0810.2333/LC.tex
1: % Please use the skeleton file you have received in the 
2: % invitation-to-submit email, where your data are already
3: % filled in. Otherwise please make sure you insert your 
4: % data according to the instructions in PoSauthmanual.pdf
5: \documentclass{PoS}
6: 
7: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
8: %           packages
9: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
10: 
11: \usepackage{amsfonts,amssymb,amsmath}
12: \usepackage{graphicx}
13: 
14: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
15: %      definitions
16: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
17: \newcommand{\be}{\begin{equation}}
18: \newcommand{\bea}{\begin{eqnarray}}
19: \newcommand{\ee}{\end{equation}}
20: \newcommand{\eea}{\end{eqnarray}}
21: \newcommand{\bpi}{\begin{picture}}
22: \newcommand{\bce}{\begin{center}}
23: \newcommand{\epi}{\end{picture}}
24: \newcommand{\ece}{\end{center}}
25: \newcommand{\D}{\displaystyle}
26: \def\chic#1{{\scriptscriptstyle #1}}
27: 
28: \def\gb{{\rm I}\hspace{-0.07cm}\Gamma}
29: \def\kb{{\rm I}\hspace{-0.07cm} {\rm K}}
30: \def\l{{\rm I}\hspace{-0.07cm} {\rm L}}
31: %\def\gb{\bm{\Gamma}}
32: %\def\kb{\bm{\rm K}}
33: %\def\l{\bm{\rm L}}
34: 
35: 
36: %\def\k{\widetilde{{\rm I}\hspace{-0.07cm}{\rm K} }}
37: \def\g{\widetilde\gb}
38: %\def\k{\widetilde{\mathbb{K}}}
39: \def\k{\widetilde\kb}
40: 
41: 
42: 
43: 
44: \def\gb{{\rm I}\hspace{-0.07cm}\Gamma}
45: \def\g{\widetilde{{\rm I}\hspace{-0.07cm}\Gamma}}
46: \def\gt{\widetilde{\Gamma}^{\rm L}_{\nu\alpha\beta}}
47: \def\gnp{{\overline g}^2_{{\chic {\rm NP}}}}
48: %\renewcommand{\theequation}{\arabic{section}.\arabic{equation}}
49: \newcommand{\Valencia}{Departamento de F\'\i sica Te\'orica
50: and IFIC, Centro Mixto, Universidad de Valencia--CSIC \\
51: E-46100, Burjassot, Valencia, Spain}
52: 
53: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
54: 
55: \title{Infrared finite effective charge of QCD}
56: 
57: \ShortTitle{Infrared finite effective charge of QCD}
58: 
59: 
60: 
61: %\author{A. C. Aguilar, \speaker{J. Papavassiliou} \\
62: %       \Valencia \\
63: %        E-mail: \email{Arlene.Aguilar@uv.es}, \email{Joannis.Papavassiliou@uv.es}}
64: 
65: 
66: %\author{D.~Binosi\\
67: %        European Centre for Theoretical Studies in Nuclear
68: %  Physics and Related Areas (ECT*), \\Villa Tambosi, Strada delle
69: %  Tabarelle 286, I-38050 Villazzano (TN), Italy. \\
70: %        E-mail: \email{binosi@ect.it}}
71: 
72: 
73: 
74: \author{A. C. Aguilar,$^a$ D.~Binosi$^{b}$
75: and  \speaker{J. Papavassiliou}$ ^a$\\
76: \llap{$^a$}\Valencia \\
77: \llap{$^b$}  European Centre for Theoretical Studies in Nuclear
78:   Physics and Related Areas (ECT*), \\Villa Tambosi, Strada delle
79:   Tabarelle 286, I-38050 Villazzano (TN), Italy. \\
80: E-mail: \email{Arlene.Aguilar@uv.es},\email{binosi@ect.it}, \email{Joannis.Papavassiliou@uv.es}}
81: 
82: 
83: 
84: 
85: \abstract{
86: 
87: We  show that  the gauge  invariant treatment  of  the Schwinger-Dyson
88: equations  of  QCD  leads  to  an infrared  finite  gluon  propagator,
89: signaling the  dynamical generation of an effective gluon mass, and
90:  a  non-enhanced ghost  propagator,  in  qualitative agreement  with
91: recent lattice data.   The truncation scheme employed is  based on the
92: synergy  between   the  pinch  technique  and   the  background  field
93: method. One of  its most powerful features is  that the transversality
94: of the gluon self-energy  is manifestly preserved, exactly as dictated
95: by the  BRST symmetry of the  theory.  We then explain,  for the first
96: time  in  the  literature,   how  to  construct  non-perturbatively  a
97: renormalization group invariant quantity out of the conventional gluon
98: propagator.   This  newly constructed  quantity  serves  as  the  natural
99: starting point  for defining  a non-perturbative effective  charge for
100: QCD,  which constitutes,  in  all respects,  the  generalization in  a
101: non-Abelian  context  of the  universal  QED  effective charge.   This
102: strong effective charge displays asymptotic freedom in the ultraviolet,
103: while in the low-energy regime  it freezes at a finite value, giving
104: rise  to an  infrared  fixed  point for  QCD.  Some possible  pitfalls
105: related to the extraction of  such an effective charge from infrared
106: finite gluon propagators, such as those found on the lattice, 
107: are briefly discussed.}
108: 
109: \FullConference{LIGHT CONE 2008 Relativistic Nuclear and Particle Physics\\
110: 		 July 7-11  2008\\
111: 		 Mulhouse, France}
112: \begin{document}
113: 
114: 
115: It  has been  known for  a long  time that  even though  the  gluon is
116: massless  at the  level  of the  fundamental  Lagrangian, and  remains
117: massless to all order in perturbation theory, the non-perturbative QCD
118: dynamics  generate  an  effective,  momentum-dependent  mass,  without
119: affecting    the   local    $SU(3)_c$   invariance,    which   remains
120: intact~\cite{Cornwall:1982zr}.    The  existence   of  this   mass  is
121: discovered by studying the Schwinger-Dyson equations (SDEs) of QCD, in
122: a  gauge-invariant framework known  as the  pinch technique  
123: (PT)~\cite{Cornwall:1982zr,Cornwall:1989gv}.
124: To obtain  massive
125: solutions it  is necessary to  include longitudinally-coupled massless
126: scalars  in the  Green's  functions,  which play  a  role rather  like
127: Goldstone  excitations, but  do not  signal any  sort of  breakdown of
128: local gauge  symmetry, which  is preserved.   Like standard
129: Goldstone bosons  these massless scalars  do not appear  explicitly in
130: the  $S$-matrix; however,  they play  a crucial  role  in confinement,
131: furnishing the required long range potential.
132: 
133: 
134: An effective low-energy field 
135: theory for describing the gluon mass   
136: is  the  gauged non-linear sigma model  known  as ``massive
137: gauge-invariant Yang-Mills''~\cite{Cornwall:1979hz}, with    
138: Lagrangian density  
139: \begin{equation}
140: {\cal L}_{MYM}= \frac{1}{2} G_{\mu\nu}^2 - 
141: m^2 {\rm Tr} \left[A_{\mu} - {g}^{-1} U(\theta)\partial_{\mu} U^{-1}(\theta) \right]^2\,,
142: \label{nlsm}
143: \end{equation}
144: where 
145: $A_{\mu}= \frac{1}{2i}\sum_{a} \lambda_a A^{a}_{\mu}$, the $\lambda_a$ are the SU(3) generators
146: (with  ${\rm Tr} \lambda_a  \lambda_b=2\delta_{ab}$), 
147: and the $N\times N$
148: unitary matrix $U(\theta) = \exp\left[i\frac{1}{2}\lambda_a\theta^{a}\right]$ 
149: describes the scalar fields $\theta_a$.  
150: Note that ${\cal L}_{MYM}$ is locally gauge-invariant under the combined gauge transformation 
151: \be
152: A^{\prime}_{\mu} = V A_{\mu} V^{-1} - {g}^{-1} \left[\partial_{\mu}V \right]V^{-1}\,, 
153: \qquad
154: U^{\,\prime} = U(\theta^{\,\prime}) = V U(\theta)\,,
155: \label{gtransfb}
156: \ee
157: for any group matrix $V= \exp\left[i\frac{1}{2}\lambda_a\omega^{a}(x)\right]$, where 
158: $\omega^{a}(x)$ are the group parameters. 
159: One might think that, by employing (\ref{gtransfb}), the fields  $\theta_a$ can always 
160: be transformed to zero, but this is not so if the $\theta_a$ contain vortices.
161: To use the ${\cal L}_{MYM}$ in (\ref{nlsm}), one solves the equations of motion for $U$ 
162: in terms of the gauge potentials and substitutes the result in the equations for the gauge potential.  
163: One then finds the Goldstone-like massless modes mentioned above.
164: This model admits  vortex
165: solutions~\cite{Cornwall:1979hz},  with a  long-range pure  gauge term  in  their potentials,
166: which endows  them with a topological quantum  number corresponding to
167: the center  of the gauge group  [$Z_N$ for $SU(N)$], and  is, in turn,
168: responsible for quark  confinement and gluon screening. 
169: Specifically, center vortices of  thickness $\sim m^{-1}$,  where $m$ is
170: the induced mass of the gluon, form a condensate because their entropy
171: (per  unit  size) is  larger  than  their  action.  This  condensation
172: furnishes an  area law to  the fundamental representation  
173: Wilson loop, thus confining quarks~\cite{Cornwall:1982zr,Cornwall:1979hz}. 
174: 
175: Of course ${\cal L}_{MYM}$ is not renormalizable, and breaks down in the ultraviolet.  
176: This breakdown simply reflects the fact that the gluon mass $m$ in (\ref{nlsm})
177: is assumed to be constant, while the solutions of the SDEs  
178: reveal that the mass is momentum-dependent, vanishing at large $q^2$~\cite{Cornwall:1982zr}.  
179: Specifically, when studying the SDE for the gluon propagator, $\Delta(q^2)$,  
180: one looks for infrared finite solutions, i.e. with  $\Delta^{-1}(0) > 0$. 
181: Such solutions may  
182: be  fitted  by     ``massive''  propagators  of   the form \mbox{$\Delta^{-1}(q^2)  =  q^2  +  m^2(q^2)$}, where  
183: $m^2(q^2)$ is  not ``hard'', but depends non-trivially  on the momentum  transfer $q^2$.
184: When the  renormalization-group 
185: logarithms are  properly taken into  account in the SDE analysis, one obtains,  in addition, 
186: the  non-perturbative  generalization  of  $\alpha(q^2)$, 
187: the  QCD  running  coupling (effective charge).
188: The presence of $m^2(q^2)$ in the argument of $\alpha(q^2)$
189: tames  the   Landau  singularity   associated   with  the
190: perturbative $\beta$  function, and the resulting  effective charge is
191: asymptotically free in  the ultraviolet , ``freezing'' at a  finite value in the infrared.
192: 
193: The general picture described above has received spectacular confirmation from 
194: a plethora of lattice studies, spanning a period of several years:
195: the gluon propagator  reaches indeed a finite (non-vanishing) value in the deep 
196: infrared, as predicted by Cornwall. This  rather characteristic  behavior 
197: was already seen in early studies~\cite{Alexandrou:2000ja}, and 
198: has been  firmly established  recently using
199: large-volume lattices, for both $SU(2)$~\cite{Cucchieri:2007md} and $SU(3)$~\cite{Bogolubsky:2007ud} 
200:  pure  Yang-Mills (no quarks included).
201: 
202: In this talk we will present recent results from a gauge-invariant study of 
203: the coupled gluon-ghost system of SDEs~\cite{Aguilar:2008xm}, yielding  
204: an infrared finite gluon propagator and a divergent (but non-enhanced) 
205: ghost propagator, in qualitative agreement with recent lattice data~\cite{Cucchieri:2007md,Bogolubsky:2007ud};
206:  this behavior has also been confirmed within the Gribov-Zwanziger formalism \cite{Dudal:2008sp}.
207: As the title suggests, we will eventually focus on the issue of the infrared finite QCD effective charge.
208:  
209: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
210: 
211: Obtaining an infrared finite result for the gluon self-energy from SDEs,
212: without violating the underlying local gauge symmetry, is far from trivial,
213: and hinges crucially on one's ability to devise
214: a  self-consistent truncation  scheme  that would  select a  tractable
215: and, at the same time, {\it physically meaningful} subset of these equations.  
216: Specifically, while in QED 
217: the Green's functions satisfy naive 
218: Ward Identities (WIs), in QCD they satisfy complicated 
219: Slavnov-Taylor identities (STIs), which 
220: involve various composite ghost operators. To see how this 
221: complicates the truncation procedure of the SDEs,
222: consider the STI of the gluon self-energy 
223: \be
224: q^{\mu} \Pi_{\mu\nu}(q) = 0 \,.
225: \label{fundtrans}
226: \ee
227: Eq.~(\ref{fundtrans}) is without a doubt the most fundamental 
228: statement at the level of Green's functions that one can obtain
229: from the BRST symmetry; it affirms the transversality of the 
230: gluon self-energy and is valid both  
231: perturbatively to all orders as well as non-perturbatively. 
232: The problem is that 
233: in the SDE governing $\Pi_{\mu\nu}(q)$ enter higher order 
234: Green's functions, 
235: namely the fully-dressed fundamental vertices of the theory, 
236: which satisfy complicated STIs.
237: Thus, whereas in QED the validity of Eq.~(\ref{fundtrans}) 
238: can be easily seen at  
239: the level of the SDE, simply because \mbox{
240: $q^{\mu}\Gamma_{\mu}(p,p+q) =e\left[ S^{-1}(p+q)-S^{-1}(p)\right]$}, in QCD 
241: proving Eq.~(\ref{fundtrans}) is very difficult, and requires the 
242: conspiracy of all full vertices appearing in the SDE.
243: Truncating the SDE naively usually amounts to leaving out some of these vertices, 
244: and, as a result, Eq.~(\ref{fundtrans}) is compromised.
245: Instead, the  gauge-invariant  truncation  scheme \cite{Binosi:2007pi},  
246: based  on  the  PT~\cite{Cornwall:1982zr, Cornwall:1989gv} and  its 
247: correspondence with the background field method (BFM)~\cite{Abbott:1980hw} 
248: maintains the validity of Eq.(\ref{fundtrans}) at every level of approximation.
249: 
250: 
251: The gluon propagator in the covariant gauges has the form
252: \mbox{ 
253: $i\Delta_{\mu\nu}(q)= \left[ {\rm P}_{\mu\nu}(q)\Delta(q^2) +\xi\frac{\D q_\mu
254: q_\nu}{\D q^4}\right]$},
255: where $\xi$ denotes the gauge-fixing parameter, 
256: \mbox{${\rm P}_{\mu\nu}(q)= g_{\mu\nu} - q_\mu q_\nu /q^2$}
257: is the usual transverse projector, and, finally, $\Delta^{-1}(q^2) = q^2 + i \Pi(q^2)$, 
258: with  $\Pi_{\mu\nu}(q)={\rm P}_{\mu\nu}(q) \,\Pi(q^2)$ the gluon self-energy. 
259: The full ghost propagator $D(p^2)$ and its self-energy $L(p^2)$ are related by
260: \mbox{$iD^{-1}(p^2)= p^2 - iL(p^2)$}. In the case of pure (quarkless) QCD,
261: the new SD series~\cite{Binosi:2007pi} for the gluon and ghost propagators reads (see also Fig.~\ref{SDeqs})
262: \bea
263: && \Delta^{-1}(q^2){\rm P}_{\mu\nu}(q) = 
264: \frac{q^2 {\rm P}_{\mu\nu}(q) + i\,\sum_{i=1}^{4}(a_i)_{\mu\nu}}{[1+G(q^2)]^2} \,,
265: \nonumber\\
266: && iD^{-1}(p^2) = p^2 +i \lambda  \int_k
267: \Gamma^{\mu}\Delta_{\mu\nu}(k)\gb^{\nu}(p,k) D(p+k) \,,
268: \nonumber\\
269: && i\Lambda_{\mu \nu}(q) = \lambda 
270: \int_k H^{(0)}_{\mu\rho}
271: D(k+q)\Delta^{\rho\sigma}(k)\, H_{\sigma\nu}(k,q)\,,
272: \label{newSDb}
273: \eea
274: where \mbox{$\lambda=g^2 C_{\rm {A}}$}, with $C_{\rm {A}}$ the Casimir eigenvalue of the adjoint representation
275: [$C_{\rm {A}}=N$ for $SU(N)$], 
276: and \mbox{$\int_{k}\equiv\mu^{2\varepsilon}(2\pi)^{-d}\int\!d^d k$}, 
277: with $d=4-\epsilon$ the dimension of space-time. 
278: $\Gamma_{\mu}$ is the standard (asymmetric) gluon-ghost vertex at tree-level,
279: and $\gb^{\nu}$ the fully-dressed one.
280: $G(q^2)$ is the $g_{\mu\nu}$ component
281: of the auxiliary two-point function $\Lambda_{\mu \nu}(q)$, and the function $H_{\sigma\nu}$ is 
282: defined diagrammatically in Fig.~\ref{SDeqs}.
283: $H_{\sigma\nu}$ is related to the full gluon-ghost vertex by 
284: \mbox{$q^{\sigma} H_{\sigma\nu}(p,r,q) = -i{\gb}_{\nu}(p,r,q)$}; 
285: at tree-level, $H_{\sigma\nu}^{(0)} = ig_{\sigma\nu}$.
286: When evaluating  the diagrams $(a_i)$  we use the  BFM Feynman
287: rules~\cite{Abbott:1980hw}; the BFM fully dressed three-gluon and
288: gluon-ghost vertices are denoted by ${\g}_{\mu\alpha\beta}$ and ${\g}_{\mu}$.
289: 
290: 
291: % vertices will be denoted by
292: %${\g}_{\mu\alpha\beta}$ (three-gluon vertex), and ${\g}_{\mu}$ (gluon-ghost vertex). 
293: 
294: %%%%%%%%%%%%%%%Fig.1%%%%%%%%
295: \begin{figure}
296: \includegraphics[width=14cm]{SDeqs.eps}
297: \caption{The PT-BFM gluon-ghost system. The circles attached 
298: to the external gluons denote that, 
299: from the point of view of Feynman rules, they are treated as background fields.}
300: \label{SDeqs}
301: \end{figure}
302: %%%%%%%%%%%%%%%%%%%%%%%%%%
303: 
304: 
305: Notice a point of paramount importance:
306: due to the Abelian all-order WIs that these two full 
307: vertices  satisfy (for all $\xi$), namely
308: \bea
309: q^{\mu}{\g}_{\mu\alpha\beta} = i\Delta^{-1}_{\alpha\beta}(k+q)
310: -i\Delta^{-1}_{\alpha\beta}(k)\,,\qquad
311: q^{\mu}{\g}_{\mu} = iD^{-1}(k+q) - iD^{-1}(k),
312: \label{PTWI}
313: \eea
314: one can demonstrate that 
315: $q^{\mu}[(a_1)+(a_2)]_{\mu\nu} =0$
316: and \mbox{$q^{\mu}[(a_3)+(a_4)]_{\mu\nu} =0$} \cite{Aguilar:2006gr}.
317: Thus, unlike other treatments in the literature, within this formalism 
318: the transversality of the gluon self-energy, i.e. Eq.~(\ref{fundtrans}), 
319: is preserved at every step, in absolute compliance with the BRST symmetry. 
320: 
321: 
322: Next, following standard techniques,
323: we express ${\g}_{\mu\alpha\beta}$ and ${\g}_{\mu}$
324: as a function of the gluon and ghost self-energy, respectively, in such a way as to 
325: automatically satisfy the crucial WIs of  Eq.~(\ref{PTWI}); failure to satisfy these WIs 
326: would invariably compromise the transversality of the answer.  
327: The Ansatz we will use is 
328: 
329: \bea
330: {\g}_{\mu\alpha\beta}= \Gamma_{\mu\alpha\beta} + i\frac{q_{\mu}}{q^2}
331: \left[\Pi_{\alpha\beta}(k+q)-\Pi_{\alpha\beta}(k)\right]\nonumber \,,  \qquad
332: {\g}_{\mu}= \widetilde{\Gamma}_{\mu} -i\frac{q_{\mu}}{q^2}
333: \left[L(k+q)-L(k)\right]\,;
334: \label{gluonv}
335: \eea
336: its essential feature, other than satisfying the aforementioned WIs, 
337: is the presence of massless, longitudinally coupled   
338: pole terms, which are instrumental for obtaining 
339: $\Delta^{-1}(0)\neq 0$~\cite{Jackiw:1973tr}. These poles are not kinematic but dynamical, 
340: corresponding to a  composite (bound-state) Goldstone excitation,  enforcing the local gauge invariance.
341: For the conventional ghost-gluon vertex ${\gb}_{\nu}$,
342: appearing in the second SDE of (\ref{newSDb}) we will use its
343: tree-level expression, {\it i.e.}, ${\gb}_{\nu}\to \Gamma_{\nu} = -p_{\nu}$;
344: this is perfectly legitimate, since in this formalism 
345: the two ghost vertices,  ${\gb}_{\nu}$ and ${\g}_{\mu}$, are different.
346: Finally, for $H_{\sigma\nu}$ we use its tree-level value, $H_{\sigma\nu}^{(0)}$.
347: 
348: 
349: In Fig.\ref{figb}, we show the numerical result for  $\Delta(q^2)$  
350: renormalized at $\mu=M_b=4.5\,\mbox{GeV}$, and the comparison 
351: with the corresponding lattice data of Ref.\cite{Bogolubsky:2007ud}.
352: In the right panel of Fig.\ref{figb}, we present the dressing function 
353: for the ghost propagator, renormalized at the same point.
354: \vspace{-1.5cm}
355: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
356: \begin{figure}[ht]
357: \hspace{-1.5cm}
358: \includegraphics[scale=2.0]{propagators.eps}
359: \vspace{-2.0cm}
360: \caption{Left Panel:The numerical solution for the gluon propagator from the SDE
361: (black continuous line) compared to the 
362: lattice data of Ref.\cite{Bogolubsky:2007ud}. Right panel:The ghost dressing function $p^2D(p^2)$ 
363: obtained from the SDE. In the insert we show  
364: the lattice data for the same quantity; notice that there is no ``enhancement''\cite{Boucaud:2008ky}.}
365: \label{figb}
366: \end{figure}
367: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
368: 
369: 
370: 
371: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
372: 
373: Let us now turn to the QCD effective charge. There are two main issues:
374: (i) how to define it consistently at the level of perturbation theory:
375: specifically, which graphs determine the running, and 
376: (ii) how to extend the (whatever) definition one reaches in (i) 
377: into the non-perturbative regime.
378: 
379: 
380: %
381: 
382: Point (i) has been addressed exhaustively in the literature~\cite{Watson:1996fg}: the upshot  is that in the 
383: context of the PT one may replicate to all-orders in perturbation theory the prototype QED construction 
384: of an effective charge. To fix the ideas,  the PT one-loop gluon self-energy 
385: reads
386: \be 
387: \widehat\Delta^{-1}(q^2)= q^2\left[1+ b g^2\ln\left(\frac{q^2}{\mu^2}\right)\right]\,,
388: \label{rightRG}
389: \ee
390: where  $b = 11 C_A/48\pi^2$  is the first coefficient of the QCD $\beta$-function. 
391: Due to the Abelian WIs satisfied by the PT effective Green's functions, the 
392:  new propagator-like quantity $\widehat\Delta^{-1}(q^2)$ absorbs all  
393: the RG-logs, exactly as happens in QED with the photon self-energy.
394: Equivalently, since $Z_{g}$ and ${\widehat Z}_{A}$, the renormalization constants 
395: of the gauge-coupling and the effective self-energy, respectively, 
396: satisfy the QED relation ${Z}_{g} = {\widehat Z}^{-1/2}_{A}$,  
397: the product 
398: ${\widehat d}(q^2) = g^2 \widehat\Delta(q^2)$ forms a RG-invariant 
399: ($\mu$-independent) quantity~\cite{Cornwall:1982zr};
400: for large momenta $q^2$,
401: \be
402: {\widehat d}(q^2) = \frac{\overline{g}^2(q^2)}{q^2}\,,
403: \label{ddef1}
404: \ee
405: where $\overline{g}^2(q^2)$ is the RG-invariant effective charge of QCD,
406: \be
407: \overline{g}^2(q^2) = \frac{g^2}{1+  b g^2\ln\left(q^2/\mu^2\right)}
408: = \frac{1}{b\ln\left(q^2/\Lambda^2\right)}\,.
409: \label{effch}
410: \ee
411: %
412: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
413: \begin{figure}[ht]
414: \vspace{-1.5cm}
415: \hspace{-2.0cm}
416: \includegraphics[scale=2.0]{rgi_procedure3.eps}
417: \vspace{-1.2cm}
418: \caption{Left panel: The solution of SDE renormalized at \mbox{$\mu=M_b=4.5\,\mbox{GeV}$}
419: (continuous blue curve) and  $\mu=M_Z=91\,\mbox{GeV}$ (red line+square curve). Right Panel: The
420:  corresponding PT-BFM $\widehat{\Delta}(q^2)$ obtained as the 
421: convolution of $\Delta(q^2)$  and the function $g^2(\mu^2)/[1+G(q^2)]^2$. }
422: \label{rgi_fig}
423: \end{figure}
424: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
425: %
426: Let us now come to point (ii): assuming that one has non-perturbative 
427: information about the infrared behavior of the conventional gluon propagator $\Delta(q^2)$, 
428: how should one extract an effective charge, which, perturbatively, 
429: will go over to Eq.~(\ref{effch})? To accomplish this, one must use an additional 
430: field-theoretic ingredient:
431: the conventional $\Delta(q^2)$ and the PT-BFM $\widehat{\Delta}(q^2)$
432: are related by the formal all-order relation~\cite{Binosi:2002ez} 
433: \be
434: \Delta(q^2) = 
435: \left[1+G(q^2)\right]^2 \widehat{\Delta}(q^2) \,.
436: \label{bqi}
437: \ee
438: Note that the  $G(q^2)$  already appears in  Eq.~(\ref{newSDb}) and Fig.\ref{SDeqs}.
439: With our approximations its SDE reads
440: \be
441: G(q^2) = - \frac{\lambda}{3}\int_k\,\left[
442: 2+\frac{(k\cdot q)^2}{k^2q^2}\right]\Delta(k)D(k+q) \,.
443: \label{gg}
444: \ee
445: %
446: 
447: First of all, it is easy to verify that at lowest order 
448: the $G(q^2)$ obtained from  Eq.~(\ref{gg})  restores the $\beta$ function coefficient  
449: in front of ultraviolet logarithm. In that limit 
450: \mbox{$1+G(q^2) = 1 +\frac{9}{4}
451: \frac{C_{\rm {A}}g^2}{48\pi^2}\ln(q^2/\mu^2) $} and \mbox{$\Delta^{-1}(q^2)= q^2 \left[1+\frac{13}{2}
452: \frac{C_{\rm {A}}g^2}{48\pi^2}\ln(q^2/\mu^2)\right]$}. Then using  
453: Eq.~(\ref{bqi}) we recover the $\widehat{\Delta}^{-1}(q^2)$ of Eq.~(\ref{rightRG}), as we should. 
454: Then, non-perturbatively, one substitutes into  Eq.~(\ref{bqi}) the $G(q^2)$ and $\Delta(q^2)$ 
455: obtained from solving the system in  Eq.~(\ref{newSDb}), to obtain $\widehat{\Delta}(q^2)$.  
456: This latter quantity is the non-perturbative generalization of Eq.~(\ref{rightRG}); 
457: for the same reasons explained above, when multiplied by $g^2$ it should form an RG-invariant quantity, e.g. the non-perturbative 
458: generalization of ${\widehat d}(q^2)$. In Fig.\ref{rgi_fig} we present the combined result of the above steps: 
459: ${\widehat d}(q^2)$ is obtained from two different sets of solutions of the system  Eq.~(\ref{newSDb}), one  
460: renormalized at $\mu=M_b=4.5\,\mbox{GeV}$ and one at $\mu=M_Z=91\,\mbox{GeV}$. Ideally
461: the two curves 
462: of ${\widehat d}(q^2)$ should be identical; even though this does not happen, due to the approximations employed 
463: when  solving the system of Eq.~(\ref{newSDb}), the two curves are fairly close, 
464: indicating that ${\widehat d}(q^2)$ is to a very good approximation an RG-invariant quantity, as it should.
465: 
466: We are now in the position to define the non-perturbative 
467: QCD effective charge from the RG-invariant quantity ${\widehat d}(q^2)$. Of course, given that 
468: ${\widehat d}(q^2)$ reaches a finite value in the deep infrared, 
469: it would be completely absurd to define the effective charge 
470: by  forcing out a factor of $1/q^2$; such a procedure would furnish a completely unphysical strong QCD coupling, 
471: namely one that would vanish in the deep infrared(!) where QCD is supposed to be strongly coupled.
472: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
473: \vspace{-1.1cm}
474: \begin{figure}
475: \begin{minipage}[c]{.45\textwidth}
476: \hspace{-1.0cm}
477: \includegraphics[scale=0.8]{effective_charge.eps}
478: \vspace{-1.4cm}
479: \begin{center}
480: \caption{The QCD effective charge, \mbox{$\alpha(q^2)=\overline{g}^2(q^2)/4\pi$},
481: extracted from Fig.3 by factoring out a gluon mass of 
482: \mbox{m(0)=500\, \mbox{MeV}}.}
483: \label{figc}
484: \end{center}
485: \end{minipage}
486: \hspace{\fill}
487: \begin{minipage}[l]{.5\textwidth}
488: {The correct thing to do is to factor out a ``massive'' propagator, i.e. write 
489: %
490: \be
491: \widehat{d}(q^2) = \frac{\overline{g}^2(q^2)}{q^2 + m^2(q^2)} \,.
492: \label{ddef}
493: \ee
494: 
495: Of course, as we have emphasized,  $m^2(q^2)$ itself is running, which must also 
496: be taken into account in a more sophisticated treatment.
497: For the purposes of this talk, however, we assume that 
498: $m^2(q^2)$ is constant, $m^2(q^2)=m^2(0)$, and use for $m(0)$ the value of $500\, \mbox{MeV}$ 
499: favored by phenomenology~\cite{Aguilar:2001zy}. 
500: The $\alpha(q^2)$ obtained is shown in Fig.(\ref{figc}); as announced, 
501: at low energies freezes to a finite value, indicating the appearance of an infrared fixed point of QCD.
502:  }
503: \end{minipage}
504: \end{figure}
505: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 
506: %
507: 
508: \acknowledgments
509:  We thank Prof. J.~M.~Cornwall for numerous illuminating discussions.
510: Work supported by the Spanish MEC grants FPA 2005-01678 and the Fundaci\'on General of the UV. 
511:  
512: \vspace{-0.4cm}
513: 
514: \begin{thebibliography}{99}
515: 
516: \vspace{-0.2cm}
517: 
518: %\cite{Cornwall:1982zr}
519: \bibitem{Cornwall:1982zr}
520: J.~M.~Cornwall,
521: %``Dynamical Mass Generation In Continuum QCD,''
522: Phys.\ Rev.\ D {\bf 26}, 1453 (1982). 
523: %%CITATION = PHRVA,D26,1453;%%
524: 
525: %\cite{Cornwall:1989gv}
526: \bibitem{Cornwall:1989gv}
527: J.~M.~Cornwall and J.~Papavassiliou,
528: %``Gauge Invariant three-gluon Vertex In QCD,''
529: Phys.\ Rev.\ D {\bf 40}, 3474 (1989).
530: %%CITATION = PHRVA,D40,3474;%%
531: 
532: 
533: %\cite{Cornwall:1979hz}
534: \bibitem{Cornwall:1979hz}
535:   J.~M.~Cornwall,
536:   %``Quark Confinement And Vortices In Massive Gauge Invariant QCD,''
537:   Nucl.\ Phys.\  B {\bf 157}, 392 (1979);
538:   %%CITATION = NUPHA,B157,392;%%
539: %\cite{Cornwall:1997ds}
540: %\bibitem{Cornwall:1997ds}
541: %  J.~M.~Cornwall,
542:   %``Center vortices and confinement vs. screening,''
543:   Phys.\ Rev.\  D {\bf 57}, 7589 (1998).
544: %  [arXiv:hep-th/9712248].
545:   %%CITATION = PHRVA,D57,7589;%%
546: 
547: 
548: 
549: %\cite{Alexandrou:2000ja}
550: \bibitem{Alexandrou:2000ja}
551:   C.~Alexandrou, P.~de Forcrand and E.~Follana,
552:   %``The gluon propagator without lattice Gribov copies,''
553:   Phys.\ Rev.\  D {\bf 63}, 094504 (2001), 
554: %  [arXiv:hep-lat/0008012].
555:   %%CITATION = PHRVA,D63,094504;%%
556: 
557: 
558: %\cite{Cucchieri:2007md}
559: \bibitem{Cucchieri:2007md}
560:   A.~Cucchieri and T.~Mendes,
561:   %``What's up with IR gluon and ghost propagators in Landau gauge? A puzzling
562:   %answer from huge lattices,''
563:   PoS {\bf LAT2007}, 297 (2007).
564: %  [arXiv:0710.0412 [hep-lat]].
565:   %%CITATION = POSCI,LAT2007,297;%%
566: 
567: 
568: %\cite{Bogolubsky:2007ud}
569: \bibitem{Bogolubsky:2007ud}
570:   I.~L.~Bogolubsky, E.~M.~Ilgenfritz, M.~Muller-Preussker and A.~Sternbeck,
571:   %``The Landau gauge gluon and ghost propagators in 4D SU(3) gluodynamics in
572:   %large lattice volumes,''
573:   PoS {\bf LAT2007}, 290 (2007).
574: %  [arXiv:0710.1968 [hep-lat]].
575:   %%CITATION = POSCI,LAT2007,290;%%
576: 
577: 
578: 
579: 
580: 
581: %\cite{Aguilar:2008xm}
582: \bibitem{Aguilar:2008xm}
583:   A.~C.~Aguilar, D.~Binosi and J.~Papavassiliou,
584:   %``Gluon and ghost propagators in the Landau gauge: Deriving lattice results
585:   %from Schwinger-Dyson equations,''
586:   Phys.\ Rev.\  D {\bf 78}, 025010 (2008).
587: %  [arXiv:0802.1870 [hep-ph]].
588:   %%CITATION = PHRVA,D78,025010;%%
589:   
590:   
591:   
592:   %\cite{Dudal:2008sp}
593: \bibitem{Dudal:2008sp}
594:   D.~Dudal, J.~A.~Gracey, S.~P.~Sorella, N.~Vandersickel and H.~Verschelde,
595:   %``A refinement of the Gribov-Zwanziger approach in the Landau gauge: infrared
596:   %propagators in harmony with the lattice results,''
597:   arXiv:0806.4348 [hep-th].
598:   %%CITATION = ARXIV:0806.4348;%%
599: 
600: 
601:  
602: 
603: %\cite{Binosi:2007pi}
604: \bibitem{Binosi:2007pi}
605:   D.~Binosi and J.~Papavassiliou,
606:   %``Gauge-invariant truncation scheme for the Schwinger-Dyson equations of
607:   %QCD,''
608:   Phys.\ Rev.\  D {\bf 77}, 061702 (2008);
609:   %%CITATION = PHRVA,D77,061702;%%
610: %\cite{Binosi:2008qk}
611: %\bibitem{Binosi:2008qk}
612: %  D.~Binosi and J.~Papavassiliou,
613:   %``New Schwinger-Dyson equations for non-Abelian gauge theories,''
614: arXiv:0805.3994 [hep-ph].
615:   %%CITATION = ARXIV:0805.3994;%%
616: 
617: 
618: %\cite{Abbott:1980hw}
619: \bibitem{Abbott:1980hw}
620: L.~F.~Abbott,
621: %``The Background Field Method Beyond One Loop,''
622: Nucl.\ Phys.\ B {\bf 185}, 189 (1981).
623: %%CITATION = NUPHA,B185,189;%%
624: 
625: 
626: %\cite{Aguilar:2006gr}
627: \bibitem{Aguilar:2006gr}
628:   A.~C.~Aguilar and J.~Papavassiliou,
629:   %``Gluon mass generation in the PT-BFM scheme,''
630:   JHEP {\bf 0612}, 012 (2006).
631: %  [arXiv:hep-ph/0610040].
632:   %%CITATION = JHEPA,0612,012;%%
633: 
634: 
635: %\cite{Jackiw:1973tr}
636: \bibitem{Jackiw:1973tr}
637:   R.~Jackiw and K.~Johnson,
638:   %``Dynamical Model Of Spontaneously Broken Gauge Symmetries,''
639:   Phys.\ Rev.\ D {\bf 8}, 2386 (1973);
640:   %%CITATION = PHRVA,D8,2386;%%
641: %\cite{Cornwall:1973ts}
642: %\bibitem{Cornwall:1973ts}
643:   J.~M.~Cornwall and R.~E.~Norton,
644:   %``Spontaneous Symmetry Breaking Without Scalar Mesons,''
645:   Phys.\ Rev.\ D {\bf 8} (1973) 3338;
646:   %%CITATION = PHRVA,D8,3338;%%
647: %\cite{Eichten:1974et}
648: %\bibitem{Eichten:1974et}
649:   E.~Eichten and F.~Feinberg,
650:   %``Dynamical Symmetry Breaking Of Nonabelian Gauge Symmetries,''
651:   Phys.\ Rev.\ D {\bf 10}, 3254 (1974).
652:   %%CITATION = PHRVA,D10,3254;%%
653: 
654: 
655: %\cite{Boucaud:2008ky}
656: \bibitem{Boucaud:2008ky}
657:   Ph.~Boucaud {\it et al.}, 
658:   %``On the IR behaviour of the Landau-gauge ghost propagator,''
659:   JHEP {\bf 0806}, 099 (2008);
660: % [arXiv:0803.2161 [hep-ph]].
661:   %%CITATION = JHEPA,0806,099;%%
662: %\cite{Boucaud:2006if}
663: %\bibitem{Boucaud:2006if}
664:   Ph.~Boucaud {\it et al.},
665:   %``Is the QCD ghost dressing function finite at zero momentum?,''
666:   JHEP {\bf 0606}, 001 (2006).
667: %  [arXiv:hep-ph/0604056].
668:   %%CITATION = JHEPA,0606,001;%%
669: 
670: 
671: 
672: %\cite{Watson:1996fg}
673: \bibitem{Watson:1996fg}
674:   N.~J.~Watson,
675:   %``The gauge-independent QCD effective charge,''
676:   Nucl.\ Phys.\  B {\bf 494}, 388 (1997);
677: %[arXiv:hep-ph/9606381].
678:   %%CITATION = NUPHA,B494,388;%%
679: %\cite{Binosi:2002vk}
680: %\bibitem{Binosi:2002vk}
681:   D.~Binosi and J.~Papavassiliou,
682:   %``The QCD effective charge to all orders,''
683:   Nucl.\ Phys.\ Proc.\ Suppl.\  {\bf 121}, 281 (2003).
684: %  [arXiv:hep-ph/0209016].
685:   %%CITATION = NUPHZ,121,281;%%
686: 
687: 
688: 
689: %\cite{Binosi:2002ez}
690: \bibitem{Binosi:2002ez}
691:   D.~Binosi and J.~Papavassiliou,
692:   %``Pinch technique and the Batalin-Vilkovisky formalism,''
693:   Phys.\ Rev.\ D {\bf 66}, 025024 (2002).
694: %  [arXiv:hep-ph/0204128].
695:   %%CITATION = HEP-PH 0204128;%%
696: 
697: 
698: 
699: %\cite{Aguilar:2001zy}
700: \bibitem{Aguilar:2001zy}
701:   A.~C.~Aguilar, A.~Mihara and A.~A.~Natale,
702:   %``Freezing of the QCD coupling constant and solutions of Schwinger-Dyson
703:   %equations,''
704:   Phys.\ Rev.\  D {\bf 65}, 054011 (2002).
705: %  [arXiv:hep-ph/0109223].
706:   %%CITATION = PHRVA,D65,054011;%%
707: 
708: 
709: \end{thebibliography} 
710: \end{document} 
711: 
712: 
713: 
714: 
715: 
716: 
717: 
718: