Infrared gluons in the stochastic quantization approach Lattice20081 Contents 1.Introduction...

Infrared gluons in the stochastic

quantization approach

Lattice2008 1

Contents1. Introduction2. Method: Stochastic gauge fixing3. Gluon propagators4. Numerical results5. Summary

Takuya Saito （ Kochi), Nakagawa Yoshiyuki 　(Osaka),

Nakamura Atsushi 　 (Hiroshima), Toki Hiroshi 　 (Osaka)

Introduction(1)

Lattice2008 2

Confinement

Quarks and gluons are basic quantities of QCD. In ultraviolet region, the perturbative QCD works well but in the confining region, some non-perturbative modes dominates hadron physics.

Infrared physics of QCD: Confinement, Chiral symmetry breaking; these non-perturbative phenomena are deeply related to infrared singularities of QCD.

Infrared (transverse) gluon propagators

If confinement exists, one can expects that a transverse gluon propagator has an infinite mass, and will vanish in the IR limit.

On the other hands, the ghost propagator diverges in the IR limit.

We can find many lattice studies for these in many references; however, there are no distinctive signals, particularly for gluons.

Introduction(2)

Lattice2008 3

Numerical difficulty :

Finite volume size effect; the infrared physics requires large lattices.

Gauge fixing computation on the large lattices is very hard, time-consuming simulations if we use the iterative gauge fixing.

Conceptual difficulty:

Lattice configuration can not be gauge-fixed uniquely due to Gribov ambiguity.

We expect that the Gribov copy configuration will fade the infrared physics we are interested in.

Gribov copy problem is not fully understood now.

=== Some difficulties for lattice calculations for gluons ===

Introduction(3)

Lattice2008 4

Calculations of the gluon propagator in the stochastic quantization with the Coulomb gauge

This method has some advantage:

We do not use the iterative gauge fixing method.

Gauge configurations go to the Gribov region automatically.

Gauge parameter is easy to change.

Measure of the transverse gluon propagators

Transverse gluon propagator is a physical quantity.

We expect that the gluon propagator in the infrared limit will be suppressed with an infinite effective masses. This means gluons are confining.

=== Aim in this study ===

Method(1)

Lattice2008 5

=== Stochastic quantization with the gauge fixing ===

Stochastic Gauge fixing ：D.Zwanziger,Nucl.Phys.B192(1981)

ababa

a

AADA

S

d

dA

)(1

Langevin equation for the gauge theory with the gauge fixing ( a la Zwanziger)

: Virtual time for the hypothetical stochastic process

::

Gauge parameter

Gaussian white noise

0),( x )()(2),(),( '''' xxxx

Method(2)

Lattice2008 6

=== Stochastic quantization on the lattice ===

),ˆ(),()exp(),(),( xxUtifxxU aa

a

aa

A

Sf

exp(iat a / )

Lattice generalization of stochastic gauge fixing ： A.Nakamura and M. Mizutani, Vistas in Astronomy (Pergamon Press,1993), vol.37 p.305. , M. Mizutani and A.Nakamura, Nucl. Phys. B (Proc.Suppl.)34(1994),253.

Driving force

Gauge rotation

Method(3)

Lattice2008 7

=== Conceptual reason for using SGF ===

Conceptual reason

Gauge copy problem

Gauge configurations not fixed completely on the non-perturbative lattice calculation

Gauge fixing term of SGF

1. It makes gauge configurations go to the Gribov region.

2. This term works as an attractive driving force.

3. More effective approach

Gribov region0..,0)( PFxA

)(xA

Method(4)

Lattice2008 8

=== Practical reason for using SGF ===

Practical reason

For a gauge fixing, we don’t use any iterative methods and so there is no critical slowing down of this algorithm. It is a great advantage for large lattice simulation with gauge fixing.

Changing a gauge parameter is easier than the iterative methods.

AMonte Carlo Steps

～ Monte Carlo Quantization ～

Gauge rotations

～ Stochastic Quantization ～ A

Langevin steps

Coulomb gauge QCD

Lattice2008 9

=== basic issues ===

Hamiltonian of Coulomb gauge QCD

A transverse part makes a physics gluon field.

A source term makes a color-Coulomb instantaneous (confining ) potential among quarks, causing by a singular eigenvalue of F.P.

No negative norm : A physical interpretation is very clear.

3 2 2 3 31 1( ) ( , ) ( )

2 2i iH d x E B d xd y x D x y y

Gluon propagators(1)

Lattice2008 10

=== General form in the perturbative region ===

General form of gluon propagators

For free case, we have

If adding an anomalous dimension, we have

)()()( 22

22

qDq

qqqD

q

qqgqD Lab

22

0

1)(

qqD

22 1

~)(q

qD


Lattice2008 11

=== Assumptions in the non-perturbative region ===

Mandlestam hypothesise ( if the confining potential is linear )

Gluon propagator with an effective mass

Gluon propagator vanishes in the IR limit

0 ,1

~)( 24

2 qq

qD

222 1

~)(mq

qD

44

22 ~)(

mq

qqD

) Zwanziger todue ( 0)0(lim 2

qDN


Lattice2008 12

=== Gluon propagators on the lattice ===

Gauge field on the lattice in this calculation

Fourier transform

Gluon correlators ( we’ll measure )

aa xUxUi

xA )()(Tr2

1)(

dxxAepA axipa )()2(

1)(

4

)()()( pApAqD baab

Numerical parameters

Lattice2008 13

β α Δτ L Thermal. Confs.

6.0 1.0 0.01 12 10k 100 (100)

6.0 1.0 0.01 18 10k 100 (100)

6.0 1.0 0.01 24 10k 100 (100)

6.0 1.0 0.01 32 10k 100 (100)

6.0 1.0 0.01 48 4k 40 (10)

6.0 1.0 0.01 64 4k 40(10)

5.7 1.0 0.01 12 10k 100(100)

5.7 1.0 0.01 18 10k 100(100)

5.7 1.0 0.01 24 10k 100(100)

5.7 1.0 0.01 32 10k 100(100)

Quenched Wilson action simulations with hypercubic lattices

Simulation parameters

Numerical result (1)

Lattice2008 14

=== Volume dependence at beta=6.0 ===

Flat in the IR region, but not suppressed.

Not diverge in the IR region.

All the data are on the same line.

For largest volume (64)4=(6.4fm) 4


Lattice2008 15

=== Volume dependence at beta=5.7 ===

Flat in the IR region, but not suppressed.

Not diverge in the IR region.

All the data are on the same line.

For largest volume

(32)4=(5.4fm) 4


Lattice2008 16

=== α-parameter dependence at beta=5.7 ===

In the UV region, small variation with α

In the IR region, large change with α?

For smallest α, we got better result.

Summary

Lattice2008 17

We try to calculate gluon propagators in the confinement region in the stochastic gauge fixing method with the Coulomb gauge.

For this new calculation, we need more information and arguments.

We find sign of an infrared suppression of gluon propagators.

Larger physical volume ?

We find that the infrared gluons are strongly affected by variation of alpha-gauge parameter.

Why ?

We need investigation of the lowest eigenvalue of FP operator, the relation of the sharp gauge, etc.

Method(5)

Lattice2008 18

=== Disadvantage for using SGF ===

Langevin step dependence

Lattice2008 19

Gauge fixing term

Gauge fixing term bab AAD

d

)(

α-paramter small, dτ small more computation time

Lattice2008 20

Numerical results of Gluon propagators

Volume dependence , beta dependence , alpha parameter dependence

Lattice2008 21

Numerical results (1)

Lattice2008 22

Numerical results (1)

Lattice2008 23

JPS２００６S 24

クーロンゲージ QCD　クーロンゲージ QCD におけるハミルトニアン

　クーロンゲージ QCD におけるファデーフポボフ

　グルーオン伝播関数の時間成分

3 2 2 3 31 1( ) ( , ) ( )

2 2i iH d x E B d xd y x D x y y

3 21 1( , ) ( )

( , ) ( , )zD x y d zM x y M x y

2( )M gA

20 0( ) ( ) ( ) ( )g A x A y V x y P x y

24 4( ) ( , ) ( )V x y g D x y x y

瞬間力部分

遅延部分

Infrared gluons in the stochastic quantization approach Lattice20081 Contents 1.Introduction...

Documents

Transcript of Infrared gluons in the stochastic quantization approach Lattice20081 Contents 1.Introduction...