑛−1 )−4𝜙 (𝑚 ,𝑛)

𝛻2𝜙 (𝑟 )=−𝜖𝜙( 0⃗)[𝛿𝑟 ,0⃗−𝛿�⃗� ,𝑒1]

Non-interacting particles

The steady state equation

particle density electrostatic potential of an electric dipole

Page 10: Long-range  c orrelations in driven systems David  Mukamel

𝛻2𝜙 (𝑟 )=−𝜖𝜙( 0⃗)[𝛿𝑟 ,0⃗−𝛿�⃗� ,𝑒1]

𝛻2𝐺 (𝑟 , �⃗�𝑜 )=−𝛿�⃗� ,𝑟 𝑜Green’s function

solution

Unlike electrostatic configuration here the strength ofthe dipole should be determined self consistently.

Page 11: Long-range  c orrelations in driven systems David  Mukamel

p\q 0 1 2

0 0

1

2

Green’s function of the discrete Laplace equation

𝐺 (𝑟 ,𝑟 𝑜 )≈− 12𝜋 ln ∨ �⃗�− 𝑟0∨¿

Page 12: Long-range  c orrelations in driven systems David  Mukamel

To find one uses the values

𝜙 (0⃗ )= 𝜌

1− 𝜖4

determining

Page 13: Long-range  c orrelations in driven systems David  Mukamel

𝜙 ( �⃗� )=𝜌− 𝜖𝜙 (0⃗ )2𝜋

�⃗�1𝑟𝑟 2

+𝑂 ( 1𝑟2

)

𝑗 (𝑟 )=𝜖𝜙 (0⃗ )2𝜋

1𝑟2

¿

density:

current:

at large

Page 14: Long-range  c orrelations in driven systems David  Mukamel

Multiple driven bonds

Using the Green’s function one can solve for , …by solving the set of linear equations for

Page 15: Long-range  c orrelations in driven systems David  Mukamel

The steady state equation:

Two oppositely directed driven bonds – quadrupole field

𝜙 ( �⃗� )=𝜌− 𝜖𝜙 (0⃗ )2𝜋 [ 1𝑟 2 −2( �⃗�1𝑟

𝑟2 )2]+𝑂 (

1𝑟 4

)

Page 16: Long-range  c orrelations in driven systems David  Mukamel

dimensions

Page 17: Long-range  c orrelations in driven systems David  Mukamel

𝜕𝑡 𝜙 (𝑟 , 𝑡 )=𝛻2𝜙 (𝑟 ,𝑡 )+𝜖 ⟨𝜏 ( 0⃗ ) {1−𝜏 ( �⃗�1 ) }⟩ [𝛿𝑟 ,0⃗−𝛿�⃗� ,𝑒1]

The model of local drive with exclusion

Here the steady state measure is not known however one candetermine the behavior of the density.

is the occupation variable

𝜙 ( �⃗� )=𝜌−𝜖 ⟨𝜏 (0⃗ ) {1−𝜏 (�⃗�1 )}⟩

2𝜋�⃗�1𝑟𝑟2

+𝑂(1𝑟 2

)

Page 18: Long-range  c orrelations in driven systems David  Mukamel

𝜙 ( �⃗� )=𝜌−𝜖 ⟨𝜏 (0⃗ ) {1−𝜏 (�⃗�1 )}⟩

2𝜋�⃗�1𝑟𝑟2

+𝑂(1𝑟 2

)

The density profile is that of the dipole potential with a dipolestrength which can only be computed numerically.

Page 19: Long-range  c orrelations in driven systems David  Mukamel

Simulation on a lattice with

For the interacting case the strength of the dipole was measured separately .

Simulation results

Page 20: Long-range  c orrelations in driven systems David  Mukamel

T. Bodineau, B. Derrida, J.L. Lebowitz, JSP, 140 648 (2010).

Two-point correlation function

- (r)

g( , )

In d=1 dimension, in the hydrodynamic limit

Page 21: Long-range  c orrelations in driven systems David  Mukamel

In higher dimensions local currents do not vanish for largeL and the correlation function does not vanish in this limit.

T. Sadhu, S. Majumdar, DM, in progress

Page 22: Long-range  c orrelations in driven systems David  Mukamel

Symmetry of the correlation function:

( Δ𝑟 +Δ𝑠 )𝐶 (𝑟 ,𝑠)=𝜎❑ (𝑟 , 𝑠)

𝐶𝜖 (r , s )=C−𝜖(−𝑟 ,−𝑠)

𝐶𝜖 ,𝜌 (r , s )=C−𝜖 ,1−𝜌(r , s )inversionparticle-hole

∫𝑑𝑑𝑟 𝐶 (𝑟 ,𝑠 )=0at 𝐶𝜖 (𝑟 ,𝑠 )=𝐶−𝜖 (𝑟 ,𝑠)

corresponds to an electrostatic potential in induced by

- (r)

Page 23: Long-range  c orrelations in driven systems David  Mukamel

Consequences of the symmetry:

The net charge =0

At is even in

Thus the charge cannot support a dipole and the leading contribution in multipole expansion is that of a quadrupole (in 2d dimensions).

𝐶 (𝑟 ,𝑠) 1 / (𝑟2+𝑠2 )𝑑

Page 24: Long-range  c orrelations in driven systems David  Mukamel

For one can expand in powers of

One finds: The leading contribution to is of order implying no dipolar contribution, with the correlation decaying as

𝐶 (𝑟 ,𝑠) 1 / (𝑟2+𝑠2 )𝑑

Page 25: Long-range  c orrelations in driven systems David  Mukamel

𝐶 (𝑟 ,𝑠 )=𝜖 𝛼1 (𝑟 ,𝑠 )+𝜖2𝛼2 (𝑟 ,𝑠 )+…

Since (no dipole) and the net charge is zero the leading contribution is quadrupolar

Page 26: Long-range  c orrelations in driven systems David  Mukamel

( Δ𝑟 +Δ𝑠 )𝐶 (𝑟 ,𝑠 )=𝜎 1 (𝑟 ,𝑠 )+𝜎2 (𝑟 ,𝑠)+𝜎3(𝑟 ,𝑠)

+

𝑄=𝑛 (0 ) (1−𝑛 (−𝑒1 ))+𝑛(−𝑒1)(1−𝑛 (0))

Page 27: Long-range  c orrelations in driven systems David  Mukamel

Summary

Local drive in dimensions results in:

Density profile corresponds to a dipole in d dimensions

Two-point correlation function corresponds to a quadrupolein 2d dimensions

𝐶 (𝑟 ,𝑠) 1 / (𝑟2+𝑠2 )𝑑

At density to all orders in At other densities to leading order

Page 28: Long-range  c orrelations in driven systems David  Mukamel

Example II: a two dimensional model with a driven line

T. Sadhu, Z. Shapira, DM PRL 109, 130601 (2012)

The effect of a drive on a fluctuating interface

Page 29: Long-range  c orrelations in driven systems David  Mukamel

Motivated by an experimental study of the effect of shear oncolloidal liquid-gas interface.

D. Derks, D. G. A. L. Aarts, D. Bonn, H. N. W. Lekkerkerker, A. Imhof,PRL 97, 038301 (2006).

T.H.R. Smith, O. Vasilyev, D.B. Abraham, A. Maciolek, M. Schmidt,PRL 101, 067203 (2008).

Page 30: Long-range  c orrelations in driven systems David  Mukamel

+

-

What is the effect of a driving line on an interface?

Page 31: Long-range  c orrelations in driven systems David  Mukamel

Local potential localizes the interface at any temperature Transfer matrix: 1d quantum particle in a local attractive potential, the wave-function is localized.

no localizing potential: with localizing potential:

+

-

In equilibrium- under local attractive potential

Page 32: Long-range  c orrelations in driven systems David  Mukamel

2 0 1 0 1 0 2 0 y

1 .0

0 .5

0 .5

1 .0

m y

Schematic magnetization profile

The magnetization profile is antisymmetric with respect to the zeroline with

Page 33: Long-range  c orrelations in driven systems David  Mukamel

+ - - + with rate

- + + - with rate

𝐻=− 𝐽 ∑¿ 𝑖 𝑗>¿ 𝑆𝑖𝑆 𝑗¿

¿

Consider now a driving line

Ising model with Kawasaki dynamics which is biased on the middle row

Page 34: Long-range  c orrelations in driven systems David  Mukamel

Main results

The interface width is finite (localized)

A spontaneous symmetry breaking takes placeby which the magnetization of the driven lineis non-zero and the magnetization profile is notsymmetric.

The fluctuation of the interface are not symmetric aroundthe driven line.

These results can be demonstrated analytically in certain limit.

Page 35: Long-range  c orrelations in driven systems David  Mukamel

Example of configurations in the two mesoscopic states for a 100X101 with fixed boundary at T=0.85Tc

Results of numerical simulations

Page 36: Long-range  c orrelations in driven systems David  Mukamel

20 10 10 20 y

1 .0

0 .5

0 .5

1 .0

m y

20 10 10 20 y

1 .0

0 .5

0 .5

1 .0

m y

Schematic magnetization profiles

2 0 1 0 1 0 2 0 y

1 . 0

0 . 5

0 . 5

1 . 0

m y

unlike the equilibrium antisymmetric profile

Page 37: Long-range  c orrelations in driven systems David  Mukamel

L=100 T=0.85Tc

Averaged magnetization profile in the two states

Page 38: Long-range  c orrelations in driven systems David  Mukamel
Page 39: Long-range  c orrelations in driven systems David  Mukamel

Time series of Magnetization of driven lane for a 100X101 lattice at T= 0.6Tc.

Page 40: Long-range  c orrelations in driven systems David  Mukamel

Switching time on a square LX(L+1) lattice with Fixed boundary at T=0.6Tc.

Page 41: Long-range  c orrelations in driven systems David  Mukamel

Typically one is interested in calculating the large deviation function of a magnetization profile

We show that in some limit a restricted large deviation function,that of the driven line magnetization, , can be computed

Analytical approach

In general one cannot calculate the steady state measure of this system.However in a certain limit, the steady state distribution (the large deviations function) of the magnetization of the driven line can be calculated.

𝑃 (𝑚 ( 𝑥 , 𝑦 )) 𝑒−𝐿2𝐹 (𝑚 ( 𝑥 , 𝑦 ) )

𝑃 (𝑚0 )=𝑒−𝐿𝑈 (𝑚0)

Page 42: Long-range  c orrelations in driven systems David  Mukamel

In this limit the probability distribution of is where the potential (large deviations function) can be computed.

Large driving field

Slow exchange rate between the driven line and the rest of the system

Low temperature

The following limit is considered

Page 43: Long-range  c orrelations in driven systems David  Mukamel

𝑈

The large deviations function

𝑃 (𝑚0 )=𝑒−𝐿𝑈 (𝑚0)

Page 44: Long-range  c orrelations in driven systems David  Mukamel

Slow exchange between the line and the rest of the system

In between exchange processes the systems iscomposed of 3 sub-systems evolving independently

+−⇆−+¿𝑟𝑤 (Δ𝐻 )

𝑟𝑤 (− Δ𝐻 )𝑟<𝑂(

1𝐿3 )

𝑤 (Δ𝐻 )=min (1 ,𝑒−𝛽Δ𝐻 )

Page 45: Long-range  c orrelations in driven systems David  Mukamel

Fast drive

the coupling within the lane can be ignored. As a resultthe spins on the driven lane become uncorrelated and they arerandomly distributed (TASEP)

The driven lane applies a boundary field on the two otherparts

Due to the slow exchange rate with the bulk, the two bulk sub-systems reach the equilibrium distribution of an Ising modelwith a boundary field

Low temperature limit

In this limit the steady state of the bulk sub systems canbe expanded in T and the exchange rate with the driven line canbe computed.

Page 46: Long-range  c orrelations in driven systems David  Mukamel

with rate

with rate

performs a random walk with a rate which depends on

𝑈 (𝑚𝑜 )=− ∑𝑘=0

𝑚𝑜 𝐿2

−1

ln𝑝 ( 2𝑘𝐿 )+ ∑𝑘=1

𝑚𝑜 𝐿2

𝑞( 2𝑘𝐿 )

𝑃 (𝑚𝑜=2𝑘𝐿 )=

𝑝 (0)⋯⋯𝑝 (2 (𝑘−1 )

𝐿 )

𝑞 ( 2𝐿)⋯⋯𝑞( 2𝑘𝐿 )

≡𝑒−𝑈(𝑚𝑜)

𝑝 (𝑚0)𝑞 (𝑚0)

Page 47: Long-range  c orrelations in driven systems David  Mukamel

+ - ++ + + + +

- - - - -

+ + + + +

- - - - -

Calculate p at low temperature

Page 48: Long-range  c orrelations in driven systems David  Mukamel

+ - ++ + + + +

- - - - -

+ + + + +

- - - - -

contribution to p

is the exchange rate between the driven line and the adjacent lines

Page 49: Long-range  c orrelations in driven systems David  Mukamel

The magnetization of the driven lane changes in steps of

Expression for rate of increase, p

𝑈 (𝑚 )=−∫0

𝑚

ln𝑝 (𝑘) 𝑑𝑘+∫0

𝑚

ln𝑞 (𝑘 ) 𝑑𝑘

𝑞 (𝑚𝑜 )=𝑝 (−𝑚𝑜)

Page 50: Long-range  c orrelations in driven systems David  Mukamel

This form of the large deviation function demonstrates the spontaneous symmetry breaking. It also yield theexponential flipping time at finite

𝑈

𝑃 (𝑚0 )=𝑒−𝐿𝑈 (𝑚0)

¿𝑚𝑜>1−𝑂 (𝑒− 4 𝛽 𝐽)

Page 51: Long-range  c orrelations in driven systems David  Mukamel

Summary

Simple examples of the effect of long range correlations in drivenmodels have been presented.

A limit of slow exchange rate is discussed which enablesthe evaluation of some large deviation functions far fromequilibrium.