Transcript Document

Yang-Mills ground-state wavefunctional
in 2+1 dimensions
(in collaboration with Jeff Greensite)
J. Greensite, ŠO, Phys. Rev. D 77 (2008) 065003, arXiv:0707.2860 [hep-lat]
7.5.2008
Seminar, Institut f. Physik, Karl-Franzens-Universität, Graz
1
Motivation



“QCD field theory with six flavors of quarks with three colors, each represented
by a Dirac spinor of four components, and with eight four-vector gluons, is a
quantum theory of amplitudes for configurations each of which is 104 numbers at
each point in space and time. To visualize all this qualitatively is too difficult. The
thing to do is to take some qualitative feature to try to explain, and then to
simplify the real situation as much as possible by replacing it by a model which is
likely to have the same qualitative feature for analogous physical reasons.
The feature we try to understand is confinement of quarks.
We simplify the model in a number of ways.



7.5.2008
First, we change from three to two colors as the number of colors does not seem to be
essential.
Next we suppose there are no quarks. Our problem of the confinement of quarks when
there are no dynamic quarks can be converted, as Wilson has argued, to a question of the
expectation of a loop integral. Or again even with no quarks, there is a confinement problem,
namely the confinement of gluons.
The next simplification may be more serious. We go from the 3+1 dimensions of the real
world to 2+1. There is no good reason to think understanding what goes on in 2+1 can
immediately be carried by analogy to 3+1, nor even that the two cases behave similarly at
all. There is a serious risk that in working in 2+1 dimensions you are wasting your time, or
even that you are getting false impressions of how things work in 3+1. Nevertheless, the
ease of visualization is so much greater that I think it worth the risk. So, unfortunately, we
describe the situation in 2+1 dimensions, and we shall have to leave it to future work to see
what can be carried over to 3+1.”
Seminar, Institut f. Physik, Karl-Franzens-Universität, Graz
2
7.5.2008
Seminar, Institut f. Physik, Karl-Franzens-Universität, Graz
3
Introduction

Confinement is the property of the vacuum of quantized non-abelian
gauge theories. In the hamiltonian formulation in D=d+1 dimensions and
temporal gauge:
7.5.2008
Seminar, Institut f. Physik, Karl-Franzens-Universität, Graz
4

At large distance scales one expects:



Halpern (1979), Greensite

Greensite, Iwasaki
Kawamura, Maeda, Sakamoto

Karabali, Kim, Nair
(1979)
(1989)
(1997)
(1998)
Property of dimensional reduction: Computation of a spacelike loop in
d+1 dimensions reduces to the calculation of a Wilson loop in Yang-Mills
theory in d Euclidean dimensions.
7.5.2008
Seminar, Institut f. Physik, Karl-Franzens-Universität, Graz
5
Suggestion for an approximate vacuum wavefunctional
7.5.2008
Seminar, Institut f. Physik, Karl-Franzens-Universität, Graz
8
Warm-up example: Abelian ED
7.5.2008
Seminar, Institut f. Physik, Karl-Franzens-Universität, Graz
9
7.5.2008
Seminar, Institut f. Physik, Karl-Franzens-Universität, Graz
10
Free-field limit (g!0)
7.5.2008
Seminar, Institut f. Physik, Karl-Franzens-Universität, Graz
11
Zero-mode, strong-field limit (D=2+1)


D. Diakonov (private communication to JG)
Let’s assume we keep only the zero-mode of the A-field, i.e. fields constant in space,
varying in time. The lagrangian is
and the hamiltonian operator

Natural choice - 1/V expansion:
7.5.2008
Seminar, Institut f. Physik, Karl-Franzens-Universität, Graz
12

Keeping the leading term in V only:

The equation is solved by:
since
7.5.2008
Seminar, Institut f. Physik, Karl-Franzens-Universität, Graz
13

Now the proposed vacuum state coincides with this solution in the strong-field limit,
assuming

The covariant laplacian is then

Let’s choose color axes so that both color vectors lie in, say, (12)-plane:
7.5.2008
Seminar, Institut f. Physik, Karl-Franzens-Universität, Graz
14

The eigenvalues of M are obtained from

Our wavefunctional becomes

In the strong-field limit
7.5.2008
Seminar, Institut f. Physik, Karl-Franzens-Universität, Graz
15
D=3+1
7.5.2008
Seminar, Institut f. Physik, Karl-Franzens-Universität, Graz
16
7.5.2008
Seminar, Institut f. Physik, Karl-Franzens-Universität, Graz
17
Dimensional reduction and confinement

What about confinement with such a vacuum state?
Define “slow” and “fast” components using a mode-number cutoff:

Then:

7.5.2008
Seminar, Institut f. Physik, Karl-Franzens-Universität, Graz
18

Effectively for “slow” components
we then get the probability distribution of a 2D YM theory and can compute the
string tension analytically (in lattice units):


Non-zero value of m implies non-zero string tension  and confinement!
Let’s revert the logic: to get  with the right scaling behavior ~ 1/2, we need to
choose
7.5.2008
Seminar, Institut f. Physik, Karl-Franzens-Universität, Graz
19
Why m02 = -0 + m2 ?

7.5.2008
Seminar, Institut f. Physik, Karl-Franzens-Universität, Graz
Samuel (1997)
20
Non-zero m is energetically preferred


Take m as a variational parameter and minimize <H > with respect to m:
Assuming the variation of K with A in the neighborhood of thermalized configurations
is small, and neglecting therefore functional derivatives of K w.r.t. A one gets:
7.5.2008
Seminar, Institut f. Physik, Karl-Franzens-Universität, Graz
21

Abelian free-field limit: minimum at m2 = 0 → 0.
7.5.2008
Seminar, Institut f. Physik, Karl-Franzens-Universität, Graz
22


Non-abelian case: Minimum at non-zero m2 (~ 0.3), though a
higher value (~ 0.5) would be required to get the right string
tension.
Could (and should) be improved!
7.5.2008
Seminar, Institut f. Physik, Karl-Franzens-Universität, Graz
23
Calculation of the mass gap

To extract the mass gap, one would like to compute
in the probability distribution:


Looks hopeless, K[A] is highly non-local, not even known for arbitrary fields.
But if - after choosing a gauge - K[A] does not vary a lot among
thermalized configurations … then something can be done.

7.5.2008
Seminar, Institut f. Physik, Karl-Franzens-Universität, Graz
Numerical simulation
24
Numerical simulation of |0|2

Define:

Hypothesis:

Iterative procedure:
7.5.2008
Seminar, Institut f. Physik, Karl-Franzens-Universität, Graz
25

Practical implementation:
choose e.g. axial A1=0 gauge, change variables from A2 to B. Then
1.
2.
3.
4.


given A2, set A2’=A2,
the probability P [A;K[A’]] is gaussian in B, diagonalize K[A’] and generate new Bfield (set of Bs) stochastically;
from B, calculate A2 in axial gauge, and compute everything of interest;
go back to the first step, repeat as many times as necessary.
All this is done on a lattice.
Of interest:



Spiral gauge
Eigenspectrum of the adjoint covariant laplacian.
Connected field-strength correlator, to get the mass gap:
For comparison the same computed on 2D slices of 3D lattices generated by
Monte Carlo.
7.5.2008
Seminar, Institut f. Physik, Karl-Franzens-Universität, Graz
26
Eigenspectrum of the adjoint covariant laplacian
7.5.2008
Seminar, Institut f. Physik, Karl-Franzens-Universität, Graz
27
Mass gap
7.5.2008
Seminar, Institut f. Physik, Karl-Franzens-Universität, Graz
28
7.5.2008
Seminar, Institut f. Physik, Karl-Franzens-Universität, Graz
29
Summary (of apparent pros)

Our simple approximate form of the confining YM vacuum
wavefunctional in 2+1 dimensions has the following properties:





7.5.2008
It is a solution of the YM Schrödinger equation in the weak-coupling
limit …
… and also in the zero-mode, strong-field limit.
Dimensional reduction works: There is confinement (non-zero string
tension) if the free mass parameter m is larger than 0.
m > 0 seems energetically preferred.
If the free parameter m is adjusted to give the correct string tension at
the given coupling, then the correct value of the mass gap is also
obtained.
Seminar, Institut f. Physik, Karl-Franzens-Universität, Graz
30
Open questions (or contras?)



Can one improve (systematically) our vacuum wavefunctional Ansatz?
Can one make a more reliable variational estimate of m?
Comparison to other proposals?





Karabali, Kim, Nair (1998)
Leigh, Minic, Yelnikov (2007)
What about N-ality?
Knowing the (approximate) ground state, can one construct an
(approximate) flux-tube state, estimate its energy as a function of
separation, and get the right value of the string tension?
How to go to 3+1 dimensions?


Much more challenging (Bianchi identity, numerical treatment very CPU time
consuming).
The zero-mode, strong-field limit argument valid (in certain approximation) also
in D=3+1.
Comparison to KKN
N-ality
Flux-tube state
7.5.2008
Seminar, Institut f. Physik, Karl-Franzens-Universität, Graz
31
Elements of the KKN approach

Matrix parametrisation:

Jakobian of the transformation leads to appearance of a WZW-like term in the action.
7.5.2008
Seminar, Institut f. Physik, Karl-Franzens-Universität, Graz
32
Comparison to KKN

Wavefunctional expressed in terms of still another variable:

It’s argued that the part bilinear in field variables has the form:

The KKN string tension following from the above differs from string tensions obtained by
standard MC methods, and the disagreement worsens with increasing .
7.5.2008
Seminar, Institut f. Physik, Karl-Franzens-Universität, Graz
33
N-ality



Dimensional reduction form at large distances implies area law
for large Wilson loops, but also Casimir scaling of higherrepresentation Wilson loops.
How does Casimir scaling turn into N-ality dependence, how
does color screening enter the game?
A possibility: Necessity to introduce additional term(s), e.g. a gaugeinvariant mass term

Cornwall (2007)
… but color screening may be contained!

Strong-coupling:

7.5.2008
Seminar, Institut f. Physik, Karl-Franzens-Universität, Graz
Greensite (1980)
34

7.5.2008
Seminar, Institut f. Physik, Karl-Franzens-Universität, Graz
Guo, Chen, Li (1994)
35
7.5.2008
Seminar, Institut f. Physik, Karl-Franzens-Universität, Graz
36
Flux-tube state


A variational trial state:
The energy of such a state for a given quark-antiquark separation can be computed
from:

On a lattice:

Work in progress.
7.5.2008
Seminar, Institut f. Physik, Karl-Franzens-Universität, Graz
37
Epilogue/Apologies
“It is normal for the true physicist not to worry too much about mathematical
rigor. And why? Because one will have a test at the end of the day which is
the confrontation with experiment. This does not mean that sloppiness is
admissible: an experimentalist once told me that they check their
computations ten times more than the theoreticians! However it’s normal
not to be too formalist. This goes with a certain attitude of physicists
towards mathematics: loosely speaking, they treat mathematics as a kind of
prostitute. They use it in an absolutely free and shameless manner, taking
any subject or part of a subject, without having the attitude of the
mathematician who will only use something after some real understanding.”

7.5.2008
Alain Connes in an interview with C. Goldstein and G. Skandalis, EMS Newsletter, March 2008.
Seminar, Institut f. Physik, Karl-Franzens-Universität, Graz
38