Gap Equation and Schwinger Dyson Equation: Pok Man Lo

Gap Equation and Schwinger Dyson Equation
Pok Man Lo
University of Pittsburgh
High Energy Journal Club, 9-11-2008

personal information
institution: University of Pittsburgh

advisor: Eric Swanson
outline
1 Gap equation and the study of non-perturbative physics
2 Tools to study the Gap Equation

Diagrammatics
Hartree-Fock approximation
Beyond Hartree-Fock
side track: the case for QED
Schwinger Dyson Equation
3 Numerical Results
4 hidden track: finite temperature gap equation

Survey of basic formalism of Finite Temperature of Field
Theory
Finite Temperature Gap Equation
outline

Diagrammatics
Beyond Hartree-Fock
3 Numerical Results

Theory
dynamically generated mass
The mass of the constituents of the proton is expected to be

1
3 Mproton

1
3 Mproton
It is believed that the (relatively large) constituent quark mass is

due to the fact that chiral symmetry is spontaneously broken

1
3 Mproton

current quark mass is small

the mass of the quark is dynamically generated

1
3 Mproton

current quark mass is small

the mass of the quark is dynamically generated
Mass is generated through interaction, and can generally be M (k)

gap equation
Gap equation dictates how the interaction give rise to the

dynamical mass
...that is, how the interaction dress the particle in question

gap equation

dynamical mass
study of spontaneous symmetry breaking mandates the use of

non-perturbative techniques
gap equation

dynamical mass
study of spontaneous symmetry breaking mandates the use of

non-perturbative techniques
non-perturbative physics
exact solution: dream of a theorist...?
Newtonian physics: 3-body problem was insoluble

QED: 2-body and 1-body problem was insoluble

now: 0-body, namely, the vacuum is insoluble

now: 0-body, namely, the vacuum is insoluble
No body is too many!!
we are tranined to make sensible approximation

illustration of the failure of perturbation
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
0 2 4 6 8 10
...an innocent looking function

1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
0 2 4 6 8 10

using Taylor expansion, we write
f (x) = A0 + A1 x + A2 x2 + ...
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
0 2 4 6 8 10

f (x) = A0 + A1 x + A2 x2 + ...
but what you will find is ...

1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
0 2 4 6 8 10

f (x) = A0 + A1 x + A2 x2 + ...
but what you will find is ...

A0 , A1 , A2 , ... are all strightly 0!
in view of the differential equation...

f (x) + x3 f 0 = 0
f (x) + x3 f 0 = 0
1
f (x) = exp− x2
f (x) + x3 f 0 = 0
1
f (x) = exp− x2
any other method will work, other than perturbation!

different philosophy
Perturbative Vs Non-perturbative:
perturbative method: sum all diagrams up to certain order in α

non-perturbative method: sum a certain class of diagram to all

order,

order, ...or more generally, a truncaton

in some sense, non-perturbative method includes the whole

perturbation method as the latter is just a classification of diagram
by the coupling constant α


perturbation is not the only way to make approximation!



perturbation is not the only way to make approximation!
QFT contains more than just the S-matrix and perturbation!

side track: bound state problem
another important area of non-perturbative physics is the

formation of bound state

easy to motivate the necessity to include all orders in coupling

constant

easy to motivate the necessity to include all orders in coupling

constant
classical
q mechanics example: mass attached to a spring:
k
A sin( m t)
outline

Diagrammatics
Beyond Hartree-Fock
3 Numerical Results

Theory
outline

Diagrammatics
Beyond Hartree-Fock
3 Numerical Results

Theory
method of partial sum
summing a particular class of diagrams to all order

method of partial sum
summing a particular class of diagrams to all order
to be explicit, let’s consider the Hamiltonian:
Z Z
H= 3
d xψ~x† γ 0 [−i~γ · ∇ + m]ψ~x − G d3 xd3 yV~x~y ψ~x† T a ψ~x ψ~y† T a ψ~y
this corresonds to the case where the quark exchange an

instantaneous gluon locally
we investigate the diagrams for constructing a propagator

diagrammatics of propagator
in general, to construct the propagator, we need to sum...

in general, we cannot sum them all!

but let’s focus on the following class of diagram:

but let’s focus on the following class of diagram:

the above procedure should be compared with the summing of

geometric series:
1
= 1 + x + x2 + x3 + ...
1−x
inspired by the series, we know we can do better:

inspired by the series, we know we can do better:

also, to take care of

the above corresponds to the Hartree-Fock approximation of the

the inclusion of ring diagram
the above is pretty standard, it would be interesting to see how one

can go beyond Hartree-Fock approximation

for example, we can include the ring diagram...
but where do this come from?


for example, we can include the ring diagram...
but where do this come from?

in fact the full Schwinger Dyson Equation for the above case reads
like:
we see the ring structure comes out if we expand...

we see the ring structure comes out if we expand...

the Hartree Fock approximation is an approximation in the sense

that it effectively truncate the Schwinger Dyson Equation,
namely, by neglecting the connected four point function
the Hartree Fock approximation is an approximation in the sense

that it effectively truncate the Schwinger Dyson Equation,
namely, by neglecting the connected four point function
in general, the four point function itself depends on some higher

function, hence there will be an infinite tower of Schwinger Dyson
Equation
in perturbation, we approximate by expanding up to a certain order

in α
in Schwinger Dyson approach, we approximate by truncation
note that the sum for each truncaton is up to all order in α, hence
the result is non-perturbative
side track: the case for qed
the Schwinger Dyson Equations for QED are:

the ring diagram in QED
in this case, the ring obviously comes form the expansion of the
dressed photon propagator in the fermion Schwinger Dyson
equation
it is interesting to compare the case we have above: the four point

function do not appear explicitly in the fermion propagator, in
QED, there is no need for that, the dressed photon contain such a
term
of course, at some point, one will need to truncate these towers of

some comments
1.) the diagrammatic approach is intuitive in the sense that you

can describe the various contribution using physically meaning
Feynman diagrams
2.) but it’s hard to decide which terms to include in the expansion
one needs a more systematic method to generate these equations:
Schwinger Dyson Equation directly from the generating functional

outline

Diagrammatics
Beyond Hartree-Fock
3 Numerical Results

Theory
Schwinger Dyson equation
represents an exact relations among n-point functions

effectively derives an expression for full propagator

by expanding Vint in terms of functional derivatives, we can obtain

the complete perturbative expansion of the theoretical quantity
by expanding Vint in terms of functional derivatives, we can obtain

the complete perturbative expansion of the theoretical quantity
basic idea: Z
d
dx f (x) → 0
dx
R
DψDψDGei(S+ ηψ+ψη+...)
R
Z= R
DψDψDGei(S)
R
R
Z= R
DψDψDGei(S)
using the fact
Z
δ i(S+R ηψ+ψη+...)
DψDψDG e =0
δψ
we get
δS δ δ δ
{ [−i , i , −i ] + η}Z = 0
δψ δη δη δjµ
R
R
Z= R
DψDψDGei(S)
using the fact
Z
δ i(S+R ηψ+ψη+...)
DψDψDG e =0
δψ
we get
δS δ δ δ
{ [−i , i , −i ] + η}Z = 0
δψ δη δη δjµ
for connected diagram, it is useful to define W by Z = eiW
as an illustration, we derive the Gap Equation for our old

Hamiltonian:
Z Z
H= 3
d xψ~x† γ 0 [−i~γ · ∇ + m]ψ~x + G d3 xd3 yV~x~y ψ~x† T a ψ~x ψ~y† T a ψ~y
δ2W
= (iγ · ∂x − m)−1 δ +
δηδη
2
W 2W
V S0 γ 0 T δδηδη T r[γ 0 T δδηδη
R
G ]
2 W 0 δ W 2
G V S0 γ 0 T δδηδη
R
γ T δηδη
δ4 W
G V S0 γ 0 γ 0 T T δηδηδηδη
R
corresponding to the diagrams ...

gap equation
for Hartree-Fock approximation, we neglect the 4 point function

and get:
gap equation
for Hartree-Fock approximation, we neglect the 4 point function

and get:
δ2W δ2W 0 a δ2W

Z
(iγ · ∂x − m) + 2G d4 zV γ 0 T a i γ T = δxy
δηx δηy δηx δηz δηz δηy
gap equation for Hartree Fock approximation
gap equation
d3 k 0 M (k~0 ) |k~0 |M (~k)

Z
T r[T T ]
M (~k) = m − G V ~0 ~ [ − k̂ 0 · k̂ ]
Nc (2π)3 k −k Ek~0 E ~0 |~k|
k
it dictates how the mass is dynamically generated

outline

Diagrammatics
Beyond Hartree-Fock
3 Numerical Results

Theory
numerical results for gap equation
contact potential: V (~x) = δ (3) (~x)
linear potential: V (~x) = br

contact potential
Dynamical Mass Generation

4
|G| > G_critical
|G| < G_critical
3.5
2.5
M
1.5
0.5
0
0 0.5 1 1.5 2 2.5 3 3.5 4
current mass m
linear potential
gap equation for linear case

0.35
0.3
0.25
0.2
M(k)
0.15
0.1
0.05
0
1e-05 0.0001 0.001 0.01 0.1 1 10
log k
outline

Diagrammatics
Beyond Hartree-Fock
3 Numerical Results

Theory
motivation
FTFT is needed to study the physics of QGP, deconfinement and

chiral restoration
when calculating observables in QFT, we only calculate the

vacuum expectation value
at finite temperature, excited states start to contribute, the

interesting quantity should be the thermal average of an
observable
we expect nE~k to enter QFT
partition function dictates the equilibrium Finite Temperature QFT

outline

Diagrammatics
Beyond Hartree-Fock
3 Numerical Results

Theory
basics of FTFT
the partition function:

Z
−βH
Z = T r[e ]= Dφ < φ|e−βH |φ >
basics of FTFT

Z
−βH
Z = T r[e ]= Dφ < φ|e−βH |φ >
Z R
−βH
Z = T r[e ]= DψDψe− ψψ
< −ψ|e−βH |ψ >
basics of FTFT

Z
−βH
Z = T r[e ]= Dφ < φ|e−βH |φ >
Z R
−βH
< −ψ|e−βH |ψ >
observables are given by
T r[e−βH O]
hhOii =
T r[e−βH ]
basics of FTFT
Z
−βH
Z = T r[e ]= Dφ < φ|e−βH |φ >
Z R
−βH
< −ψ|e−βH |ψ >
the form of partition function motivates the use of imaginary time

τ = it
a corresponding path integral representation of the partition

function can be obtained, with periodic or antiperiodic
boundary condition
the origin of this is of course due to the statistics of the particle

basics of FTFT
the statistics of the particles, together with the boundary condition

motivates the use of Matsubara Green’s function: G[ωn~k]
basics of FTFT

2nπ (2n+1)π
with ωn = β or β
basics of FTFT

2nπ (2n+1)π
with ωn = β or β
Z
1
dk 0 → Σω
β n
similarly, one can derive the Schwinger Dyson Equation
techniques in performing the Matsubara frequency sum

→ which in turns give you the thermodynamics...

→ which in turns give you the thermodynamics...
in the non-interacting limit, it reduces to statistical mechanics...

outline

Diagrammatics
Beyond Hartree-Fock
3 Numerical Results

Theory
gap equation in finite temperature
Generalized Gap Equation at Finite Temperature
d3 k 0 M (k~0 ) |k~0 |M (~k)

Z
T r[T T ]
M (~k) = m − G 3
Vk~0 −~k [ − k̂ 0 · k̂ ]
Nc (2π) Ek~0 E ~0 |~k|
k
(1 − 2nE ~0 )
k
1
with nE~k = βE~
e k +1
numerical result for finite temperature gap equation
again, results for contact case and linear potential

contact case
finite temperature contact mass

1
G_eff = 3.6
0.9
0.8
0.7
dynamical mass
0.6
0.5
0.4
0.3
0.2
0.1
0
0 2 4 6 8 10
beta
linear potential case
M(k) for various Temperature

0.35
beta = 500
beta = 100
beta = 30
0.3 beta = 7
beta = 6.85
beta = 6.35
0.25
0.2
0.15
0.1
0.05
0
1e-05 0.0001 0.001 0.01 0.1 1 10
linear potential case
M(0) Vs Temperature
350
300
250
200
M(0) (MeV)
150
100
50
0
0 20 40 60 80 100 120 140 160
Temperature (MeV)
thank you

Gap Equation and Schwinger Dyson Equation: Pok Man Lo

Uploaded by

Copyright:

Available Formats

Gap Equation and Schwinger Dyson Equation: Pok Man Lo

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Gap Equation and Schwinger Dyson Equation: Pok Man Lo

Uploaded by

Copyright:

Available Formats

Gap Equation and Schwinger Dyson Equation

High Energy Journal Club, 9-11-2008

institution: University of Pittsburgh

1 Gap equation and the study of non-perturbative physics

2 Tools to study the Gap Equation

4 hidden track: finite temperature gap equation

1 Gap equation and the study of non-perturbative physics

2 Tools to study the Gap Equation

4 hidden track: finite temperature gap equation

The mass of the constituents of the proton is expected to be

The mass of the constituents of the proton is expected to be

It is believed that the (relatively large) constituent quark mass is

The mass of the constituents of the proton is expected to be

It is believed that the (relatively large) constituent quark mass is

current quark mass is small

The mass of the constituents of the proton is expected to be

It is believed that the (relatively large) constituent quark mass is

current quark mass is small

Mass is generated through interaction, and can generally be M (k)

Gap equation dictates how the interaction give rise to the

...that is, how the interaction dress the particle in question

Gap equation dictates how the interaction give rise to the

...that is, how the interaction dress the particle in question

study of spontaneous symmetry breaking mandates the use of

Gap equation dictates how the interaction give rise to the

...that is, how the interaction dress the particle in question

study of spontaneous symmetry breaking mandates the use of

exact solution: dream of a theorist...?

Newtonian physics: 3-body problem was insoluble

exact solution: dream of a theorist...?

Newtonian physics: 3-body problem was insoluble

QED: 2-body and 1-body problem was insoluble

exact solution: dream of a theorist...?

Newtonian physics: 3-body problem was insoluble

QED: 2-body and 1-body problem was insoluble

now: 0-body, namely, the vacuum is insoluble

exact solution: dream of a theorist...?

Newtonian physics: 3-body problem was insoluble

QED: 2-body and 1-body problem was insoluble

now: 0-body, namely, the vacuum is insoluble

No body is too many!!

we are tranined to make sensible approximation

...an innocent looking function

...an innocent looking function

...an innocent looking function

but what you will find is ...

...an innocent looking function

but what you will find is ...

in view of the differential equation...

in view of the differential equation...

in view of the differential equation...

in view of the differential equation...

any other method will work, other than perturbation!

perturbative method: sum all diagrams up to certain order in α

perturbative method: sum all diagrams up to certain order in α

non-perturbative method: sum a certain class of diagram to all

perturbative method: sum all diagrams up to certain order in α

non-perturbative method: sum a certain class of diagram to all