1: \documentclass[aps,article,groupedaddress]{revtex4}
2: %\documentclass[aps,twocolumn,groupedaddress]{revtex4}
3: \usepackage{graphicx}
4: \begin{document}
5: \bibliographystyle{apsrev}
6:
7: \newcommand{\be}{\begin{eqnarray}}
8: \newcommand{\ee}{\end{eqnarray}}
9: \newcommand{\ds}[1]{
10: #1{\hskip-2.0mm}/
11: }
12: \newcommand{\lb}[1]{
13: \label{#1}
14: }
15:
16:
17: \title{Systematic Study of the Single Instanton Approximation in QCD}
18:
19: \author{P. Faccioli, E. V. Shuryak}
20: \email{faccioli@tonic.physics.sunysb.edu,shuryak@dau.physics.sunysb.edu}
21: \affiliation{Department of Physics and Astronomy\
22: State University of New York at Stony Brook/
23: Stony Brook, New York 11794, USA}
24: \date{\today}
25:
26: %---------------------------------------------------------------------%
27:
28: \begin{abstract}
29:
30: Single-instanton
31: approximation (SIA) is often used to evaluate analytically
32: instanton contributions
33: euclidean correlation function in QCD at small distances.
34: We discuss how this approximation can be
35: consistently derived from the theory of
36: instanton ensemble and give precise definitions to a number of
37: different ``quark effective masses'', generalizing the parameter
38: $m^*$, which was introduced long ago to account for the
39: collective contribution of the whole ensemble.
40: We test numerically the range of applicability of the SIA
41: for different quantities. Furthermore,
42: we determine all the effective masses (for random and interacting
43: instanton liquid models) as well as from phenomenology,
44: and discuss to what extent those are universal.lan
45: \end{abstract}
46: \pacs{xxx}
47: %\keywords{Instantons, hadronic correlation functions}
48: \maketitle
49:
50: %---------------------------------------------------------------------%
51: \section{Introduction}
52: \label{intro}
53:
54: The instanton liquid model of the QCD vacuum \cite{shuryak82}
55: is based on a
56: semiclassical approximation, in which
57: %has been proven to be a theoretically grounded and phenomenologically
58: %successful description of the QCD vacuum.
59: all gauge configurations are replaced by
60: an ensemble of topologically non-trivial fields, instantons
61: and anti-instantons. It remains a model because
62: we do not still understand why large-size instantons are not present
63: in
64: the ensemble.
65: %Fits to phenomenology and later by lattice studies,
66: Fits to phenomenology and later lattice studies showed that
67: their total density is $n_0\simeq 1 fm^{-4}$ while the typical
68: size of about $\rho\sim 1/3 fm$,
69: leading to small diluteness parameter $n_0\rho^3\sim 10^{-2}$
70: \cite{shuryak82}.
71: With these parameters, the model quantitatively explains such
72: important phenomena as spontaneous $SU(N_f)$
73: chiral symmetry breaking for $N_f$ quark flavors,
74: the explicit U(1) symmetry breaking, and
75: many more other details of hadronic correlators and spectroscopy
76: (for a recent example see discussion of vector and axial correlators
77: \cite{SS_01}, for a
78: review see \cite{shuryakrev}).
79: The main feature of the instanton\footnote{For simplicity, we shall often use the term ``instanton'' to denote
80: instantons and/or
81: anti-instantons.}
82: ensemble is that each pseudo-particle
83: is an effective vertex with $2N_f$ quark lines \cite{thooftzm1},
84: which are exchanged between them and fill the vacuum.
85: A theory is developed, called Interacting Instanton Liquid Model
86: (IILM) which include these 't Hooft interactions {\em to all orders}
87: \cite{shuryakrev}.
88:
89: If new sources (external currents) are added, they produce
90: extra quarks which
91: interact with those in vacuum and
92: produce non-trivial correlation functions. In particular,
93: many (Lorentz scalar) chirally odd local operators obtain non-zero
94: vacuum expectation values.
95: In general, all of those
96: ``condensates'' and correlation functions are determined by
97: the interaction of instantons and thus
98: depend on the global (collective)
99: properties of the ensemble.
100:
101: On the other hand, as the instanton vacuum is
102: fairly dilute, one may think that the correlation
103: functions at distances short compared to instanton spacing $x\ll R=n^{-1/4}\sim 1 fm$
104: may be dominated by a $single$ instanton, the closest (or leading) one
105: (LI).
106: This framework ( which we shall refer to as the
107: ``single instanton approximation'', SIA ) has the advantage to allow
108: to carry out calculations analytically.
109: It is therefore possible to obtain closed expressions
110: for instanton contribution to Green's functions
111: in momentum or in Borel space.
112:
113: In SIA
114: collective contribution of all
115: instantons {\em other than the leading one}
116: are taken care of by a single effective
117: parameter, usually called effective mass, $m^*$. In the simplest
118: approximation, it can be associated with an $average$ value of the
119: quark condesate \cite{shifman80}:
120: \be
121: \label{mstar}
122: m^*= m - \frac{2}{3}\pi^2\,\rho^2 \,<\bar{u}\,u>,
123: \ee
124: which leads to
125: the value $m^*\simeq 170 MeV$ \cite{shuryak82}.
126: Note that it is already very different from what one infers from the same
127: model for long distance
128: (or zero Euclidean momentum) limit of the quark propagator, which
129: gives {\em constituent quark mass} of the order of 400 MeV.
130:
131: Furthermore,
132: although the SIA has been used in several phenomenological
133: studies (e.g. \cite{shuryak82}, \cite{forkel,forkel2,forkel3},
134: and references therein), its derivation
135: was never discussed in detail, its range of applicability
136: was never quantitatively checked, and the values of relevant effective
137: masses
138: well specified.
139: And
140: indeed, if one uses the value $m^*\simeq 170 MeV$ the
141: correlation functions, evaluated in the SIA,
142: do not agree with the results of the random and interacting
143: instanton liquid \cite{shuryakrev}.
144:
145: In this paper we identify the origin of such discrepancy
146: and calculate the values of effective mass appropriate for different
147: observables. This analysis reveals that the discrepancy between
148: SIA and full liquid calculations is due to an incorrect estimate
149: of the effective mass, $m^*$.
150: We also
151: present a systematic study of the SIA in QCD by itself.
152: We show that the approach is really accurate
153: only for calculations that involve operators of dimension six or more,
154: or correlators with
155: more than one zero-mode propagator.
156: We shall also prove that the mass terms, appearing
157: in matrix elements involving different
158: numbers of zero-mode propagators, are indeed independent parameters that have
159: to be fixed separately.
160: We provide with the definitions of all such mass factors in
161: terms of averages of the instanton ensembles
162: and prove that they are nearly universal,
163: i.e. the same for all similar correlation functions.
164:
165: %In particular, we argue that such parameter can not be extracted
166: %from the quark condensate
167: %but it can be obtained from correlations receiving contribution from at
168: %least two quark propagators in the zero-mode.
169:
170: The paper is organized as follows.
171: In section \ref{theory} we derive the SIA from the theory of the
172: instanton ensemble,
173: in section \ref{reliability} we
174: present the results of our numerical simulations that
175: estimate the contribution from the leading-instanton to several
176: correlation functions.
177: In section \ref{mass}, we evaluate the effective mass terms both from the
178: random and interacting instanton liquid and compare it
179: with the values obtained phenomenologically from the
180: pion sum-rule.
181: In section \ref{determass} we compare our effective masses with the
182: so-called ``determinantal masses'', which are other effective parameters that
183: can be defined in terms of averages of
184: the fermionic determinant.
185: The main results of our analysis are summarized in
186: section \ref{conclusions}.
187:
188: %---------------------------------------------------------------------------%
189:
190: \section{Quark Propagator}
191: \label{theory}
192:
193: In this section we review how the quark propagator in the instanton vacuum
194: is obtained and present consistent derivation of the SIA.
195:
196:
197: The quark propagator in general background field is
198: \begin{eqnarray}
199: \label{propdefinition}
200: S_I(x,y)=<x|(i\ds{D}_I+i\,m)^{-1}|y>,
201: \end{eqnarray}
202: where $\ds{D}_I$ denotes the Dirac operator. The inverse
203: (\ref{propdefinition}) can be formally represented as an
204: expansion in eigenmodes of the Dirac operator:
205: \be
206: \label{propspectral}
207: S_I(x,y) = \sum_\lambda \, \frac{\psi_\lambda (x)\, \psi^\dagger_\lambda(y)}
208: {\lambda+ i\, m}, \qquad i\,\ds{D}_I\psi_\lambda(x)
209: = \lambda \psi_\lambda(x).
210: \ee
211: From eq. (\ref{propspectral}) it follows that the propagator of light quarks
212: is dominated by eigenmodes with small virtuality.
213:
214: We begin by considering the academic case in which the
215: vacuum contains only one isolated instanton.
216: One eigenmode of $\ds{D}_I$
217: with zero virtuality (zero-modes) is given by 't Hooft
218: \cite{thooftzm1}, \cite{thooftzm2}:
219: \be
220: \label{zmodes}
221: i\ds{D}\,\psi_0(x) &=& 0, \nonumber\\
222: \psi_{0\,\,a\,\nu}(x;z) = \frac{\rho}{\pi}
223: \frac{1}{((x-z)^2+\rho^2)^{3/2}}&\cdot&\left[
224: \frac{1 - \gamma_5}{2}\, \frac{\ds{x}-\ds{z}}{\sqrt{(x-z)^2}}
225: \right]_{\alpha\,\beta} U_{a\,b}\,\epsilon_{\beta\,b},
226: \ee
227: where $z$ denotes the instanton position,
228: $\alpha, \beta= 1,\cdots
229: 4$ are spinor indices and $U_{a b}$ represents a
230: general group element.
231:
232: Isolating the contribution from zero-modes we can write:
233: \be
234: \label{S_I}
235: S_I(x,y;z) & = &\frac{\psi_0(x-z)\,\psi^\dagger_0(y-z)}{i\,m} +
236: \sum_{\lambda\ne 0} \, \frac{\psi_\lambda (x-z)\,
237: \psi^\dagger_\lambda(y-z)}{\lambda + i\, m}\nonumber\\
238: & = & S^{zm}_I(x,y;z) + S^{nzm}_I(x,y;z).
239: \ee
240: The zero-mode part of the propagator in the field of one
241: instanton can be evaluated from (\ref{S_I}) and
242: (\ref{zmodes}) to give, \cite{dyakonov86}:
243: \be
244: \label{S_zmI}
245: S^{zm}_I(x,y;z)=
246: \frac{(\ds{x}-\ds{z})\gamma_\mu \gamma_\nu
247: (\ds{y}-\ds{z})}{8 m}
248: \left[
249: \tau_\mu^- \tau_\nu^+ \,\frac{1-\gamma_5}{2}
250: \right]
251: \, \phi(x-z)\, \phi(y-z),
252: \ee
253: where
254: \be
255: \phi(t):=\frac{\rho}{\pi}\frac{1}{|t|\,(t^2+\rho^2)^{3/2} },
256: \qquad \tau_\mu^{\pm} := ({\bf \tau}, \mp i )
257: \ee
258: The corresponding expression in the field of one anti-instanton is obtained
259: through the substitution:
260: \be
261: \frac{1-\gamma_5}{2}\longleftrightarrow \frac{1+\gamma_5}{2}
262: \qquad
263: \tau^- \longleftrightarrow \tau^+.
264: \ee
265:
266: In the chiral limit, $m\to 0$, the expression for $S^{nzm}_I(x,y;z)$
267: is also known exactly \cite{brown78}.
268: In the limit of small distances ($|x-y|\to 0$), or if the instanton
269: is very far away ($|x-z|\to \infty$) one has:
270: \be
271: \label{nonzm}
272: S^{nzm}_I(x,y;z) \simeq S_0(x,y),
273: \ee
274: where $S_0$ denotes the free propagator.
275: Typically, corrections to eq. (\ref{nonzm}) lead to small contributions
276: and will be neglected in what follows.
277: Once the propagator has been calculated, one can in principle evaluate any
278: correlation function in the single-instanton background.
279:
280: Now, let's turn to the realistic vacuum of QCD.
281: Here, any configuration with a non-zero net topological charge
282: would be highly disfavored by the small value of the $\theta$-angle.
283: Therefore, one is lead to picture the vacuum as an ensemble with equal
284: density of instanton
285: and anti-instantons.
286: If the vacuum is dilute enough, the classical back-ground field can
287: be approximatively taken to be a superposition of
288: separated instantons and anti-instantons \footnote{For sake of simplicity, we
289: are explicitly dropping all collective coordinates, except the
290: instanton position $z_i$; moreover the use of the singular gauge
291: is assumed everywhere.}:
292: \begin{eqnarray}
293: \label{sumansatz}
294: A_\mu(x,\{\Omega_i\}_i)=\sum_{I}A^I_\mu(x,\{\Omega_i^I\}_i)
295: +\sum_{A}A^A_\mu(x,\{\Omega_i^A\}_i),
296: \end{eqnarray}
297: where $\{\Omega_i\}_i$ denotes the set of all collective coordinates.
298:
299: The propagator in such back-ground field can then be evaluated as follows
300: \cite{dyakonov86}. Let's consider the expansion:
301: \begin{eqnarray}
302: \label{Sexp1}
303: S= S_0 + S_0\,\ds{A}\,S_0 + S_0\,\ds{A}\,S_0\,\ds{A}\,S_0 + ...,
304: \end{eqnarray}
305: where integrations over
306: the positions of each background field insertion is understood.
307: The series (\ref{Sexp1}) can be rearranged so that all terms depending on
308: the collective coordinates of one
309: instanton field only are summed up first, followed by all terms depending
310: on two instantons and so on. One gets:
311: \be
312: \label{Sexp2}
313: S= S_0 + \sum_I ( S_I - S_0 ) + \sum_{I\ne J} ( S_I - S_0) S_0^{-1} (S_J-S_0)
314: +...,
315: \ee
316: where $S_I$ denotes the full propagator in the field of
317: the instanton $I$ so, in
318: the approximation ($\ref{nonzm}$) one has:
319: \be
320: \label{S_I-S_0}
321: (S_I-S_0)_{i\,j}(x,y) \simeq \frac{\psi^I_{0\,i}(x)\,
322: \psi^{\dagger\,I}_{0\,j}(y)}{i\,m},
323: \ee
324: where we have dropped all collective coordinates indices.
325: Inserting (\ref{S_I-S_0}) in (\ref{Sexp2}) and dropping also all
326: spinor indices we get:
327: \be
328: S(x,y) & \simeq & S^0(x,y) +
329: \sum_I \frac{\psi_0(x)\,\psi^\dagger_0(y)}{i\,m}
330: + \nonumber\\
331: & &\sum_{I,\,J} \frac{\psi_{0\,I}(x)}{i\,m}
332: \left(
333: \int d^4 z \psi^\dagger_{0\, I}(z)(i\ds{\partial}_z +i\,m)
334: \psi_{J\,0}(z) - i\,m \delta_{I\,J}
335: \right)
336: \frac{\psi_{0\,J}^\dagger(y)}{i\,m}+...,
337: \ee
338: where $- i m\, \delta_{I,J}$ has been added in order to relax the
339: $J\ne I$ constraint in the summation.
340: All the terms, starting from the second on, form a geometrical progression,
341: which can be re-summed to give:
342: \be
343: \label{S}
344: S(x,y) & \simeq & S^0(x,y)
345: + \sum_{I,\,J}
346: \psi_{0\,I}(x)\,
347: \left(
348: \frac{1}{T
349: %+ im
350: %\left(
351: %\int d^4 z \psi^\dagger(z)\psi_J(z)-\delta_{I\,J}
352: %\right) -im\delta{I\,J}}
353: + o(m)}
354: \right)_{I J}
355: \psi_{0\,J}^\dagger(y),
356: \ee
357: where $T_{I J}$ denotes the overlap matrix in zero-modes subspace
358: \be
359: T_{I J}=\int d^4 z \psi^\dagger(z)_I(i\ds{\partial})\psi(z)_J .
360: \ee
361:
362: In (\ref{S}), the zero-mode part the quark propagator is approximatively
363: written as a
364: bilinear form in the space spanned by the quark zero-mode
365: wave functions.
366: From (\ref{zmodes}) it follows that the contribution coming
367: from all the terms in the sum
368: associated to instantons very far away from the points $x$ and $y$ will be
369: negligible.
370: In particular, the biggest term in (\ref{S}) is associated to
371: the closest instanton, $I^*$.
372: Such instanton is dominating if the average of the correlation
373: function calculated retaining only the $(I^*,I^*)$ term in (\ref{S})
374: is much larger than
375: the average of the same quantity calculated from all other terms in
376: the sum (\ref{S}). Notice that this is a much weaker assumption than
377: demanding
378: \be
379: %\label{SIA}
380: \psi_{0\,I^*}(x)\,
381: \left(
382: \frac{1}{T + o(m)}
383: \right)_{I^* I^*}
384: \psi_{0\,I^*}^\dagger(y)
385: \gg \sum_{I\ne I^*,\,J\ne I^*}
386: \psi_{0\,I}(x)\,
387: \left(
388: \frac{1}{T + o(m)}
389: \right)_{I J}
390: \psi_{0\,J}^\dagger(y),
391: \ee
392: for \emph{each} configuration.
393:
394:
395: Let us summarize the framework developed so far.
396: First of all, the inverse matrix
397: $\left(
398: \frac{1}{T}
399: \right)_{I J}$
400: contains all the information about the particular configuration of the
401: instanton ensemble. In order to evaluate correlation functions,
402: one needs to average over all possible configurations.
403: Since contributions from distant instantons are suppressed by their
404: zero-mode wave functions, one expects correlation
405: functions with the
406: highest number of zero-modes to be most influenced by the
407: leading-instanton $I^*$.
408: %Also, for the same reason, one expects contribution from far
409: % instatons to scale with density and therefore become important
410: %for $|x-y|$ of the order of the typical distance between two instantons.
411: If it is possible to retain only
412: the contribution from $I^*$, the global properties of the ensemble
413: are present in the
414: matrix element
415: $\left(
416: \frac{1}{T}
417: \right)_{I^* I^*}$.
418:
419: As it was suggested long time ago by one of us \cite{shuryak82},
420: one can represent collective contribution of all other instantons,
421: by introducing an effective mass associated to quark propagating in the
422: zero-modes.
423: In other words, one assumes that for $|x-y|< 1 fm$,
424: the quark propagator can be written as:
425: \be
426: \label{SIAprop}
427: S(x,y)=\frac{\psi_0(x)\psi_0^\dagger(y)}{i m^*}.
428: \ee
429: With such propagator
430: all quark correlation functions
431: in the instanton back-ground could be evaluated simply by computing
432: all relevant Feynman diagrams and then
433: averaging over the instanton collective coordinates
434: \footnote{Notice that for all gauge invariant matrix elements,
435: the average over the color orientation is trivial.}.
436:
437: More specifically, in Random Instanton Liquid Model (RILM)
438: one introduces a model instanton density $n(\rho)$,
439: \be
440: n_I(\rho):= \bar{n}_I \,d(\rho),
441: \ee
442: where $\bar{n}_I=\bar{n}_A\simeq \frac{1}{2} fm^{-4}$ and
443: $d(\rho)$ represents the instanton size distribution. The latter is
444: schematically taken to be:
445: \be
446: \label{deltarho}
447: d(\rho)=\delta(\rho-\bar{\rho}).
448: \ee
449: with $\bar{\rho}\simeq 1/3 fm$.
450: This approach has the advantage to be considerably simple and was also
451: proven to be quite phenomenologically successful \cite{forkel,forkel2}.
452: However, we show below that
453: the effective mass defined in
454: (\ref{SIAprop}) is a quantity quite different from its naive estimate
455: (\ref{mstar}).
456:
457: In order to clarify the statement, let us first consider the quark condensate:
458: \be
459: \label{uu}
460: \chi_{uu}= <0|Tr \,\bar{u}(x) u(x) |0> = <Tr \,S(x,x)>,
461: \ee
462: where, in general, the average is done over all possible gauge
463: field configurations.
464: In the SIA is easily evaluated:
465: \be
466: <0| \bar{u} (x) \,u(x) |0>=
467: \int d^4 z \,\int d\rho\, \bar{n} \,d(\rho)
468: \left[
469: \frac{-2\,\rho^2}{[(z-x)^2+\rho^2]^3\,\pi^2\,m_{uu}},
470: \right]
471: \ee
472: where, for reasons that will become clear shortly, we have denoted with
473: $m_{uu}$ the quark effective mass and
474: $\bar{n}:= \bar{n}_I +\bar{n}_A$.
475: After performing the integrations one finds:
476: \be
477: \label{uuSIA}
478: \chi_{uu}= -\frac{\bar{n}}{m_{uu}},
479: \ee
480: for any normalized $d(\rho)$.
481:
482: Now, repeating the same calculation in the full liquid
483: \footnote{Here, we have neglected all
484: small current quark mass
485: terms in $1/T$.} gives:
486: \be
487: \label{uuRIL}
488: \chi_{uu} =
489: \left<
490: Tr
491: \left[
492: \sum_{I,J} \psi_{0\,I}(x) \,
493: \left(
494: \frac{1}{T}
495: \right)_{I\,J}
496: \psi_{0\, J}^\dagger(x)
497: \right]
498: \right>,
499: \ee
500: where, again, the average is made over all possible configurations of the
501: ensemble.
502: A comparison between (\ref{uuSIA}) and (\ref{uuRIL}) gives:
503: \be
504: \label{mstar1}
505: m_{uu}:=-\frac{\bar{n}}
506: {
507: \left<
508: Tr
509: \left[
510: \sum_{I,J} \psi_{0\,I}(x) \,
511: \left(
512: \frac{1}{T}
513: \right)_{I\,J}
514: \psi_{0\, J}^\dagger(x)
515: \right]
516: \right>
517: }
518: \ee
519:
520: Let us now consider another quark condensate:
521: \be
522: \label{uudd}
523: \chi_{uudd}:= <0| Tr\,[ \bar{u}(x) u(x) ]\cdot Tr\,[ \bar{d}(x) d(x) ] |0>
524: = < [ \,Tr\, S(x,x)]^2>.
525: \ee
526: Such condensate receives double contribution from zero-modes. In the SIA one
527: obtains:
528: \be
529: \chi_{uudd}=
530: \int d\rho \, d(\rho) \frac{\bar{n}}{5\,\pi^2\,\rho^4\,m_{uudd}^2},
531: \ee
532: where we have now denoted with $m_{uudd}$ the quark effective mass.
533:
534: Comparing, as before, with the result of full liquid calculations leads to:
535: \be
536: \label{mstar2}
537: m^2_{uudd}=
538: \left(\int\,d\rho\, d(\rho)
539: \frac{\bar{n}}{5\, \pi^2\,\rho^4}
540: \right)
541: \frac{1}{
542: \left<
543: \left[
544: Tr
545: \sum_{I,J} \psi_{0\,I}(x)
546: \left(
547: \frac{1}{T}
548: \right)_{I\,J}
549: \psi_{0\, J}^\dagger(x)
550: \right]^2
551: \right>
552: }
553: \ee
554:
555: Now, if the effective mass is universal,
556: $(m_{uu})^2 = m^2_{uudd}$, it would imply:
557: \be
558: \label{universality}
559: \frac{
560: \left<
561: Tr
562: \left[
563: \sum_{I,J} \psi_{0\,I}(x) \,
564: \left(
565: \frac{1}{T}
566: \right)_{I\,J}
567: \psi_{0\, J}^\dagger(x)
568: \right]
569: \right>^2}
570: {\left<
571: \left[
572: Tr
573: \sum_{I,J} \psi_{0\,I}(x)
574: \left(
575: \frac{1}{T}
576: \right)_{I\,J}
577: \psi_{0\, J}^\dagger(x)
578: \right]^2
579: \right>}=
580: \frac{5\,\pi^2 \bar{n}}{\int\,d\rho\, d(\rho)\frac{1}{\rho^4}}\simeq
581: 5\,\pi^2 \bar{n}\bar{\rho}^4\sim \frac{5}{8},
582: \ee
583: where we have used the ansatz (\ref{deltarho})
584: \footnote{Alternatively, we repeated the calculation
585: using a parameterization of the lattice
586: measurements of $d(\rho)$, which is peaked about somewhat higher
587: values of $\rho$ ($\rho\simeq 3.9$).
588: Both calculation give basically the same result.}.
589: Some comments on eq. (\ref{universality})
590: are in order.
591: First of all, in general quark condensate is rather inhomogeneous,
592: and for parametrically dilute instanton ensemble
593: this ratio is small. However, with empirical diluteness it happens to
594: be not so small, about 0.6 . In principle,
595: by measuring the left-hand side and right-hand-side
596: of (\ref{universality}) on the lattice $separately$,
597: one can estimate the accuracy of the
598: universality of the effective mass.
599:
600: However, since different configurations and even points have different
601: leading instanton, the corresponding value $T_{I^* I^*}$ fluctuates,
602: and the average of its different powers in general leads to different
603: effective masses. (This effect should not be confused with
604: the inhomogeneity of the condensates discussed above.)
605: Let us define a parameter $R_m$, such that $R_m=1$ means universal mass
606: $(m_{uu})^2 = m_{uudd}^2$:
607: \be
608: \label{Rm}
609: R_m:=
610: \frac{
611: \left<
612: Tr
613: \left[
614: \sum_{I,J} \psi_{0\,I}(x) \,
615: \left(
616: \frac{1}{T}
617: \right)_{I\,J}
618: \psi_{0\, J}^\dagger(x)
619: \right]
620: \right>^2}
621: {5\,\pi^2\,\bar{\rho}^4\, \bar{n}
622: \left<
623: \left[
624: Tr
625: \sum_{I,J} \psi_{0\,I}(x)
626: \left(
627: \frac{1}{T}
628: \right)_{I\,J}
629: \psi_{0\, J}^\dagger(x)
630: \right]^2
631: \right>}.
632: \ee
633:
634: %-----------------------------------------------------------------------%
635:
636: \section{Numerical study of the Single Instanton Approximation.}
637: \label{reliability}
638:
639: In general, reliability of the SIA
640: depends on the vacuum diluteness.
641: In this section, we want to establish whether the
642: QCD vacuum with realistic density is actually
643: dilute enough for the leading-instanton to be dominant, at least for
644: some observables.
645:
646: For this purpose we have performed numerical analysis of several correlation
647: functions, measured in the random instanton liquid model.
648: In such ensemble, the vacuum expectation values are obtained by averaging
649: over configurations of
650: randomly distributed instantons of size $\rho = 1/3 fm$.
651: The contribution from the leading-instanton is evaluated by retaining
652: only the largest term in (\ref{S}), for each configuration.
653:
654: We begin by considering two quark condensates $\chi_{uu}$ and
655: $\chi_{uudd}$, introduced in (\ref{uu}) and (\ref{uudd}).
656: We will show later that they represent all generic
657: observables which receive contribution from
658: one and two zero-mode propagators, respectively.
659:
660: In this calculation we average
661: $5000$ configurations of 20 instantons in
662: a box of volume $3.4\times 1.8^3 fm^4$.
663: The results of this simulation
664: are presented in table \ref{condensates}.
665:
666: \begin{table}%[h]
667: \caption{Quark condensates evaluated in the full instanton ensemble
668: and from the leading-instanton, only.}
669: \label{condensates}
670: %\begin{ruledtabular}
671: \begin{tabular}{|c|c|c|}\hline
672: \multicolumn{1}{|c|}{condensate} &
673: \multicolumn{1}{|c|}{complete calculation} &
674: \multicolumn{1}{|c|}{LI } \\ \hline \hline
675: $\chi_{uu}$ & $ ( - 232 \pm 5 MeV )^3 $& $(- 198 \pm 1 MeV)^3$ \\ \cline{2-3}
676: $\chi_{uudd}$ &$ ( 310 \pm 7 MeV)^6 $& $ (309 \pm 3 MeV)^6$ \\ \hline
677: \end{tabular}
678: % \end{ruledtabular}
679: \end{table}
680:
681: From these results one can see how the accuracy of SIA
682: (keeping only the closest instanton) depends on the particular matrix
683: element being evaluated.
684: Naturally, the accuracy increases with the dimension of the operator
685: involved, because it diminish the contribution of distant instantons.
686: Specifically, SIA for dimension-six local operators which
687: receive contribution from two zero-mode propagators
688: agree with full calculation
689: within a few percent.
690: On the other hand, prediction for operators/correlators with
691: only one zero-mode propagators are not really accurate: the
692: error in quark condesate is large ( $\gtrsim 35\%$ ).
693:
694: Next we consider two-point correlation functions.
695: This allows us to determine the scale at which the closest instanton is no
696: longer dominant.
697: At this purpose we have measured the pion pseudo-scalar two point function,
698: \be
699: \label{pion2point}
700: P(x) := <0| J_5(x) J^\dagger_5 (0)|0>,
701: \ee
702: where,
703: \be
704: J_5(x) := \bar{u}(x)\,\gamma_5 \,d(x).
705: \ee
706:
707: This particular choice is motivated by the fact that such correlation
708: function is known to receive maximal contribution from quark zero-modes
709: \cite{shuryakcorr}.
710: One expects many instantons effects to become important for $|x|$
711: larger than the instanton size
712: and smaller than the typical distance between two neighbor instantons:
713: \be
714: \label{SIArange}
715: 1/3 fm \lesssim |x| \lesssim 1 fm
716: \ee
717: Results of simulations including the contribution from all
718: instantons and form the leading-instanton only are reported in figure
719: (\ref{RILvsLI}).
720: One can see that the agreement is lost for rather large values of $|x|$
721: ($|x| \gtrsim 0.6 fm$).
722:
723: In the last section we saw that the SIA does not only assume
724: leading-instanton
725: dominance, but involves some effective mass parameters, which
726: collectively describe the effects of all other instantons
727: (see eqs. (\ref{mstar1}) and (\ref{mstar2})).
728: In the next section we shall determine numerically
729: such parameters.
730:
731: %-----------------------------------------------------------------------%
732:
733: \section{Numerical study of the Quark Effective Mass Parameters}
734: \label{mass}
735: In section \ref{theory} we argued that the universality of the
736: effective mass, which collectively describes the effects of all
737: non-leading-instantons, can be put in relation to the fluctuations
738: of the quark condensates through eq. (\ref{universality}).
739:
740: Obviously, the accuracy of calculations in the SIA
741: depends on the value of $R_m$ (defined in (\ref{Rm}))
742: in realistic ensembles.
743:
744: We have have evaluated $R_m$ and the corresponding effective masses,
745: numerically
746: \footnote{Here we have averaged on 5000 configuration of
747: 256 instantons in a $4^4 fm^4$ box.}
748: in the random instanton liquid and
749: in the interacting liquid
750: (for a review of these ensembles see \cite{shuryakrev}).
751: Our results are summarized in table \ref{massresult}:
752: \begin{table}%[h]
753: \caption{Universality parameter, $R_m$ and the effective masses evaluated in
754: the RILM and in the IILM}
755: \label{massresult}
756: \begin{tabular}{|c|c|c|}\hline
757: \multicolumn{1}{|c|}{Quantity} &
758: \multicolumn{1}{|c|}{RILM calculation} &
759: \multicolumn{1}{|c|}{IILM calculation} \\ \hline \hline
760: $R_m$ & $0.4$ & $0.2$ \\ \cline{2-3}
761: $m_{uu}$ & $\ 120 MeV$ & $ 177 MeV$ \\ \cline{2-3}
762: $\sqrt{m_{uudd}^2}$&$ 65 MeV$ & $ 91 MeV$ \\ \hline
763: \end{tabular}
764: \end{table}
765:
766: These results show that, in the instanton vacuum with realistic density,
767: the {\em universality does not hold}
768: \be
769: \label{inequality}
770: m_{uu}^2\ne m^2_{uudd}.
771: \ee
772: This implies that an effective mass extracted from the quark condesate can
773: not be used in calculations involving more than one zero-mode propagator.
774:
775: On the other hand, the results of numerical simulations presented in
776: section \ref{reliability} have shown that matrix elements involving
777: only one zero-mode propagator (like the quark condensate)
778: can not be reliably evaluated in the SIA, simply because the leading
779: instanton is not dominant.
780: As a consequence, one is forced to consider only
781: correlation functions involving at least two
782: such propagators and therefore $m_{uu}$ is of no practical usefulness.
783:
784: In more general terms, one may address the question
785: whether the effective mass parameter
786: depends on the particular correlation
787: function being evaluated.
788: If so, this feature would spoil much of the predictive power
789: of the SIA.
790: In such pessimistic scenario
791: the SIA would only allow
792: to work out the functional expressions of small-sized correlations,
793: but not their overall normalization.
794: However we will show that
795: the effective mass parameters depend essentially on
796: the number of zero-mode propagators involved, and that
797: $m^2_{uudd}$ is in a way universal for a number of applications.
798: In this case, SIA is predictive including the normalization.
799: %For example, in calculations involving $u$ and $d$ quarks,
800: %it would be necessary to know
801: %$m_1:=m_{uu}$, appearing in calculations with one
802: %zero-mode propagators, and $m_2^{\,2}:=m_{uudd}^2$ appearing when two
803: %quark zero-mode propagators contribute.
804: To check that
805: we have extracted
806: $m^2_2$
807: from the analysis of
808: several hadronic two-point functions evaluated in SIA and in the liquid.
809: In particular, we considered the pion pseudo-scalar the scalar diquark and the
810: a nucleon scalar correlation functions:
811: \be
812: P(x) &=& <0| J_5(x) J^\dagger_5 (0)|0>,\\
813: D(x) &=& <0| J^a_{C5}(x) J^{a\,\dagger}_{C5} (0)|0>,\\
814: N(x) &=&<0| Tr\,[ \eta (x)\,\bar{\eta}(0) \gamma_4 ]|0>,
815: \label{proton}
816: \ee
817: where,
818: \be
819: J_5(x) &:=& \bar{u}(x)\,\gamma_5 \,d(x).\\
820: J^a_{C5}(x) &:=& \epsilon^{a\,b\,c}u_b(x)\,C \gamma_{5} \,d_c(x).\\
821: \eta_\alpha (x) &:=& \epsilon^{a\,b\,c}
822: ( u^a(x) C\gamma_5 u^b(x)) u_\alpha^c(x).
823: \ee
824: All these correlations function are known to receive contribution
825: from two propagators in the zero-mode.
826:
827: The comparison between results obtained in the SIA, in the random instanton
828: liquid model (RILM) and in the interacting instanton liquid model (IILM)
829: are reported in figs. (\ref{figps2point}),
830: (\ref{figdiquark2point}) and
831: (\ref{fignucleon2point}).
832: The corresponding values for $\sqrt{m_2^2}$ are presented in
833: table (\ref{manymasses}).
834: These
835: values are indeed rather different from
836: the traditionally adopted estimate
837: $m^*=170 MeV$, extracted from the quark condensate.
838:
839: The general reason why these masses are rather small is the
840: following.
841: Instantons have fluctuating strength of interaction with others in the
842: ensemble: some of them are ``hermits'' and have small matrix
843: elements
844: in the corresponding entries of the
845: overlap matrix $T$. As in all expressions we average the {\em inverse}
846: of this
847: matrix, the contribution of such ``hermits'' is enhanced.
848: This lowers the value of the effective masses.
849: Furthermore, because random ensemble of RILM has more
850: such ``hermits'', as compared to
851: IILM (where the fermionic determinant in the statistical weight
852: suppresses them), these masses are smaller in RILM as compared to IILM.
853: Such discrepancy reflects the fact that the two ensembles give
854: actually quite different correlation functions \cite{shuryakrev}.
855:
856:
857: \begin{table}%[h]
858: \caption{Estimates of the quark effective mass $m^2_2$ from several correlation functions.}
859: \label{manymasses}
860: \begin{tabular}{|c|c|c|}\hline
861: \multicolumn{1}{|c|}{Correlation function} &
862: \multicolumn{1}{|c|}{ $\,m_2^2 \,[MeV^2]\,$ (RILM)} &
863: \multicolumn{1}{|c|}{ $\,m_2^2 \,[MeV^2]\,$ (IILM)} \\ \hline \hline
864: $\chi_{uudd}$ condensate & $ ( 65 )^2$ & $( 91 )^2$\\ \cline{2-3}
865: pion pseudo-scalar & $ ( 65 )^2$ & $( 105 )^2$\\ \cline{2-3}
866: diquark scalar & $ ( 69 )^2$ & $( 105 )^2$\\ \cline{2-3}
867: nucleon scalar & $ ( 67 )^2$ & $( 105)^2$\\ \hline
868: \end{tabular}
869: \end{table}
870:
871: From these results we conclude that $m_2^2$ seems to be a
872: universal parameter, describing the collective many-instanton effects.
873:
874: It is important to know what value of $m_2^{\,2}$ is suggested
875: by the available phenomenology.
876: As before, we chose to consider the pion
877: pseudo-scalar correlator, because it receives maximal contribution from
878: instanton zero-modes.
879: The traditional ``pole-plus-continuum'' model for the spectral
880: decomposition of $P(x)$, gives \cite{meson2pt}, \cite{shuryakcorr}:
881: \be
882: \label{pssumrule}
883: P(x)= \lambda_\pi^2 D(m_\pi;x) + \frac{3}{8\,\pi^2}\int_{s_0}^\infty
884: \, ds\, s D(\sqrt{s};x),
885: \ee
886: where $D(m;x)$ is the scalar propagator,
887: $s_0$ is the threshold for the continuum ( $\sqrt{s_0} \simeq 1.6 GeV $) and
888: the pseudo-scalar decay constant $\lambda_\pi$ is given by:
889: \be
890: \lambda_\pi =<0| \bar{u}\gamma_5d|\pi>= \frac{f_\pi\,m_\pi^2}{m_u+m_d}
891: \simeq (480 MeV)^2.
892: \ee
893:
894: \begin{table}%[h]
895: \caption{Determinantal masses evaluated in the RILM and in the IILM as
896: compared to $m_1$ and $m_2^2$, defined in section \ref{mass}.}
897: \label{determinantalmass}
898: %\begin{ruledtabular}
899: \begin{tabular}{|c|c|c|}\hline
900: \multicolumn{1}{|c|}{mass} &
901: \multicolumn{1}{|c|}{RILM calculation} &
902: \multicolumn{1}{|c|}{IILM calculation} \\ \hline \hline
903: $m_1$ & $120 MeV $ & $177 MeV$ \\ \cline{2-3}
904: $m_{det}$ & $ 63 MeV $ & $102 MeV$ \\ \hline
905: $ m_2^2 $ & $ (65 MeV)^2 $ & $(103 MeV)^2$ \\ \cline{2-3}
906: $m_{det}^2$
907: &$ (64 MeV)^2 $& $(103 MeV)^2$ \\ \hline
908: \end{tabular}
909: % \end{ruledtabular}
910: \end{table}
911:
912: We determined $m_2^2$, by fitting the SIA prediction
913: to the phenomenological curve
914: obtained from (\ref{pssumrule}).
915: We found (see fig. \ref{figps2point}):
916: \be
917: \label{m2phen}
918: m_{2\, phen.}^2= (86 MeV)^2.
919: \ee
920: To further check the approach,
921: we have evaluated the scalar proton two-point function $N(x)$, using the
922: value (\ref{m2phen}) and we have compared with the phenomenological curve
923: (see fig. (\ref{fignucleon2point})
924: \footnote{Notice that, in this case, the pole contribution depends on the
925: nucleon ``decay-constant'' $\Lambda_s$, which is
926: not known experimentally. We have therefore used an
927: estimate ($\Lambda_s=2.5 fm^{-3}$) which reasonably agrees
928: with several model calculations
929: \cite{shuryakrev} .}).
930: In summary: with this value
931: we obtained good very good agreement with phenomenology
932: and therefore we suggest that (\ref{m2phen}) should be used f
933: the applications of the SIA, when two zero-mode propagators are involved.
934:
935:
936: %-----------------------------------------------------------------------%
937: \section{Evaluation of an effective mass in the fermionic determinant}
938: \label{determass}
939:
940: The propagator is not the only place where the Dirac operator appears:
941: the QCD statistical sum contains its $determinant$, appearing in power
942: given by the number of light quark flavors $N_f$. If one considers
943: the academic vacuum with only one instanton, this determinat
944: contains the product of ``current'' quark masses for all
945: quarks \cite{thooftzm1}.
946: If this would be the final answer for the instanton density,
947: instanton effect would be strongly suppressed by their small values.
948:
949: However, in physical
950: vacuum there are sufficiently many instantons to break chiral
951: symmetry and produce non-zero quark condesates and effective quark
952: masses,
953: which substitute for much smaller ``current'' masses and make instanton
954: effects significantly stronger. The interplay between these effective
955: masses
956: and current quark masses is especially interesting for strange quark,
957: since the former and the latter, $m_s$, are of comparable magnitude.
958: This issue has been studied e.g. in recent paper \cite{Musakhanov},
959: where it was concluded that the usual additive formula for total
960: effective quark mass of the strange quark
961: $M_s^{tot}=m_{eff}(m_s=0)+m_s$ is wrong, and the true value
962: of
963: $M_s^{tot}$ is not very different from that for u,d quarks because
964: $m_{eff}(m_s)$ strongly decreases with $m_s$.
965:
966: Apart of the role of strange quark mass in general, there is also a
967: general issue of correct connection of units and vacuum parameters
968: (with the instanton density being one of them) for QCD with different
969: number of flavors. (For example, between no-quark or quenched QCD and
970: the physical world.)
971: In order to study all of this, it important to know what is the
972: absolute
973: magnitude of the fermionic determinants in the
974: instanton-based vacuum models considered. Some of those are repoted in
975: this section.
976:
977: In the instanton-based model context, the fermionic determinant
978: is usually represented by the determinant of the overlap matrix $T$
979: (see description e.g. in \cite{shuryakrev}) in the zero mode subspace.
980: After averaging over appropriate ensemble, one can
981: define the so-called ``determinantal masses'':
982: \be
983: m_{det}^i:=
984: \frac{<(\,det [\ds{D}]\,)^{i/N}\rho^i >}{<\rho^i>},\qquad i=1,2,...
985: \ee
986: where index $i$ refers to number of flavors and
987: $N$ denotes the number of instantons.
988: Their values tell us how much the presence of fermions reduces
989: the instanton density, compared to the same ensemble without them.
990:
991:
992: Originally, in \cite{shifman80}, \cite{shuryak86}
993: an estimate for the determinant
994: effective mass was estracted from the averaging of the 't Hooft Lagrangian
995: assuming factorization of quark condesates, and using the same
996: $m^*=170 \, MeV$. If so, each flavor reduces instanton density
997: by the factor $m^*\bar \rho\approx 0.28$.
998: As we will see shortly, the corresponding reduction
999: factor is actually even smaller.
1000:
1001: In principle, there is no reason why
1002: the values of $m_{det}$ and $m_{det}^2$ should agree
1003: with $m_1$ and $m_2^2$, defined in the previous section:
1004: we now averahe positive rather than negative powers of the
1005: overlap matrix.
1006:
1007: We have evaluated the determinantal masses
1008: in the RILM and in the IILM.
1009: Results are reported in table \ref{determinantalmass}.
1010: Some comments are in order.
1011: First of all note that, in both ensembles, the values of
1012: $m_{det}^2$ turn out to be quite consistent
1013: with the values of $m_2^2$.
1014: Furthermore, the fluctuations of the determinantal mass,
1015: $m_{det}$ and $(m_{det})^2$ are very small:
1016: \be
1017: m_{det}^2- (m_{det})^2 \ll m_2^2 - (m_1)^2,
1018: \ee
1019: implying essentially that $m_1$ is inconsistent with $m_{det}$.
1020: This fact could have two possible explanations.
1021: On the one hand, one could argue that $m_1$ is a somewhat
1022: ill-defined parameter, because the SIA can not be used to
1023: evaluate quark condensate.
1024: On the other hand, one could observe that larger fluctuations
1025: for the effective masses defined
1026: in \ref{mass} should not be surprising, since such parameters appear
1027: always in denominators of SIA calculations.
1028:
1029: %-----------------------------------------------------------------------%
1030:
1031: \section{Conclusions and Outlook}
1032: \label{conclusions}
1033:
1034: Summarizing our study of the SIA approximation
1035: in QCD, we first notice that this
1036: approach has been related to the theory of the full
1037: ensemble and all the effective parameters
1038: previously loosely called ``effective masses'' are defined.
1039: All of them describe different aspects of
1040: collective interaction between the
1041: ``leading'' instanton (the closest to the observation points) and
1042: all others, and related to the overlap matrix $T$ .
1043: Different effective mass values simply follow from different
1044: ensemble averaging. In particular,
1045: the factor $\frac{1}{m_1}$, appearing in SIA calculations
1046: with one
1047: propagator in the zero-mode, does not correspond to the square root of the
1048: factor $\frac{1}{m^2_2} $, appearing when two such propagators
1049: are involved.
1050: %Moreover, we showed that this difference is not due the fluctuations
1051: %of the ensemble, since it emerges already in a mean-field calculation
1052: %($R_m\simeq \frac{8}{5}$,
1053: %where all fluctuations are neglected).
1054:
1055: We have made numerical simulations in the RILM and IILM and found
1056: that the contribution of the leading-instanton actually dominates all
1057: condensates of operators of dimension six or more, as well as
1058: short-distance correlation functions ($|x|\lesssim 0.6 fm$).
1059: This however is true only for correlation functions with
1060: at least two zero-mode propagators involved.
1061: Earlier estimates extracted from the quark
1062: condensate are not accurate.
1063:
1064: Furthermore, the parameter
1065: $\frac{1}{m^2_2}$ is approximatively universal
1066: for several correlation functions with \emph{two}
1067: zero-mode propagators involved.
1068: We have also extracted a phenomenological estimate
1069: of its value from the analysis of the
1070: pion pseudo-scalar correlator.
1071: We found $m_{2\, phen.}^2 \simeq (86 MeV)^2$, much
1072: smaller than the value originally obtained from the quark condensate.
1073: Our new value should
1074: be used in many applications of the SIA.
1075:
1076: Finally, we have compared our estimates for the effective mass parameters
1077: $m_1$ and $m_2^2$, with
1078: the measurements of the ``determinantal'' masses,
1079: introduced in \cite{shuryak86}.
1080: We observed substantial agreement between $m_2^2$ and $m_{det}^2$
1081: both in the RILM and the IILM, but different
1082: from $m_1$ extracted from the quark condesate alone.
1083: This implies that light quarks are about twice more effective
1084: (per flavor) in diluting
1085: the instanton vacuum density.
1086:
1087:
1088: \begin{figure}[h|t]
1089: \includegraphics[height=7.0cm,width=8.0cm]{RIL_vsLI.eps}
1090: \caption{Pion pseudo-scalar correlation function in the RILM,
1091: normalized to the same correlation function in the free theory.
1092: The solid line corresponds to the full RILM simulation,
1093: the dashed line denotes the leading-instanton contribution.}
1094: \label{RILvsLI}
1095: \end{figure}
1096:
1097: \begin{figure}[h|t]
1098: \includegraphics[height=7.0cm,width=8.0cm]{pion.eps}
1099: \caption{Pion pseudo-scalar two-point function
1100: normalized to the same correlation function in the free theory.
1101: The open circles (squares) represent RILM (IILM) points, the
1102: dashed lines represent SIA calculations with masses given in table
1103: \ref{manymasses} and the dotted line
1104: is the phenomenological curve obtained from the spectral decomposition.}
1105: \label{figps2point}
1106: \end{figure}
1107: \begin{figure}[h|t]
1108: \includegraphics[height=7.0cm,width=8.0cm]{diquark.eps}
1109: \caption{Diquark scalar two-point function
1110: normalized to the same correlation function in the free theory.
1111: The open circles (squares) represent RILM (IILM) points and the
1112: dashed lines represent SIA calculations with the effective
1113: masses given in table \ref{manymasses}}.
1114: \label{figdiquark2point}
1115: \end{figure}
1116: \begin{figure}[h|t]
1117: \includegraphics[height=7.0cm,width=8.0cm]{nucl.eps}
1118: \caption{Nucleon scalar two-point function normalized to the same
1119: correlation function in the free theory.
1120: The open circles (squares) represent RILM (IILM) points and the
1121: dashed lines represent SIA calculations with the effective
1122: masses given in table \ref{manymasses}.}
1123: \label{fignucleon2point}
1124: \end{figure}
1125: %-------------------------------------------------------------------%
1126: \begin{acknowledgments}
1127: We would like to thank H.Forkel whose questions initiated this work, and
1128: T. Sch\"afer for many helpful discussions and numerical help.
1129: The work is partly supported by the US DOE grant No. DE-FG02-88ER40388.
1130: \end{acknowledgments}
1131: %-----------------------------------------------------------------------%
1132:
1133:
1134: %\bibliography{instanton}
1135:
1136: \begin{thebibliography}{99}
1137: \bibitem{shuryak82}
1138: E.V.Shuryak, Nucl. Phys., \textbf{B214} (1982), 237.
1139: \bibitem{SS_01}
1140: T.~Schafer and E.~V.~Shuryak,
1141: %``Implications of the ALEPH tau lepton decay data for perturbative and non-perturbative QCD,''
1142: Phys.\ Rev.\ Lett.\ {\bf 86}, 3973 (2001)
1143: [hep-ph/0010116].
1144: %%CITATION = HEP-PH 0010116;%%
1145: \bibitem{shuryakrev}
1146: T.Schaefer and E.V.Shuryak, Rev. Mod. Phys., \textbf{70} (1998) 323.
1147: \bibitem{shifman80}
1148: M. A. Shifman, A.I. Vainshtein, A.I. Zakharov, Nucl. Phys., \textbf{B163}
1149: (1980) 46.
1150: \bibitem{forkel}
1151: H. Forkel and M. Nielsen, Phys. Lett., \textbf{B345} (1997) 55.
1152: \bibitem{forkel2}
1153: H. Forkel and M. Nielsen, Phys. Rev., \textbf{D55} (1997) 1417.
1154: \bibitem{forkel3}
1155: M. AW, M.K. Banerjee and H. Forkel, Phys. Lett., \textbf{B454} (1999) 147.
1156: \bibitem{thooftzm1}
1157: G.'t Hooft, Phys. Rev. Lett., \textbf{37} (1976), 8.
1158: \bibitem{thooftzm2}
1159: G.'t Hooft, Phys. Rev., \textbf{D14}(1976), 3432.
1160: \bibitem{dyakonov86}
1161: D.J. Dyakonov and V.Yu. Petrov,
1162: Nucl. Phys.,
1163: \textbf{B272} (1986) 457.
1164: \bibitem{brown78}
1165: L.S. Brown, R.D. Carlitz, D.B. Craemer and C. Lee,
1166: Phys. Rev.,
1167: \textbf{D17} (1978) 1583.
1168: \bibitem{chu94}
1169: M.C. Chu, J.M. Grandy, S. Huang and J.W. Negele,
1170: Phys. Rev., \textbf{D49} (1994) 6039.
1171: \bibitem{shuryakcorr}
1172: E.V.Shuryak, Rev. Mod. Phys., \textbf{65} (1993) 1.
1173: \bibitem{shuryak86}
1174: E.V.Shuryak, Nucl. Phys., \textbf{B302} (1986) 599.
1175: \bibitem{meson2pt}
1176: E.V. Shuryak and J.J.M. Verbaarschot, Nucl. Phys.,\textbf{B140} (1993) 37.
1177: \bibitem{Musakhanov}
1178: M.~Musakhanov,
1179: ``Current mass dependence of the quark condensate and the
1180: constituent quark mass,''
1181: hep-ph/0104163.
1182: %%CITATION = HEP-PH 0104163;%%
1183:
1184: \end{thebibliography}
1185: \end{document}
1186:
1187:
1188:
1189:
1190:
1191:
1192:
1193:
1194:
1195:
1196:
1197:
1198:
1199:
1200:
1201:
1202:
1203:
1204:
1205:
1206:
1207:
1208: