1:
2:
3: \documentclass[aps,nofootinbib,showpacs,showkeys,preprintnumbers,onecolumn,superscriptaddress]{revtex4}
4: % \documentstyle[prl,preprint,tighten,aps,12pt,times,epsfig]{revtex}
5: \usepackage{graphicx}
6: \usepackage{dcolumn}
7: \usepackage{amsmath}
8: \usepackage{colordvi}
9: \newcommand{\UMLAUT}[1]{\"{#1}}
10: \newcommand{\tr}{\mbox{\rm tr}}
11: \newcommand{\eV}{\mbox{\rm eV}}
12: \newcommand{\keV}{\mbox{\rm keV}}
13: \newcommand{\MeV}{\mbox{\rm MeV}}
14: \newcommand{\GeV}{\mbox{\rm GeV}}
15: \newcommand{\indexentry}[2]{\item{{\bf #1}{, #2}}}
16: \newcommand{\lsim}[1]{
17: \setlength{\unitlength}{10pt}
18: \begin{picture}(1.4,1.)
19: \put(.7,-0.3){\makebox(0.0,1.)[t]{$<$}}
20: \put(.7,-0.3){\makebox(0.0,1.)[b]{$\sim$}}
21: \end{picture}#1}
22: \newcommand{\gsim}[2]{
23: \setlength{\unitlength}{10pt}
24: \begin{picture}(1.4,1.)
25: \put(.7,-0.3){\makebox(0.0,1.)[t]{$>$}}
26: \put(.7,-0.3){\makebox(0.0,1.)[b]{$\sim$}}
27: \end{picture}#2}
28: \renewcommand{\thesection}{\arabic{section}}
29:
30: \begin{document}
31:
32:
33: % \date{January 31, 2005.}
34: \title{Circumventing the axial
35: anomalies and the strong CP problem}
36:
37: \author{Dalibor Kekez}
38: \affiliation{\footnotesize Rudjer Bo\v{s}kovi\'{c} Institute,
39: P.O.B. 180, 10002 Zagreb, Croatia}
40:
41: \author{Dubravko Klabu\v{c}ar\footnote{Senior Associate of Abdus Salam ICTP}}
42: \affiliation{\footnotesize Physics Department, Faculty of Science,
43: University of Zagreb, P.O.B. 331, 10002 Zagreb, Croatia}
44:
45: \author{M. D. Scadron}
46: \affiliation{\footnotesize Physics Department, University of Arizona,
47: Tucson Az 85721 USA}
48:
49:
50: \begin{abstract}
51: Many meson processes are related to the $U_A(1)$
52: axial anomaly, present in the Feynman graphs where
53: fermion loops connect axial vertices with vector
54: vertices.
55: However, the coupling of pseudoscalar mesons to quarks
56: does not have to be formulated via axial vertices.
57: The pseudoscalar coupling is also possible, and this
58: approach is especially natural on the level of
59: the quark substructure of hadrons.
60: In this paper we point out the advantages of calculating
61: these processes using (instead of the anomalous graphs)
62: the graphs where axial vertices are replaced by
63: pseudoscalar vertices.
64: We elaborate especially the case of the processes related to the
65: Abelian axial anomaly of QED, but we speculate that
66: it seems possible that effects of the non-Abelian
67: axial anomaly of QCD can be accounted for in an
68: analogous way.
69: \end{abstract}
70: \pacs{14.40 -n, 12.39.Fe, 13.20.-v, 11.10.St}
71: \keywords{axial anomaly, quark loops, radiative and hadronic decays of mesons}
72:
73: \maketitle
74:
75: \section{Introduction}
76:
77:
78: Numerous processes in meson physics are related to the
79: Adler-Bell-Jackiw (ABJ) axial anomaly \cite{Adler69,BellJackiw69}
80: appearing in the fermion loops connecting certain number of axial (A)
81: and vector (V) vertices.
82: Concretely, in this paper we will deal with the processes
83: related to the AVV (``triangle", Fig. 1) and VAAA
84: (``box", Fig. 2) anomaly,
85: exemplified by the famous $\pi^0 \to \gamma\gamma$ and
86: $\gamma \to \pi^+ \pi^0 \pi^-$ transitions.
87:
88: \begin{figure}
89: \includegraphics[height=60mm,angle=0]{triangle.eps}
90: \caption{The triangle graph and its crossed graph
91: relevant for the interaction of the neutral pseudoscalar meson
92: of momentum $P$ with two photons of momenta $k$ and $k^\prime$.
93: The quark-photon coupling is in general given by dressed vector
94: vertices $\Gamma_\mu(q_1,q_2)$, which in the free limit reduce
95: to $e {\cal Q} \gamma_\mu$.}
96: \label{fig:triangle}
97: \end{figure}
98:
99: \begin{figure}
100: \includegraphics[height=70mm,angle=0]{box.eps}
101: \caption{One of the box diagrams for the process
102: $\gamma \to \pi^+ \pi^0 \pi^-$.
103: There are six different contributing graphs, obtained from Fig.~\ref{fig:box}
104: by the permutations of the vertices of the three different pions.
105: The position of the $u$ and $d$
106: quark flavors on the internal lines, as well as $Q_u$ or $Q_d$
107: quark charges in the quark-photon vertex, varies from graph to
108: graph, depending on the position of the quark-pion vertices.
109: The physical pion fields are $\pi^\pm=(\pi_1\mp i \pi_2)/\sqrt{2}$
110: and $\pi^0\equiv\pi_3$. Thus, in Eq. (\ref{1Leff}) one has
111: $\pi_a \tau_a = \sqrt{2}(\pi^+\tau_+ + \pi^- \tau_-) + \pi^0 \tau_3$
112: where $\tau_{\pm}=(\tau_1 \pm i\tau_2$)/2.
113: The momenta flowing through the four sections of the quark loop
114: are conveniently given by various combinations of the symbols
115: $\alpha, \beta, \gamma = +, 0, -$ in $k_{\alpha\beta\gamma} \equiv
116: k + \alpha p_{1} + \beta p_{2} + \gamma p_{3}$.}
117: \label{fig:box}
118: \end{figure}
119:
120: Suppose one wants to describe such processes using QCD-related
121: effective chiral meson Lagrangians \cite{seeGeorgi,Georgi:1985kw}
122: without adding ad hoc interactions of mesons with external
123: gauge fields to reproduce empirical results. For example, one
124: can add by hand
125: \begin{equation}
126: \Delta {\cal L} = g_{\pi\gamma\gamma} \pi^0
127: \epsilon_{\mu\nu\rho\sigma} F^{\mu\nu} F^{\rho\sigma} \, ,
128: \label{L_pi2gamma}
129: \end{equation}
130: and this would reproduce the observed $\pi^0 \to \gamma\gamma$
131: width for the favorable value of the $\pi^0\gamma\gamma$
132: coupling $g_{\pi\gamma\gamma}$. However, if one does not want
133: to add such ad hoc terms in the effective meson Lagrangians,
134: one must describe such ``anomalous'' processes through the term
135: derived by Wess and Zumino (WZ) \cite{WZ}. On the other hand, if one
136: wants to utilize and explicitly take into account the fact that
137: mesons are composed of quarks, another way of describing these
138: processes is optimal in our opinion, and the main purpose of this
139: paper is to stress and elucidate this.
140:
141: Axial vertices in the anomalous graphs such as the AVV and VAAA ones,
142: couple the quarks with pseudoscalar mesons.
143: Instead of anomalous graphs, another way to study the related
144: amplitudes involving pseudoscalar mesons, is to calculate the
145: corresponding graphs where axial vertices (A) are replaced by
146: pseudoscalar (P) ones. Thereby, for example, the
147: $\pi^0 \to \gamma\gamma$
148: decay amplitude due to the AVV ``triangle anomaly",
149: \begin{equation}
150: F_{m_\pi=0}(\pi^0\to 2\gamma)
151: = \frac{e^2 N_c}{12\pi^2f_\pi}~,
152: \label{pi2gammaAmp}
153: \end{equation}
154: is reproduced by the calculation of the PVV triangle graph.
155: [Eq. (\ref{pi2gammaAmp}) pertains to the chiral limit, where
156: the pion mass $m_\pi=0$. Also, $f_\pi\approx 93~\MeV$
157: is the pion decay constant,
158: $e$ is the proton charge, and $N_c=3$ is the number of quark colors.]
159: A survey of this P coupling method is given in Sec.~\ref{sec:SurveyP}.
160:
161: The PVV triangle graph calculation of Eq. (\ref{pi2gammaAmp})
162: can most simply be done essentially {\it {\` a} la} Steinberger
163: \cite{S49}, that is,
164: with a loop of ``free'' constituent quarks with the point
165: pseudoscalar coupling (i.e., $g \gamma_5$, where $g = constant$)
166: to quasi-elementary pion fields.
167: However, since the development of the Dyson-Schwinger (DS)
168: approach to quark-hadron physics \cite{Alkofer:2000wg,Maris:2003vk},
169: the presently advocated method becomes even more convincing.
170: Namely, the DS approach clearly shows how the light pseudoscalar
171: mesons simultaneously appear both as quark-antiquark ($q\bar q$)
172: bound states and as Goldstone bosons of the dynamical chiral
173: symmetry breaking (D$\chi$SB) of nonperturbative QCD. The
174: solutions of Bethe-Salpeter (BS) equations for the bound-state
175: vertices of pseudoscalar mesons then enter in the PVV triangle
176: graph instead of the point $g \gamma_5$ coupling, and the
177: current algebra result (\ref{pi2gammaAmp}) is again reproduced
178: exactly and analytically, which is unique among the bound-state
179: approaches. That the (almost massless) pseudoscalars are
180: (quasi-)Goldstone bosons, is also a unique feature among
181: the bound-state approaches to mesons.
182:
183: A reason why the P-coupling method is simpler both technically
184: and conceptually is that the PVV triangle graph amplitude is
185: {\em finite}, unlike the AVV one, which is divergent and therefore also
186: ambiguous with respect to the momentum routing.
187: Also, the PVV quark triangle amplitude leads to many (over 15)
188: decay amplitudes in agreement with data to within 3\% and not
189: involving free parameters \cite{Delbourgo,delbourgo95,bramon98}.
190: This will be elaborated
191: in more detail in Sec. \ref{manyProc}. Additional advantages
192: of this method is that its treatment of the $\eta$-$\eta'$ complex
193: and resolution of the $\rm U_A(1)$ problem, goes well with the
194: absence of axions (which were predicted to solve the strong CP
195: problem but have {\it not} yet been observed \cite{PDG2004}) and
196: with the arguments of Ref. \cite{Mitra}, that there is really
197: no strong CP problem. All this will be discussed in Sec. 4.
198: We state our conclusions in Sec.~\ref{sec:summary}.
199:
200: However, we will first give, in the next section, a more
201: detailed discussion of the P-coupling method and why is that it
202: is equivalent to the anomaly calculations. We illustrate this on
203: the examples of the well-known decay $\pi^0 \to \gamma \gamma$
204: and processes of the type $\gamma \to \pi^+ \pi^0 \pi^-$.
205:
206:
207:
208: \section{Survey of the P-coupling method }
209: \label{sec:SurveyP}
210:
211: The analysis of the Abelian ABJ axial anomaly \cite{Adler69,BellJackiw69}
212: shows that the $\pi^0\to\gamma\gamma$ amplitude in the chiral and
213: soft limit of pions of vanishing mass $m_\pi$,
214: $F_{m_\pi=0}(\pi^0\to 2\gamma)$,
215: is exactly given by Eq. (\ref{pi2gammaAmp}). This anomaly is relevant
216: also for some other process, including some which are even not given
217: by the three-point functions. Notably, the amplitude for the anomalous
218: processes of the type $\gamma \to \pi^+ \pi^0 \pi^-$ is related to
219: $F_{m_\pi=0}(\pi^0\to 2\gamma)$
220: and is given \cite{Ad+al71Te72Av+Z72} by
221: \begin{equation}
222: F_\gamma^{3\pi}(0,0,0) \, =
223: \, \frac{1}{e f_\pi^2} \,
224: F_{m_\pi=0}(\pi^0\to 2\gamma) \, =
225: \, \frac{e N_c}{12 \pi^2 f_\pi^3} \, .
226: \label{g3piAnomAmp}
227: \end{equation}
228: The arguments of the anomalous amplitude (\ref{g3piAnomAmp}),
229: namely the momenta $\{ p_1,p_2,p_3 \}$ of the three pions
230: $\{\pi^+,\pi^0, \pi^-\}$,
231: are all set to zero, because Eq.~(\ref{g3piAnomAmp}) is also a soft
232: limit and chiral limit result, giving the form factor
233: $F_\gamma^{3\pi}(p_1,p_2,p_3)$ at the soft point.
234:
235:
236: \subsection{Point coupling of mesons to loops of
237: simple constituent quarks}
238:
239: Suppose that the relevant fermion propagators are the ones
240: of the effectively free constituent quarks,
241: \begin{equation}
242: S(k) = \frac{1}{\rlap{$k$}/ - M}~,
243: \label{freeSq}
244: \end{equation}
245: where $M$ is a constant effective constituent quark mass parameter.
246: Then the simple ``free" quark loop (QL) calculation of the PVV
247: ``triangle" graph also reproduces successfully the chiral-limit
248: $\pi^0 \to \gamma\gamma$ amplitude
249: $F_{m_\pi=0}(\pi^0\to 2\gamma)$,
250: provided one uses the quark-level Goldberger-Treiman (GT) relation
251: \begin{equation}
252: \frac{g}{M} = \frac{1}{f_{\pi}}
253: \label{GTrel}
254: \end{equation}
255: to express the effective constituent quark mass
256: $M$ and quark-pion coupling strength $g$ in terms of the pion decay
257: constant $f_{\pi}$. (Recall that the Goldstone boson coupling in
258: the Wess-Zumino term is proportional to $1/f_{\pi}$.)
259: The analogous treatment of the VPPP ``box'' graph, Fig. 2, gives
260: the amplitude $F_\gamma^{3\pi}(0,0,0)$ in Eq. (\ref{g3piAnomAmp}).
261:
262: These calculations (essentially
263: {\it {\` a} la} Steinberger \cite{S49}) is the same as the lowest
264: (one-loop) order calculation \cite{BellJackiw69} in the
265: quark--level $\sigma$-model which was constructed
266: to realize current algebra explicitly \cite{HakiogluAndOthers}.
267: By ``free'' quarks we mean that there are no interactions
268: between the effective constituent quarks in the loop, while they
269: {\it do} couple to external fields, presently the photons $A_\mu$
270: and the pion $\pi_a$. Our effective QL model Lagrangian is thus
271: \begin{equation}
272: {\mathcal{L}}_{\mathit{eff}} =
273: \overline\Psi\left( i \partial \hskip-0.55em{\slash}
274: - e {\cal Q} A \hskip-0.5em\slash -M\right)\Psi
275: - i \,g\; \overline\Psi \gamma_5 \pi_a\tau_a\; \Psi + ... \; .
276: \label{1Leff}
277: \end{equation}
278: In the SU(2) case,
279: ${\cal Q}\equiv\mbox{\rm diag}(Q_u,Q_d)= \mbox{\rm diag}
280: (\frac{2}{3},-\frac{1}{3})$ is the quark charge
281: matrix, and $\tau_a$ are the Pauli SU(2)-isospin matrices acting
282: on the quark iso-doublets $\Psi = (u, d)^T$.
283: This can be extended to the SU(3)-flavor case, where
284: ${\cal Q}\equiv\mbox{\rm diag}(Q_u,Q_d,Q_s) =
285: \mbox{\rm diag} (\frac{2}{3},-\frac{1}{3},-\frac{1}{3})$,
286: if $\tau_a$'s are replaced by the Gell-Mann matrices $\lambda_a$
287: acting on the quark flavor triplets $\Psi = (u, d, s)^T$.
288: The ellipsis in ${\mathcal{L}}_{\mathit{eff}}$
289: serve to remind us that Eq. (\ref{1Leff}) also represents the
290: lowest order terms from the $\sigma$-model Lagrangian which are
291: pertinent for calculating photon-pion processes. The same holds
292: for all chiral quark models ($\chi$QM) -- considered in,
293: {\it e.g.}, Ref. \cite{Andrianov+al98} -- which has the mass term
294: containing the quark-meson coupling
295: \begin{equation}
296: - M{\overline \Psi}(UP_L+U^\dagger P_R)\Psi
297: \label{massTerm}
298: \end{equation}
299: with the projectors
300: \begin{equation}
301: P_{L,R} \equiv \frac{ 1\pm \gamma_5 }{2} \, .
302: \end{equation}
303: Namely, expanding
304: \begin{equation}
305: U^{(\dagger)} \equiv \exp[(-)i \pi_a\tau_a/f_\pi]
306: \end{equation}
307: to the lowest order in
308: $\pi_a$ and invoking the GT relation, again returns the QL model
309: Lagrangian (\ref{1Leff}).
310:
311: This simple QL model (and hence also the lowest order $\chi$QM and
312: the $\sigma$-model) provides an analytic expression ({\it e.g.}, see
313: Ref. \cite{Ametller+al83}) for the amplitude
314: $F(\pi^0\to 2\gamma)$
315: also for $m_\pi > 0$ (but restricted to $m_\pi < 2 M$, which anyway
316: must hold for the light, pseudo-Goldstone pion), namely
317: \begin{equation}
318: F(\pi^0\to 2\gamma)=
319: \frac{e^2 N_c}{12\pi^2f_\pi}
320: \left[ \frac{\arcsin(m_\pi/2M)}{(m_\pi/2M)}\right]^2 =
321: \frac{e^2 N_c}{12\pi^2f_\pi} \left[ 1 + \frac{m_\pi^2}{12M^2} +
322: \dots \right]~.
323: \label{freeLoopAmp}
324: \end{equation}
325:
326:
327: In the QL model, one can similarly go beyond the chiral and
328: soft-point limit in the case of the anomalous process of the type
329: $\gamma \to \pi^+ \pi^0 \pi^-$. Ref. \cite{Bistrovic:1999yy}
330: extended the amplitude (\ref{g3piAnomAmp}) obtained by calculating
331: the ``box" graph, Fig. 2, to the case of nonvanishing pion mass
332: and/or nonvanishing pion momenta.
333:
334:
335:
336: \subsection{Mesons as bound states of quarks dressed by
337: D$\chi$SB}
338:
339: In the aforementioned DS approach, one does not postulate
340: constituent quarks, i.e., effective free quasiparticles
341: with propagators (\ref{freeSq}). Instead, in the DS approach
342: one constructs constituent quarks by solving the DS equation
343: (the ``gap equation'') for the quark propagator. Namely,
344: in this way, starting from the current quarks which in the
345: QCD Lagrangian break chiral symmetry explicitly just by
346: relatively small current mass $m$, one obtains the dynamically
347: dressed quark propagator
348: \begin{equation}
349: S(k)= \frac{1}{\rlap{$k$}/ \,A(k^2) - m - B(k^2)}
350: \equiv \frac{Z(k^2)}{\rlap{$k$}/ - {\cal M}(k^2) }~.
351: \label{EuclS}
352: \end{equation}
353:
354: Even in the chiral limit, where $m = 0$ so that chiral
355: symmetry is not broken explicitly but only dynamically,
356: D$\chi$SB gives the dressing functions $A(k^2) = 1/Z(k^2)$
357: and $B(k^2)\ne 0$ leading to the dynamically generated,
358: momentum-dependent quark mass
359: \begin{equation}
360: {\cal M}(k^2) \equiv \frac{m + B(k^2)}{A(k^2)}
361: \end{equation}
362: which, at small $k^2$, takes values close to a
363: phenomenologically required constituent mass
364: \begin{equation}
365: M \sim \frac{1}{3} \,\, {\rm nucleon} \,\, {\rm mass}
366: \sim \frac{1}{2} \,\, \mbox{\rm rho--meson mass} \, .
367: \end{equation}
368: In this way, the DS approach provides one with a modern
369: constituent quark model possessing many remarkable features.
370: Its presently interesting feature is its relation with the Abelian
371: axial anomaly. Other bound state approaches generally have problems
372: with describing anomalous processes such as the famous
373: $\pi^0 \to \gamma\gamma$ and related anomalous decays.
374: (See Ref. \cite{KeBiKl98} for a comparative discussion thereof.)
375: Thus, it was a significant advance in the theory of bound states,
376: when Roberts \cite{Roberts} and Bando {\it et al.} \cite{bando94}
377: showed that the DS approach, in the chiral and soft limit, reproduces
378: exactly the famous $\pi^0\to\gamma\gamma$ ``triangle"-amplitude
379: (\ref{pi2gammaAmp}).
380: Later, in the same approach and limits, the reproduction of
381: the related ``box"-amplitude (\ref{g3piAnomAmp}) for the
382: $\gamma \to \pi^+ \pi^0 \pi^-$ process was also achieved
383: and clarified \cite{AR96,Bistrovic:1999dy}.
384: Just as the triangle amplitude (\ref{pi2gammaAmp}),
385: the box amplitude (\ref{g3piAnomAmp})
386: is in the DS approach evaluated analytically and without any fine tuning
387: of the bound-state description of the pions~\cite{AR96}. This happens
388: because the DS approach incorporates D$\chi$SB into the bound states
389: consistently, so that the pion,
390: although constructed as a quark--antiquark composite described
391: by its BS bound-state vertex
392: $\Gamma_{\pi}(p,k_{\pi})$, also appears
393: as a Goldstone boson in the chiral limit
394: ($k_{\pi}$ denotes the
395: relative momentum of the quark and antiquark constituents of the pion
396: bound state).
397:
398: Technically, DS calculations of transition amplitudes are much
399: more complicated than the corresponding free QL calculations;
400: not only more complicated, dressed quark propagators (\ref{EuclS})
401: are used instead of (\ref{freeSq}), but the related
402: momentum-dependent $q\bar q$ pseudoscalar pion bound state
403: BS vertex solutions $\Gamma_{\pi^a}$ replace
404: $g\gamma_5\tau_a$ quark-pion Yukawa point couplings used in
405: QL calculations. Still, these ingredients of
406: the DS approach conspire together so that
407: any dependence on what precisely the solutions for
408: the dressed quark propagator (\ref{EuclS})
409: and the BS vertex $\Gamma_{\pi}(p,k_{\pi})$
410: are, drops out in the course of the analytical derivation of
411: Eqs.~(\ref{pi2gammaAmp}) and (\ref{g3piAnomAmp}) in the chiral
412: and soft limit. This is as
413: it should be, because the amplitudes predicted by the anomaly (again
414: in the chiral limit $m=0=m_\pi$ and the
415: soft limit, {\it i.e.}, at zero four-momentum) are independent of the
416: bound-state structure, so that the DS approach is the bound-state
417: approach that correctly incorporates the Abelian axial anomaly.
418:
419: Another crucial requirement for reproducing the Abelian
420: axial anomaly amplitudes in Eqs. (\ref{pi2gammaAmp}) and
421: (\ref{g3piAnomAmp}), is that the electromagnetic interactions
422: are embedded in the context of the DS approach in a way
423: satisfying the vector Ward--Takahashi identity (WTI)
424: \begin{equation}
425: (k^\prime-k)_\mu \Gamma^\mu(k^\prime,k)=S^{-1}(k^\prime)-S^{-1}(k)
426: \label{vWTI}
427: \end{equation}
428: for the dressed quark-photon-quark ($qq\gamma$) vertex
429: $\Gamma_\mu (k,k^\prime)$. The so-called generalized impulse
430: approximation (GIA) (used, for example, by
431: Refs. \cite{bando94,Roberts,AR96,KeBiKl98,KeKl1,KlKe2,KeKl3,Kekez:2003ri})
432: is such a framework.
433: There, the quark-photon-quark ($qq\gamma$) vertex
434: $\Gamma_\mu (k,k^\prime)$ is dressed so that it satisfies
435: the vector WTI (\ref{vWTI})
436: together with the quark propagators (\ref{EuclS}),
437: which are in turn dressed consistently with the solutions for
438: the pion bound state BS vertices $\Gamma_{\pi}$.
439: The triangle graph for $\pi^0 \to \gamma\gamma$ in Fig. 1
440: and the box graph for $\gamma\to 3\pi$ in Fig. 2 is a GIA graph
441: if all its propagators and vertices are dressed like this.
442: (On the example of $\pi^0\to\gamma\gamma$, Table 1 of Ref.~\cite{KeKl1}
443: illustrates quantitatively the consequences of using,
444: instead of a WTI-preserving dressed $qq\gamma$ vertex,
445: the bare vertex $\gamma^\mu$, which violates the vector
446: WTI (\ref{vWTI}) in the context of the DS approach.)
447:
448:
449: In practice, one usually uses
450: \cite{Roberts,AR96,KeBiKl98,KeKl1,KlKe2,KeKl3,Kekez:2003ri}
451: realistic WTI-preserving {\it Ans\"{a}tze} for $\Gamma^\mu(k^\prime,k)$.
452: Following Ref.~\cite{AR96}, we employ the widely used
453: Ball--Chiu ~\cite{BC} vertex, which is fully given
454: in terms of the quark propagator functions of Eq. (\ref{EuclS}):
455: \begin{eqnarray}
456: \Gamma^\mu(k^\prime,k) =
457: [A(k^{\prime 2}) \! + \! A(k^2)]
458: \frac{\gamma^\mu}{\textstyle 2}
459: + \frac{\textstyle (k^\prime+k)^\mu }
460: {\textstyle (k^{\prime 2} - k^2) }
461: \{[A(k^{\prime 2}) \! - \! A(k^2)] \,
462: \frac{\textstyle ({\rlap{$k$}/}^\prime + \rlap{$k$}/) }{\textstyle 2}
463: - [B(k^{\prime 2}) \! - \! B(k^2)] \, \}~.
464: \label{BC-vertex}
465: \end{eqnarray}
466:
467: The amplitude
468: $F(\pi^0\to 2\gamma)$
469: obtained in the chiral and soft limit
470: is an excellent approximation for the realistic $\pi^0\to\gamma\gamma$
471: decay. On the other hand, the already published \cite{Antipov+al87}
472: and presently planned Primakoff experiments at CERN \cite{Moinester+al99},
473: as well as the current CEBAF measurement of the
474: $\gamma \pi^+ \to \pi^+ \pi^0$ process \cite{Miskimen+al94}
475: involve values of energy and momentum transfer sufficiently
476: large to give a lot of motivation for theoretical predictions
477: of the extension of the anomalous $\gamma\to 3 \pi$ amplitude
478: away from the soft point. Ref. \cite{Bistrovic:1999dy} thus extended
479: the DS calculation of the result (\ref{g3piAnomAmp}) away from
480: the soft and chiral limit, giving the corresponding form factor
481: in the form of the expansion in the powers of the pion momenta
482: and mass. (See also Refs. \cite{Klabucar:2000yd,Klabucar:2000mk}.)
483:
484:
485: \subsection{Explanation of the equivalence of the P-coupling method
486: and anomaly calculations}
487:
488: Some confusion has resulted from the fact that anomalous
489: amplitudes (such as those of $\pi^0 \to \gamma\gamma$ and
490: $\gamma \pi^+ \to \pi^+ \pi^0$ processes) can be obtained
491: either through the anomaly analysis or through the pseudoscalar
492: coupling to quark loops as in subsections A and B above. In a
493: way, this is a continuation of an earlier confusion when the
494: Veltman-Sutherland theorem (VSTh) \cite{Sutherland:1967vf,VeltmanPRSLA67}
495: was perceived to require the vanishing $\pi^0 \to \gamma\gamma$
496: amplitude
497: $F_{m_\pi=0}(\pi^0\to 2\gamma)$,
498: in conflict with experiment. Subsequently, VSTh seemed to
499: some to be invalidated by the anomaly which also explains the
500: experimentally found $\pi^0 \to \gamma\gamma$ width. But
501: the Steinberger-like calculation, i.e., the P-coupling method,
502: also explains the experimental $\pi^0 \to \gamma\gamma$ width,
503: and VSTh is of course a valid mathematical result.
504:
505: To be precise, VSTh is the exact statement that the quantity
506: (\ref{T_mu_nu}), constructed from the vector electromagnetic
507: current $J^\mu(x)$ and the third isospin component of the
508: isovector axial current
509: $A^\rho_3(x) = \overline\Psi(x) \gamma^\rho \gamma_5 \tau_3 \Psi(x)$
510: as follows:
511: \begin{equation}
512: \frac{1}{2} \int d^4x \, d^4y \, {\rm e}^{{\rm i}(x\cdot k_1
513: + y\cdot k_2)} \langle 0 | {\rm T} [ J^\mu(x) J^\nu(y)
514: \partial_\rho A^\rho_3 (0) ] | 0 \rangle =
515: \epsilon^{\mu\nu\alpha\beta} k_{1\alpha} k_{2\beta}
516: \Phi(k_1 \cdot k_2, k_1^2, k_2^2)
517: + {\cal O}[(k)^3] \, ,
518: \label{T_mu_nu}
519: \end{equation}
520: vanishes in the chiral limit as $\Phi \propto k_1 \cdot k_2
521: \propto m_\pi^2 \propto m$ \cite{seeRef}.
522: (Throughout, $k_1$ and $k_2$ are the momenta of the two photons.)
523: Then, when the PCAC relation for the third isospin component,
524: $\partial_\mu A^\mu_3(x) = 2 f_\pi m_\pi^2 \pi^0(x)$,
525: is modified by Abelian anomaly to read
526: \begin{equation}
527: \partial_\mu A^\mu_3 = 2 f_\pi m_\pi^2 \pi^0 +
528: \frac{e^2 N_c}{16\pi^2} \tr(\tau_3 {\cal Q}^2)
529: \epsilon_{\mu\nu\alpha\beta} F^{\mu\nu} F^{\alpha\beta} \\
530: \\
531: = {\rm i} 2 m \overline\Psi \gamma_5 \tau_3 \Psi +
532: \frac{e^2 N_c}{16\pi^2} \tr(\tau_3 {\cal Q}^2)
533: \epsilon_{\mu\nu\alpha\beta} F^{\mu\nu} F^{\alpha\beta}\, ,
534: \label{divAanom}
535: \end{equation}
536: it becomes clear that VSTh, i.e., the vanishing of Eq. (\ref{T_mu_nu}),
537: does {\it not} imply
538: $F_{m_\pi=0}(\pi^0\to 2\gamma)=0$,
539: but that VSTh relates the Steinberger-like calculation of the
540: PVV amplitude to the anomaly. That is, VSTh dictates that
541: in the chiral limit, the PVV $\pi^0 \to \gamma\gamma$ amplitude
542: is given exactly by the coefficient of the anomaly term.
543: This is precisely the result (\ref{pi2gammaAmp}),
544: empirically successful and of the order ${\cal O}[(k)^0]$.
545:
546: Note that together with the result (\ref{g3piAnomAmp}),
547: the above discussion also clarifies the relationship of
548: the anomaly and the PVVV ``box'' calculation of the
549: $\gamma \to \pi^+ \pi^0 \pi^-$ amplitude.
550:
551: Even with the above understanding, one may wonder when
552: and why the WZ term should or should not be included in
553: one's Lagrangian. The WZ term naturally appears when the
554: quarks are integrated out so that one obtains a low-energy
555: theory containing only the meson fields.
556: The situation is more subtle when the quarks are left in the theory.
557: Georgi explains pedagogically \cite{Georgi6a} the relationship
558: and equivalence between the following two distinct cases.
559: {\it (i)} If the quarks transform nonlinearly under the chiral
560: transformations, in which case all their interactions explicitly
561: involve derivatives so that one has axial couplings of the
562: quarks to mesons, but no such pseudoscalar couplings, the
563: WZ term must be included.
564: {\it (ii)} Equivalently, the quarks can transform linearly,
565: and in this case the WZ term is not present. The quarks
566: chirally transforming linearly are related to the quarks
567: transforming nonlinearly by a chiral transformation. In this
568: case the quark mass term assumes the form (\ref{massTerm}),
569: which contains nonderivative, pseudoscalar ($\gamma_5$)
570: couplings of the quarks to the Goldstone bosons. This is
571: seen by comparing Eq. (\ref{massTerm}) with the expansion
572: (\ref{1Leff}) if one takes into account that the couplings
573: are determined by the quark-level GT relation (\ref{GTrel}).
574:
575: On the basis of the above experience with the
576: $\pi^0\to \gamma\gamma$ and $\gamma \to \pi^+ \pi^0 \pi^-$
577: amplitudes, we can expect the complete equivalence of the
578: cases {\it (i)} and {\it (ii)}, that is, of the anomaly
579: and P-coupling calculations. For that, the P-coupling
580: (``Steinberger-like'') calculations with the coupling
581: (\ref{massTerm}), should reproduce the effects of the WZ term.
582: Indeed, Georgi shows that one can obtain any coupling in
583: the WZ term from a Steinberger-like quark loop calculation
584: \cite{Georgi6a}.
585: Here, it suffices to illustrate this on the example of
586: the $\pi^0\to \gamma\gamma$ ``triangle''PVV calculation,
587: where squeezing the quark loop to a point would amount
588: to having the effective $\pi^0 \gamma\gamma$ interaction
589: (\ref{L_pi2gamma}) but with the coupling predicted to be
590: (in the chiral limit)
591: \begin{equation}
592: g_{\pi\gamma\gamma} = \frac{1}{8}
593: F_{m_\pi=0}(\pi^0\to 2\gamma)
594: =
595: \frac{e^2 N_c}{96\pi^2f_\pi} \, ,
596: \label{gpi2gamma}
597: \end{equation}
598: which makes Eq. (\ref{L_pi2gamma}) exactly equal to that piece of the
599: WZ term \cite{WZ} which is relevant for the $\pi^0\to \gamma\gamma$
600: decay.
601:
602:
603: \section{Processes going through the quark triangle}
604: \label{manyProc}
605:
606: In this section we calculate the amplitudes for a number of processes
607: using the quark triangle graphs.
608: Figures 1 and 3 show three such PVV processes.
609: First we consider $\pi^0\to \gamma\gamma$ decay via the $u$ and $d$
610: quark triangle graph for $\pi^0=(\bar{u}u-\bar{d}d)/\sqrt{2}$, $N_c=3$
611: and GT relation (\ref{GTrel}) leading to the pion decay constant:
612: $f_\pi=\hat{m}/g_{\pi qq}$.
613: This amplitude is finite and for the experimental value of the pion
614: decay constant, $f_\pi = (92.42\pm 0.26) \, \MeV$ \cite{PDG2004},
615: gives \cite{Delbourgo} the chiral-limit amplitude (\ref{pi2gammaAmp})
616: of magnitude
617:
618: \begin{figure}
619: \begin{tabular}{cc}
620: \includegraphics[height=44mm,angle=0]{OmegaToPi0Gamma.eps}
621: &
622: \includegraphics[height=44mm,angle=0]{RhoToPi0Gamma.eps}
623: \end{tabular}
624: \caption{Two examples of the PVV triangle graphs
625: where just one of the vector vertices couples to a photon,
626: whereas the other couples to a vector meson. These two graphs
627: describe the decays of $\omega$ and $\rho$ mesons into a photon
628: and a pion.}
629: \label{fig:PVVdiagrams}
630: \end{figure}
631:
632: \begin{equation}
633: |F_{m_\pi=0}(\pi^0\to 2\gamma)|
634: =
635: \frac{e^2}{4\pi^2 f_\pi}=0.0251~\GeV^{-1}
636: \end{equation}
637:
638: \noindent very close to experimental data \cite{PDG2004}
639:
640: \begin{equation}
641: |F_{\rm exp}(\pi^0\to 2\gamma)| =
642: \left[\frac{64\pi\Gamma(\pi^0\to\gamma\gamma)}{m_\pi^3}\right]^{1/2}
643: =
644: (0.0252\pm 0.0009)~\GeV^{-1}~.
645: \end{equation}
646:
647: \noindent Likewise, the $u$, $d$ quark triangles
648: for $\rho\to\pi\gamma$ decay give \cite{Delbourgo}
649:
650: \begin{equation}
651: |F(\rho\to\pi\gamma)|=\frac{e g_\rho}{8\pi^2 f_\pi}
652: = 0.206~\GeV^{-1}
653: \end{equation}
654:
655: \noindent for $g_\rho=4.965\pm 0.002$
656: found from $\rho^0\to e^-e^+$ decay
657: \cite{PDG2004}:
658:
659: \begin{equation}
660: \Gamma(\rho^0\to e^-e^+)
661: =
662: \frac{e^4 m_\rho}{12\pi g_\rho^2}
663: =
664: (7.02\pm 0.11)~\keV~.
665: \end{equation}
666:
667: \noindent The calculated $|F(\rho\to\pi\gamma)|$
668: is also near data
669: \cite{PDG2004},
670: \begin{equation}
671: |F_{\rm exp}(\rho\to\pi\gamma)|
672: =
673: \left[\frac{12\pi\Gamma(\rho\to\pi\gamma)}{q^3}\right]^{1/2}
674: =(0.225\pm 0.011)~\GeV^{-1}~,
675: \end{equation}
676: where $q=(m_\rho^2-m_\pi^2)/(2m_\rho)$ is the photon momentum.
677: [Actually, the above value is a weighted average of
678: $F_{\rm exp}(\rho^0\to\pi^0\gamma)$
679: and $F_{\rm exp}(\rho^\pm\to\pi^\pm\gamma)$ amplitudes.]
680:
681: Next we predict
682: the $u$, $d$ quark triangle amplitude for $\omega\to\pi\gamma$
683: taking $\omega$
684: as 99\% nonstrange \cite{PDG2004} ($\cos^2\phi_V \approx 0.99$)
685:
686: \begin{equation}
687: |F(\omega\to\pi\gamma)|
688: =
689: \frac{\cos\phi_V\, e\, g_\omega}{8\pi^2 f_\pi} = 0.705~\GeV^{-1}
690: \label{FOmegaToPi0Gamma}
691: \end{equation}
692:
693: \noindent for $g_\omega=17.06\pm 0.28$
694: found from $\omega\to e^-e^+$ decay.
695: The mixing angle is\footnote{We use quadratic mass formulae
696: for mesons (See, {\em e.g.}, Ref.~\cite{PDG2002} and earlier).
697: However, the input experimental meson masses are newest,
698: taken from Ref.~\cite{PDG2004}.}
699:
700: \begin{equation}
701: \phi_V
702: =
703: \theta_V - \arctan(\frac{1}{\sqrt{2}}) =
704: \arctan\sqrt{\frac{\frac{1}{3}(4 m_{K^\star}^2 - m_\rho^2) - m_\varphi^2 }
705: {m_\omega^2-\frac{1}{3}(4 m_{K^\star}^2 - m_\rho^2) } }
706: -\arctan(\frac{1}{\sqrt{2}})
707: = (5.208\pm 0.092)^\circ~.
708: \label{phiV}
709: \end{equation}
710:
711: \noindent Again this theory in Eq.~(\ref{FOmegaToPi0Gamma}) is near data
712: $(0.722\pm 0.012)~\GeV^{-1}$
713: \cite{PDG2004}.
714:
715: Other PVV photon decays involve the $\eta$ and $\eta^\prime$
716: mixed non--strange and $\bar{s}s$ pseudoscalar mesons. Again the
717: quark triangle amplitudes are a close match with data
718: \cite{Delbourgo,delbourgo95,bramon98}.
719:
720: The quark--triangle (QT) calculation gives reliable predictions
721: also for the $\eta$ and $\eta^\prime$ two--photon decays:
722:
723: \begin{eqnarray}
724: |F(\eta\to\gamma\gamma)|
725: =
726: \frac{e^2}{4\pi^2 f_\pi}
727: \frac{N_c}{9}
728: (5 \cos\phi_P - \sqrt{2} \frac{\hat{m}}{m_s}\sin\phi_P)
729: =
730: 0.0255\GeV^{-1}~,
731: \\
732: |F(\eta^\prime\to\gamma\gamma)|
733: =
734: \frac{e^2}{4\pi^2 f_\pi}
735: \frac{N_c}{9}
736: (5 \sin\phi_P + \sqrt{2} \frac{\hat{m}}{m_s}\cos\phi_P)
737: =
738: 0.0345~\GeV^{-1}~.
739: \end{eqnarray}
740:
741: \noindent This should be compared with the experimental data:
742:
743: \begin{eqnarray}
744: |F_{\rm exp}(\eta\to\gamma\gamma)|
745: =
746: \left[\frac{64\pi\Gamma(\eta\to\gamma\gamma)}{m_\eta^3}\right]^{1/2}
747: =(0.02498\pm 0.00064)~\GeV^{-1}~,
748: \\
749: |F_{\rm exp}(\eta^\prime\to\gamma\gamma)|
750: =
751: \left[\frac{64\pi\Gamma(\eta^\prime\to\gamma\gamma)}{m_{\eta^\prime}^3}\right]^{1/2}
752: =(0.03133\pm 0.00055)~\GeV^{-1}~,
753: \end{eqnarray}
754:
755: \noindent where $\Gamma(\eta\to\gamma\gamma)
756: =(0.5108\pm 0.0268)~\keV$
757: and
758: $\Gamma(\eta^\prime\to\gamma\gamma)
759: =(4.29\pm 0.15)~\keV$. The ratio of the constituent quark masses
760: is $m_s/m=2f_K/f_\pi-1=1.445\pm 0.024$
761: for $f_{\pi^\pm}=(92.4\pm 0.3)~\MeV$
762: and $f_K=(113.0\pm 1.0)~\MeV$ \cite{PDG2004}.
763: The mixing angle is \cite{Jones:1979ez,Kekez:2000aw}
764:
765: \begin{equation}
766: \phi_P =
767: \theta_P + \arctan({\sqrt{2}}) =
768: \arctan
769: \sqrt{
770: \frac {(m_{\eta'}^2 - 2m_K^2 + m_\pi^2) (m_\eta^2 -
771: m_\pi^2)}
772: {(2m_K^2 - m_\pi^2 - m_\eta^2) (m_{\eta'}^2
773: - m_\pi^2)}
774: }
775: = (42.441\pm 0.019)^\circ~.
776: \label{phiP}
777: \end{equation}
778:
779: Next, we can calculate the $\rho^0\to\eta\gamma$ amplitude employing
780: the quark--triangle diagram,
781:
782: \begin{equation}
783: |F(\rho^0\to\eta\gamma)|
784: =
785: \frac{e g_\rho}{8\pi^2 f_\pi} 3\cos\phi_P
786: =0.456~\GeV^{-1}~.
787: \end{equation}
788:
789: \noindent Again, this is close to the experimental data,
790:
791: \begin{equation}
792: |F_{\rm exp}(\rho^0\to\eta\gamma)|
793: =
794: \left[\frac{12\pi\Gamma(\rho^0\to\eta\gamma)}{q^3}\right]^{1/2}
795: =(0.48\pm 0.03)~\GeV^{-1}~,
796: \end{equation}
797:
798: \noindent where
799: $q=(m_\rho^2-m_\eta^2)/(2m_\rho)=(194.5\pm 0.4)~\MeV$
800: is the photon momentum and
801: $\Gamma(\rho^0\to\eta\gamma)
802: =(45.1\pm 6.0)~\keV$.
803: A similar situation is with the $\eta^\prime\to\rho\gamma$ amplitude,
804: for which the quark--triangle calculation gives
805:
806: \begin{equation}
807: |F(\eta^\prime\to\rho^0\gamma)|
808: =
809: \frac{e g_\rho}{8\pi^2 f_\pi} 3\sin\phi_P
810: =
811: 0.417~\GeV^{-1}~.
812: \end{equation}
813:
814: \noindent The corresponding experimental value is
815:
816: \begin{equation}
817: |F_{\rm exp}(\eta^\prime\to\rho^0\gamma)|
818: =
819: \left[\frac{4\pi\Gamma(\eta^\prime\to\rho^0\gamma)}{q^3}\right]^{1/2}
820: =(0.411\pm 0.017)~\GeV^{-1}~,
821: \end{equation}
822:
823: \noindent where
824: $q=(m_{\eta^\prime}^2-m_\rho^2)/(2m_{\eta^\prime})
825: =(164.7\pm 0.4)~\MeV$ is the photon momentum and
826:
827: \begin{equation}
828: \Gamma(\eta^\prime\to\rho^0\gamma\,\,\,\mbox{\rm including non--resonant}\,\,\, \pi^+\pi^-\gamma)
829: =(60.0\pm 5.0)~\keV
830: \end{equation}
831:
832: \noindent is the experimental decay width \cite{PDG2004}.
833:
834:
835: The $\eta\to\pi\pi\gamma$ amplitude is
836:
837: \begin{equation}
838: |M^{\mbox{\rm\scriptsize VMD}}_{\eta\to\pi\pi\gamma}|
839: =
840: |\frac{2g_{\rho\pi\pi} M^{\mbox{\rm\scriptsize QT}}_{\rho^0\to\eta\gamma}}
841: {m_\rho^2-s}|
842: =9.80~\GeV^{-3}
843: \end{equation}
844:
845: \noindent where $s=m_\pi^2$. The $\eta\to\pi\pi\gamma$ decay width is
846:
847: \begin{equation}
848: \Gamma(\eta\to\pi\pi\gamma)
849: =
850: \frac{|M_{\eta\to\pi\pi\gamma}|^2}{(2\pi)^3}
851: m_\eta^{7} Y_\eta
852: =56.2~\eV~,
853: \end{equation}
854:
855: \noindent where $Y_\eta = 0.98\cdot 10^{-5}$ \cite{thew}.
856: This is in a good agreement with the experimental value
857:
858: \begin{equation}
859: \Gamma(\eta\to\pi\pi\gamma)
860: =(60.4\pm 3.6)~\eV~,
861: \end{equation}
862: \noindent revealing that the vector meson dominance
863: is the main effect, while the coupling through VPPP quark box loop
864: (``contact term") contributes little.
865:
866: It is known that $\omega\to 3\pi$ decay is dominated by
867: $\rho$--meson poles. The required $\omega\to\rho\pi$ amplitude can be estimated
868: as
869:
870: \begin{equation}
871: |M^{\mbox{\rm\scriptsize VMD}}(\omega\to\rho\pi)|
872: =
873: \left(\frac{g_\rho}{e}\right)
874: |F(\omega\to\pi^0\gamma)|
875: \sim 12~\GeV^{-1}~,
876: \end{equation}
877:
878: \noindent but cannot be measured because there is no phase space for this
879: process. The $\omega\to\rho\pi$ amplitude is more precisely defined with
880: QL, additionally enhanced with a meson loop associated with sigma exchange
881: \cite{delbourgo95,bramon98,freund-nandi-rudaz},
882:
883: \begin{equation}
884: |M(\omega\to\rho\pi)|_{\mbox{\rm\scriptsize QT}}
885: =
886: \frac{3 g_{\rho\pi\pi}^2}{8\pi^2 f_\pi}
887: \approx
888: 15~\GeV^{-1}~.
889: \end{equation}
890:
891: \noindent {The scalar amplitude
892: $M^{\mbox{\rm\scriptsize VMD}}(\omega\to 3\pi)$
893: is dominated by the $\rho$ meson in each of the three possible channels
894: \cite{gell-mann},
895:
896: \begin{equation}
897: |M^{\mbox{\rm\scriptsize VMD}}(\omega\to 3\pi)|
898: =
899: 2 g_{\rho\pi\pi}
900: |M(\omega\to\rho\pi)|
901: \left[
902: \frac{1}{m_\rho^2-s}
903: +
904: \frac{1}{m_\rho^2-t}
905: +
906: \frac{1}{m_\rho^2-u}
907: \right]
908: \approx 1480~\GeV^{-3}~.
909: \end{equation}
910:
911: \noindent Following Thew's phase space analysis \cite{thew}, we get
912:
913: \begin{equation}
914: \Gamma(\omega\to\ 3\pi)
915: =
916: \frac{|M^{\mbox{\rm\scriptsize VMD}}(\omega\to 3\pi)|^2}{(2\pi)^3}
917: m_\omega^7
918: Y_\omega
919: =
920: 7.3~\MeV
921: \end{equation}
922:
923: \noindent where $Y_\omega = 4.57\cdot 10^{-6}$ is used.
924: The predicted value is close to the experimental value \cite{PDG2004}
925:
926: \begin{equation}
927: \Gamma(\omega\to 3\pi)
928: =
929: (7.6\pm 0.1)~\MeV~.
930: \end{equation}
931:
932: Here we have taken $\omega$ as pure NS, although it is about 99\% NS,
933: since $\phi_V=(5.208 \pm 0.092)^\circ$ from our Eq.~(\ref{phiV}).
934:
935:
936: In the quark--level $\sigma$--model a quark box diagram contributes
937: to the $\omega\to 3\pi$ decay. This box diagram can be interpreted as a
938: contact term. It is shown that the contact contribution is small by itself,
939: but can be enlarged through the interference effect \cite{lucio00}.
940:
941: Using $\phi_P=(42.441\pm 0.019)^\circ$ from our Eq.~(\ref{phiP}),
942: we predict the tensor $T\to PP$ branching ratios for
943: $a_2(1320)$:
944:
945: \begin{eqnarray}
946: \begin{array}{ll}
947: \mbox{\rm BR}(\frac{\textstyle{a_2\to\eta\pi}}{\textstyle{a_2\to K\bar{K}}})
948: =\left(\frac{\textstyle{p_{\eta\pi}}}{\textstyle{p_{K}}}\right)^5
949: 2 \cos^2\phi_P=2.996
950: & (\mbox{\rm data}\,\,\,\,
951: 2.96\pm 0.54)~,
952: \\
953: \mbox{\rm BR}(\frac{\textstyle{a_2\to\eta^\prime\pi}}
954: {\textstyle{a_2\to K\bar{K}}})
955: =\left(\frac{\textstyle{p_{\eta^\prime\pi}}}{\textstyle{p_{K}}}\right)^5
956: 2 \sin^2\phi_P=0.1113
957: & (\mbox{\rm data}\,\,\,\,
958: 0.108\pm 0.025)~,
959: \\
960: \mbox{\rm BR}(\frac{\textstyle{a_2\to\eta^\prime\pi}}
961: {\textstyle{a_2\to\eta\pi}})
962: =\left(\frac{\textstyle{p_{\eta^\prime\pi}}}{\textstyle{p_{\eta\pi}}}\right)^5 \tan^2 \phi_P=0.0371
963: & (\mbox{\rm data}\,\,\,\,
964: 0.0366\pm 0.0069)~,
965: \end{array}
966: \end{eqnarray}
967:
968: \noindent for center of mass momenta
969: $p_{\eta\pi}=535~\MeV$,
970: $p_{\eta^\prime\pi}=287~\MeV$,
971: $p_{K}=437~\MeV$.
972: The above data branching ratios follow from $a_2(1320)$ recent fractions
973: \cite{PDG2004}:
974: $\mbox{\rm BR}(a_2\to\eta\pi) =(14.5\pm 1.2)\%$,
975: $\mbox{\rm BR}(a_2\to K\bar{K}) =( 4.9\pm 0.8)\%$ and
976: $\mbox{\rm BR}(a_2\to\eta^\prime\pi)=( 5.3\pm 0.9)\cdot 10^{-3}$.
977:
978:
979: \section{Comments related to the gluon anomaly}
980:
981: The approach using the pseudoscalar coupling is, in our opinion,
982: also relevant for the effects related to the non-Abelian, ``gluon"
983: ABJ axial anomaly. Here, we comment on this only briefly,
984: and direct the reader to the original references for details.
985:
986:
987: \subsection{Goldstone structure and $\eta$-$\eta'$ phenomenology}
988:
989: The first point concerns the $\eta$-$\eta'$ complex and the $U_A(1)$
990: problem related to it.
991:
992: In the chiral limit $m_\pi = m_K = m_{\eta_8} = 0$,
993: since all members of the flavor-SU(3) pseudoscalar meson octet
994: are massless in this theoretical, but very useful limit. The only
995: non-vanishing ground-state pseudoscalar meson mass in this limit
996: is the mass of the SU(3)-singlet pseudoscalar meson $\eta_1$.
997: This is thanks to the non-Abelian, gluon ABJ axial anomaly, i.e.,
998: to the fact that the divergence of the SU(3)-singlet axial current
999: \begin{equation}
1000: A^\mu_0(x) = \overline\Psi(x) \gamma^\mu \gamma_5 \Psi(x) \, ,
1001: \end{equation}
1002: receives the contributions from gluon fields $G_a^{\mu\nu}$
1003: similar to those of photon fields $F^{\mu\nu}$ in
1004: Eq. (\ref{divAanom}), namely
1005: \begin{equation}
1006: \partial_\mu A^\mu_0 =
1007: 2 {\rm i} m_u \, \overline{u} \gamma_5 u +
1008: 2 {\rm i} m_d \, \overline{d} \gamma_5 d +
1009: 2 {\rm i} m_s \, \overline{s} \gamma_5 s +
1010: \frac{3 \, g^2 }{32 \pi^2}
1011: \epsilon_{\mu\nu\alpha\beta} G_a^{\mu\nu} G_a^{\alpha\beta}\, .
1012: \label{divGanom}
1013: \end{equation}
1014: This removes the $U_A(1)$ symmetry and explains why only eight
1015: pseudoscalar mesons are light, and not nine; i.e., why there
1016: is an octet of (almost-)Goldstone bosons, but not a nonet.
1017: The physically observed $\eta$ and $\eta'$ are then the
1018: mixtures of the anomalously heavy $\eta_1$ and
1019: (almost-)Goldstone $\eta_8$ in such a way that $\eta'$ is
1020: predominantly $\eta_1$ and $\eta$ is predominantly $\eta_8$.
1021: This is how the gluon anomaly can save us from the $U_A(1)$ problem
1022: in principle, and the details of how
1023: we achieve a successful description of the $\eta$-$\eta'$
1024: complex, are given in the references
1025: \cite{Jones:1979ez,Kekez:2000aw,Klabucar:1997zi,Klabucar:2000me,Klabucar:2001gr}.
1026: Here we just sketch some important points.
1027: The mass matrix squared $\hat{M}^2$
1028: in the quark basis $|u\bar{u}\rangle$, $|d\bar{d}\rangle$, $|s\bar{s}\rangle$
1029: is
1030:
1031: \begin{equation}
1032: \hat{M}^2
1033: =
1034: \hat{M}_{\mbox{\rm\scriptsize NA}}^2 + \hat{M}_{\rm A}^2
1035: =
1036: \left[
1037: \begin{array}{ccc}
1038: m_{u\bar{u}}^2 & 0 & 0 \\
1039: 0 & m_{d\bar{d}}^2 & 0 \\
1040: 0 & 0 & m_{s\bar{s}}^2
1041: \end{array}
1042: \right]
1043: +
1044: \beta
1045: \left[ \begin{array}{ccl} 1 & 1 & X \\
1046: 1 & 1 & X \\
1047: X & X & X^2
1048: \end{array} \right]~,
1049: \end{equation}
1050:
1051: \noindent
1052: where $\hat{M}_{\mbox{\rm\scriptsize NA}}^2$ is the non-anomalous part of the matrix,
1053: since $m_{u\bar{u}}^2=m_{d\bar{d}}^2=m_\pi^2$ and
1054: $m_{s\bar{s}}^2=2m_K^2-m_\pi^2$ would be the masses of the respective
1055: ``non-strange" (NS) and ``strange" (S) $q\bar{q}$ mesons
1056: if there were no gluon anomaly.
1057: In the NS sector, in the isospin symmetry limit
1058: (which is very close to reality), the relevant combinations are
1059: $| \pi^0 \rangle = | u\bar{u} - d\bar{d} \rangle / \sqrt{2}$
1060: as the neutral partner of the charged pions $| \pi^\pm \rangle$
1061: in the isospin 1 triplet, and the isospin 0 combination
1062: $ | u\bar{u} + d\bar{d} \rangle / \sqrt{2} $.
1063: In the absence of gluon anomaly, but with
1064: an $s$-quark mass heavier than the isosymmetric $u$ and $d$ ones,
1065: $\eta$ would reduce to $|{\rm NS}\rangle=|u\bar{u} + d\bar{d} \rangle / \sqrt{2}$
1066: with the mass $m_{\mbox{\rm\scriptsize NS}} = m_\pi$,
1067: and $\eta'$ to $|{\rm S}\rangle = | s\bar{s} \rangle $ with the mass
1068: $m_{\mbox{\rm\scriptsize S}} = m_{s\bar{s}}$. Both of these assignments are in conflict
1069: with experiment. The realistic contributions of various flavors to
1070: $\eta$ and $\eta'$ and their masses (i.e., the realistic
1071: $\eta$-$\eta'$ mixing) are obtained only thanks to $\hat{M}_{\rm A}^2$,
1072: the anomalous contribution to the mass matrix. In $\hat{M}_{\rm A}^2$, }
1073: the quantity $\beta$ describes transitions
1074: $|q\bar{q}\rangle\to|q^\prime\bar{q}^\prime\rangle$
1075: ($q,q^\prime=u,d,s$) due to the gluon anomaly and $X$
1076: describes the effects of the SU(3) flavor symmetry breaking
1077: on these transitions.
1078: In Refs.~\cite{Jones:1979ez,Kekez:2000aw,Klabucar:2000me},
1079: as the first step
1080: in solving the $U_A(1)$ problem,
1081: we extract $\eta_8$, $\eta_1$ masses from the $\eta$, $\eta^\prime$
1082: via
1083:
1084: \begin{eqnarray}
1085: m_{\eta_8}^2=(m_\eta\cos\theta_P)^2+(m_{\eta^\prime}\sin\theta_P)^2
1086: =(572.73~\MeV)^2~,
1087: \label{meta8}
1088: \\
1089: m_{\eta_1}^2=(m_\eta\sin\theta_P)^2+(m_{\eta^\prime}\cos\theta_P)^2
1090: =(943.05~\MeV)^2~,
1091: \label{meta1}
1092: \end{eqnarray}
1093:
1094: \noindent where
1095: $\theta_P=\phi_P-\arctan(\sqrt{2})=(-12.295\pm 0.019)^\circ$.
1096: The mesons $\eta_8$ and $\eta_1$ are defined as
1097:
1098: \begin{eqnarray}
1099: |\eta_8\rangle &=& \frac{1}{\sqrt{6}} (|u\bar{u}\rangle + |d\bar{d}\rangle - 2 |s\bar{s}\rangle)~, \\
1100: |\eta_1\rangle &=& \frac{1}{\sqrt{3}} (|u\bar{u}\rangle + |d\bar{d}\rangle + |s\bar{s}\rangle)~.
1101: \end{eqnarray}
1102:
1103: \noindent The $\eta_8$ meson mass (\ref{meta8})
1104: $m_{\eta_8}=572.73~\MeV$
1105: is 4.56\% greater than the observed \cite{PDG2004}
1106: $m_\eta=(547.75\pm 0.12)~\MeV$.
1107: The singlet $\eta_1$ mass (\ref{meta1})
1108: $m_{\eta_1}=943.06~\MeV$ is only
1109: 1.56\% below the observed
1110: $m_\eta^\prime=(957.78\pm 0.14)~\MeV$ and close to the
1111: nonstrange--$\bar{s}s$ mixing $U_A(1)$ mass dictated by phenomenology
1112: \cite{Jones:1979ez,Kekez:2000aw,Klabucar:2000me}
1113: \begin{equation}
1114: m_{U_A(1)}
1115: \equiv (3\beta)^{1/2} =
1116: \left[ \frac{3}{4} \frac{(m_{\eta^\prime}^2 - m_\pi^2)
1117: (m_\eta^2 - m_\pi^2)} {m_K^2-m_\pi^2} \right]^{1/2}
1118: = 915.31~\MeV~,
1119: \label{secondU1Amass}
1120: \end{equation}
1121: (This is also close to $912~\MeV$, which is the mass found in the
1122: analogous DS approach \cite{Kekez:2000aw,Klabucar:2000me}.)
1123:
1124: We call the quantity (\ref{secondU1Amass}) the ``mixing $U_A(1)$ mass"
1125: since the mass matrix (which is especially clear in the
1126: nonstrange-strange quark basis) reveals that $m_{U_A(1)}$
1127: induces the mixing between the nonstrange isoscalar
1128: $(|\bar{u}u\rangle+|\bar{d}d\rangle/\sqrt{2}$ and $\bar{s}s$
1129: quark-antiquark states. Equivalently, $m_{U_A(1)}$ can be viewed
1130: as being generated by the transitions among the $\bar{u}u$,
1131: $\bar{d}d$ and $\bar{s}s$ pseudoscalar states; via quark loops,
1132: these pseudoscalar $\bar{q}q$ bound states can annihilate into
1133: gluons which in turn via another quark loop can again recombine
1134: into another pseudoscalar $\bar{q}'q'$ bound state of the same
1135: or different flavor. The quantity $\beta$ appearing in Eq.
1136: (\ref{secondU1Amass}) is then the annihilation strength of such
1137: transitions, in the limit of an exact SU(3) flavor symmetry.
1138: (The realistic breaking of this symmetry is easily introduced
1139: and improves our description of the $\eta$-$\eta'$ complex
1140: considerably.)
1141: The ``diamond'' graph in Fig. 4 gives just the
1142: simplest example of such an annihilation/recombination transition.
1143: Since these annihilations occur in the nonperturbative regime of
1144: QCD, all graphs with any even number of gluons instead of just
1145: those two in Fig. 4, can be just as significant in annihilating
1146: and forming a $C^+$ pseudoscalar $\bar{q}q$ meson. Indeed, this
1147: nonperturbative $U_A(1)$ mass scale,
1148: Eq.~(\ref{secondU1Amass}), is still 3 times higher than the
1149: gluon ``diamond'' graph evaluated perturbatively \cite{Choudhury}.
1150: Thus, we cannot calculate $\beta = m_{U_A(1)}^2/3$ and the
1151: situation is much more complicated and less clear than in
1152: the Abelian case, where we have seen, in Sec. 2.C, that PVV,
1153: the quark triangle graph with pseudoscalar coupling,
1154: reproduces the effect of the axial anomaly, i.e.,
1155: the WZ Lagrangian term, or equivalently, the effect of the
1156: anomalous term $({e^2 N_c}/{16\pi^2}) \tr(\tau_3 {\cal Q}^2)
1157: \epsilon_{\mu\nu\alpha\beta} F^{\mu\nu} F^{\alpha\beta}$
1158: in the divergence (\ref{divAanom}) of the current $A^\mu_3(x)$.
1159: Can it then be founded to think that the annihilation graphs with
1160: the pseudoscalar meson-quark coupling, such as the ``diamond''
1161: graph in Fig. 4, give rise to the anomalous term
1162: $({3 \, g^2 }/{32 \pi^2})
1163: \epsilon_{\mu\nu\alpha\beta} G_a^{\mu\nu} G_a^{\alpha\beta}$
1164: in the divergence (\ref{divGanom}) of the SU(3)-singlet current
1165: $A^\mu_0(x)$, and thus ultimately to the large mass of $\eta_0$
1166: (and of the observed $\eta'$)? Well, this conjecture may remain
1167: a speculation since we cannot calculate $\beta$ due to the
1168: nonperturbative nature of the problem. Nevertheless, when
1169: we use it in our approach as a parameter with the value given by
1170: Eq.~(\ref{secondU1Amass}), we obtain a very good description of
1171: the $\eta$-$\eta'$ complex phenomenology \cite{Jones:1979ez,Kekez:2000aw,Klabucar:1997zi,Klabucar:2000me,Klabucar:2001gr}.
1172: This includes not only the masses of $\eta$ and $\eta'$, but also
1173: their $\gamma\gamma$ decay widths, and the mixing angle
1174: $\theta_P \approx -13^\circ$ consistently following from
1175: the masses and $\gamma\gamma$ widths.
1176: This gives a strong motivation for the above conjecture.
1177:
1178: \begin{figure}
1179: \includegraphics[height=66mm,angle=0]{DiamondGraph.eps}
1180: \caption{Nonperturbative QCD annihilation of a
1181: quark-antiquark bound state illustrated by the diagram with
1182: two-gluon exchange. The $\bar{q}q$ pseudoscalar $P$ is coupled
1183: to a quark loop, whereby it can annihilate into gluons which in
1184: turn recombine into the pseudoscalar $P'$ having the flavor
1185: content $\bar{q'}q'$.}
1186: \label{fig:diamond}
1187: \end{figure}
1188:
1189: \subsection{Taming of strong CP problem}
1190:
1191:
1192: We should also note that our conjecture in the previous subsection
1193: goes well with the arguments of Banerjee {\it et al.}
1194: \cite{Mitra}, that there is really no strong CP problem. They
1195: find that one does {\it not} need vanishing
1196: $\Theta_{\rm eff} = \Theta - {\rm tr}\ln {\hat M}$
1197: (where ${\hat M}$ is the quark mass matrix). Thus,
1198: one does not need any fine-tuning, and all CP violation in
1199: the QCD Lagrangian can be avoided by having $\Theta = 0$
1200: in its CP-violating term
1201: \begin{equation}
1202: {\cal L}_\Theta =
1203: - \Theta \, \frac{ g^2 }{64 \pi^2}
1204: \epsilon_{\mu\nu\alpha\beta} G_a^{\mu\nu} G_a^{\alpha\beta}\, .
1205: \label{ThetaQCD}
1206: \end{equation}
1207: This term in the QCD Lagrangian breaks the $U_A(1)$ symmetry
1208: and corresponds to the anomalous term $\propto
1209: \epsilon_{\mu\nu\alpha\beta} G_a^{\mu\nu} G_a^{\alpha\beta}$
1210: in the divergence (\ref{divGanom}) of the singlet current.
1211: The term (\ref{ThetaQCD}) is allowed by gauge invariance and
1212: renormalizability, but apparent nonexistence of the strong
1213: CP violation, and also of axions,
1214: is the solid reason to have it vanishing.
1215: Our conjecture, that P-coupled annihilation graphs reproduce
1216: the effect of the gluon ABJ anomaly, naturally agrees with
1217: the vanishing of this term and with putting the case of the
1218: strong CP problem to rest {\it \`a la} Banerjee {\it et al.}
1219: \cite{Mitra}.
1220:
1221: \section{Summary/Discussion}
1222: \label{sec:summary}
1223:
1224: We have presented and surveyed in detail the method of
1225: pseudoscalar coupling of pseudoscalar mesons to the
1226: ``triangle" and ``box" quark loops. We have reviewed
1227: how this method gives the equivalent results to the
1228: anomaly calculations. The P-coupling method has also
1229: been illustrated on the example of many decay amplitudes.
1230:
1231: The AVV anomaly \cite{Adler69,BellJackiw69} involves 10 invariant
1232: amplitudes (reduced to 1 or 2 amplitudes for $\pi^0\to \gamma\gamma$
1233: decay using additional Ward identities). If instead one considers
1234: the PVV transition with a pseudoscalar
1235: coupling, then the PVV quark triangle amplitude is
1236: finite and leads to many decay amplitudes (over 15)
1237: then in agreement with data to within 3\% and not
1238: involving free parameters \cite{Delbourgo}. To solve
1239: instead the former AVV decay problem, very light axion
1240: bosons have been predicted but have {\it not} yet been
1241: observed \cite{PDG2004}.
1242:
1243: Also, there is the $U_A(1)$ and $\Theta$ problem involving gluons
1244: whereby strong interaction QCD leads to CP violation,
1245: definitely a ``strong CP problem'' because CP violation is known
1246: to occur at the $10^{-3}$ weak interaction amplitude level \cite{PDG2004}.
1247: Physicists have tried to circumvent this ``$U_A(1)$ --
1248: strong CP problem'' either via the topology of gauge
1249: fields or by investigating the $\Theta$--vacuum for this
1250: strong CP problem \cite{Mitra}.
1251:
1252: In this paper we have circumvented the need to deal directly
1253: with the above photon or gluon AVV anomalies by studying
1254: instead (finite) PVV quark triangle graphs. Then we have given
1255: our phenomenological results -- which always are in
1256: approximate agreement with the data. Next we return to
1257: the $U_A(1)$ problem and again use quark triangle
1258: diagrams coupled to 2 gluons. Invoking nonstrange--strange
1259: particle mixing, the predicted $U_A(1)$
1260: mass is within 3\% of data
1261: \cite{Jones:1979ez,Kekez:2000aw,Klabucar:2000me}.
1262:
1263: Thus we circumvent both photon and, admittedly on a
1264: much more speculative level, also the gluon ABJ anomaly
1265: without resorting either to unmeasured axions or to
1266: a strong CP violating term in the QCD Lagrangian.
1267:
1268:
1269: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1270:
1271: \section*{Acknowledgments}
1272: \noindent D. Klabu\v{c}ar gratefully acknowledges the partial support
1273: of the Abdus Salam ICTP at Trieste.
1274:
1275: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1276:
1277:
1278:
1279: \begin{thebibliography}{100}
1280:
1281: %\bibitem{ABJ} S. L. Adler, Phys. Rev. {\bf 177}, 2426 (1969); \\
1282: %J. S. Bell and R. Jackiw, Nuovo Cim. A {\bf 70}, 47 (1969).
1283: %% [1] Adler 1969; Bell, Jackiw 1969
1284:
1285: \bibitem{Adler69}
1286: S. L. Adler, {Phys. Rev.} {{\bf 177}} (1969) 2426.
1287:
1288: \bibitem{BellJackiw69}
1289: J. S. Bell and R. Jackiw, {Nouvo Cim. A} {\bf{60}} (1969) 47.
1290:
1291:
1292: \bibitem{seeGeorgi}
1293: See, for example, Ref. \cite{Georgi:1985kw} by H. Georgi,
1294: and Chap. 5 of its updated version available in electronic form at
1295: http://schwinger.harvard.edu/\~{}georgi/weak5.ps.
1296:
1297: %\bibitem{Georgi5}
1298: %H. Georgi, Chapter 5 (available at
1299: %http://schwinger.harvard.edu/~georgi/weak5.ps) of the updated version
1300: %of the monograph \cite{Georgi:1985kw} on weak interactions.
1301:
1302:
1303: %\cite{Georgi:1985kw}
1304: \bibitem{Georgi:1985kw}
1305: H.~Georgi,
1306: %``Weak Interactions And Modern Particle Theory,''
1307: %\href{http://www.slac.stanford.edu/spires/find/hep/www?irn=1459899}{SPIRES entry}
1308: ``Weak Interactions And Modern Particle Theory",
1309: Menlo Park, USA: Benjamin/Cummings (1984).
1310:
1311:
1312: \bibitem{WZ} J. Wess and B. Zumino, Phys. Lett. B {\bf 37} (1971) 95.
1313:
1314:
1315: \bibitem{S49}
1316: J. Steinberger, {Phys. Rev.} {\bf{76}} (1949) 1180.
1317:
1318:
1319: %\cite{Alkofer:2000wg}
1320: \bibitem{Alkofer:2000wg}
1321: R.~Alkofer and L.~von Smekal,
1322: %``The infrared behavior of QCD Green's functions: Confinement, dynamical
1323: %symmetry breaking, and hadrons as relativistic bound states,''
1324: Phys.\ Rept.\ {\bf 353} (2001) 281
1325: [arXiv:hep-ph/0007355].
1326: %%CITATION = HEP-PH 0007355;%%
1327:
1328:
1329: %\cite{Maris:2003vk}
1330: \bibitem{Maris:2003vk}
1331: P.~Maris and C.D.~Roberts,
1332: %``Dyson-Schwinger equations: A tool for hadron physics,''
1333: Int.\ J.\ Mod.\ Phys.\ E {\bf 12} (2003) 297
1334: [arXiv:nucl-th/0301049].
1335: %%CITATION = NUCL-TH 0301049;%%
1336:
1337:
1338: \bibitem{Delbourgo}
1339: R. Delbourgo, Dongsheng Liu and M. D. Scadron,
1340: Int. J. Mod. Phys. A {\bf 14} (1999) 4331 [arXiv:hep-ph/9905501];
1341: M. D. Scadron, F. Kleefeld, G. Rupp and E. van Beveren,
1342: Nucl. Phys. A {\bf 724} (2003) 391 [hep-ph/0211275].
1343:
1344: \bibitem{delbourgo95}R. Delbourgo and M. D. Scadron,
1345: Mod. Phys. Lett. A {\bf 10} (1995) 251 [arXiv:hep-ph/9910242].
1346:
1347: \bibitem{bramon98}A. Bramon, Riazuddin and M. D. Scadron,
1348: J. Phys. G {\bf 24} (1998) 1 [arXiv:hep-ph/9709274].
1349:
1350:
1351: \bibitem{PDG2004} S. Eidelman {\em et al.},
1352: Phys. Lett. B {\bf 592} (2004) 1.
1353:
1354: \bibitem{Mitra} H. Banerjee, D. Chatterjee and P. Mitra,
1355: Phys. Lett. B {\bf 573} (2003) 109 [arXiv:hep-ph/0012284].
1356: % [4] Mitra et al. hep-ph/0012284 ver4.
1357:
1358:
1359: \bibitem{Ad+al71Te72Av+Z72}
1360: S. L. Adler, B. W. Lee, S. Treiman and A. Zee,
1361: Phys. Rev. D {\bf 4} (1971) 3497;
1362: M. V. Terent'ev, Phys. Lett. {\bf 38B} (1972) 419;
1363: R. Aviv and A. Zee, Phys. Rev. D {\bf 5} (1972) 2372.
1364:
1365:
1366: \bibitem{HakiogluAndOthers}T. Hakioglu and M. D. Scadron,
1367: Phys. Rev. D {\bf 42} (1990) 941;
1368: T. Hakioglu and M. D. Scadron,
1369: Phys. Rev. D {\bf 43} (1991) 2439.
1370:
1371:
1372: \bibitem{Andrianov+al98}
1373: A. A. Andrianov, D. Espriu and R. Tarrach,
1374: Nucl. Phys. B {\bf 533} (1998) 429.
1375:
1376:
1377: \bibitem{Ametller+al83}
1378: Ll. Ametller, L. Bergstr\" om, A. Bramon and E. Mass\' o,
1379: Nucl. Phys. {\bf B 228} (1983) 301.
1380:
1381:
1382: %\cite{Bistrovic:1999yy}
1383: \bibitem{Bistrovic:1999yy}
1384: B.~Bistrovi\'{c} and D.~Klabu\v{c}ar,
1385: %``The quark loop calculation of the gamma $\to$ 3pi form factor,''
1386: Phys.\ Rev.\ D {\bf 61} (2000) 033006
1387: [arXiv:hep-ph/9907515].
1388: %%CITATION = HEP-PH 9907515;%%
1389:
1390:
1391: \bibitem{KeBiKl98}
1392: D. Kekez, B. Bistrovi\'{c} and D. Klabu\v{c}ar,
1393: Int. J. Mod. Phys. A {\bf 14} (1999) 161 [arXiv:hep-ph/9809245].
1394:
1395:
1396: \bibitem{Roberts}
1397: C.D. Roberts,
1398: Nucl. Phys. A {\bf 605} (1996) 475;
1399: C.D. Roberts, in: Chiral Dynamics: Theory and Experiment,
1400: eds. A. M. Bernstein and B. R. Holstein, Lecture Notes in Physics,
1401: Vol. 452, Springer, Berlin, 1995 p. 68.
1402:
1403:
1404: \bibitem{bando94}
1405: M. Bando, M. Harada and T. Kugo,
1406: Prog. Theor. Phys. {\bf 91} (1994) 927 [arXiv:hep-ph/9312343].
1407:
1408:
1409: \bibitem{AR96}
1410: R. Alkofer and C. D. Roberts, {Phys. Lett.} B {\bf{369}} (1996) 101
1411: {}[arXiv:hep-ph/9510284 ].
1412:
1413:
1414: %\cite{Bistrovic:1999dy}
1415: \bibitem{Bistrovic:1999dy}
1416: B.~Bistrovi\'{c} and D.~Klabu\v{c}ar,
1417: %``Anomalous gamma $\to$ 3pi amplitude in a bound-state approach,''
1418: Phys.\ Lett.\ B {\bf 478} (2000) 127
1419: [arXiv:hep-ph/9912452].
1420: %%CITATION = HEP-PH 9912452;%%
1421:
1422:
1423: \bibitem{KeKl1}
1424: D. Kekez and D. Klabu\v{c}ar, Phys. Lett. B {\bf 387} (1996) 14
1425: {}[arXiv:hep-ph/9605219].
1426:
1427: \bibitem{KlKe2}
1428: D. Klabu\v{c}ar and D. Kekez, Phys. Rev. D {\bf 58} (1998) 096003
1429: {}[arXiv:hep-ph/9710206].
1430:
1431: \bibitem{KeKl3}
1432: D. Kekez and D. Klabu\v{c}ar, Phys. Lett. B {\bf 457} (1999) 359
1433: {}[arXiv:hep-ph/9812495].
1434:
1435: %\cite{Kekez:2003ri}
1436: \bibitem{Kekez:2003ri}
1437: D.~Kekez and D.~Klabucar,
1438: %``Pseudoscalar q anti-q mesons and effective QCD coupling enhanced by
1439: %condensate,''
1440: Phys.\ Rev.\ D {\bf 71} (2005) 014004
1441: [arXiv:hep-ph/0307110].
1442: %%CITATION = HEP-PH 0307110;%%
1443:
1444:
1445: \bibitem{BC}
1446: J. S. Ball and T.-W. Chiu, Phys. Rev. D {\bf 22} (1980) 2542.
1447:
1448: \bibitem{Antipov+al87}
1449: Yu. M. Antipov {\it et al.}, Phys. Rev. D {\bf 36} (1987) 21.
1450:
1451: \bibitem{Moinester+al99}
1452: M. A. Moinester, V. Steiner and S. Prakhov,
1453: Hadron-Photon Interactions in COMPASS,
1454: to be published in the proceedings of 37th International Winter
1455: Meeting on Nuclear Physics, Bormio, Italy, 25-29 Jan 1999,
1456: [arXiv:hep-ex/9903017].
1457:
1458: \bibitem{Miskimen+al94}
1459: R. A. Miskimen, K. Wang and A. Yagneswaran (spokesmen),
1460: Study of the Axial Anomaly using the $\gamma\pi^+ \to \pi^+ \pi^0$
1461: Reaction Near Threshold, Letter of intent, CEBAF-experiment 94-015.
1462:
1463:
1464: %\cite{Klabucar:2000yd}
1465: \bibitem{Klabucar:2000yd}
1466: D.~Klabu\v{c}ar and B.~Bistrovi\'{c},
1467: %``The gamma $\to$ 3pi form factor as a constraint on Schwinger-Dyson modeling
1468: %of light quarks,''
1469: arXiv:hep-ph/0009259.
1470: %%CITATION = HEP-PH 0009259;%%
1471:
1472:
1473: %\cite{Klabucar:2000mk}
1474: \bibitem{Klabucar:2000mk}
1475: D.~Klabu\v{c}ar and B.~Bistrovi\'{c},
1476: %``The anomalous gamma $\to$ pi+ pi0 pi- form factor and the light-quark mass
1477: %functions at low momenta,''
1478: arXiv:hep-ph/0012273.
1479: %%CITATION = HEP-PH 0012273;%%
1480:
1481:
1482: %\cite{Sutherland:1967vf}
1483: \bibitem{Sutherland:1967vf}
1484: D.~G.~Sutherland,
1485: %``Current Algebra And Some Nonstrong Mesonic Decays,''
1486: Nucl.\ Phys.\ B {\bf 2} (1967) 433.
1487: %%CITATION = NUPHA,B2,433;%%
1488:
1489:
1490: \bibitem{VeltmanPRSLA67}
1491: M.~Veltman, Proc. R. Soc. London A{\bf 301}, 107 (1967).
1492:
1493:
1494: \bibitem{seeRef}
1495: Besides original references, some may also find helpful more
1496: pedagogical presentations such as that in Ref. \cite{Yndurain:1999ui}.
1497:
1498:
1499: \bibitem{Georgi6a}
1500: H. Georgi, Chapter 6a (available at
1501: http://schwinger.harvard.edu/~georgi/weak6a.ps) of the updated version
1502: of the monograph \cite{Georgi:1985kw} on weak interactions.
1503:
1504: \bibitem{PDG2002} K. Hagiwara {\em et al.},
1505: Phys. Rev. D {\bf 66}, 010001 (2002).
1506:
1507: %\cite{Kekez:2000aw}
1508: \bibitem{Kekez:2000aw}
1509: D.~Kekez, D.~Klabu\v{c}ar and M.~D.~Scadron,
1510: %``Revisiting the U(1)A problems,''
1511: J.\ Phys.\ G {\bf 26} (2000) 1335
1512: [arXiv:hep-ph/0003234].
1513: %%CITATION = HEP-PH 0003234;%%
1514:
1515: %\cite{Jones:1979ez}
1516: \bibitem{Jones:1979ez}
1517: H. F. Jones and M. D. Scadron,
1518: %``Gluon Effects And Nonet Mixing In The Current Quark Picture,''
1519: Nucl.\ Phys.\ B {\bf 155} (1979) 409.
1520: %%CITATION = NUPHA,B155,409;%%
1521:
1522: \bibitem{thew}
1523: R. L. Thews, Phys. Rev. D {\bf 10} (1974) 2993.
1524:
1525: \bibitem{freund-nandi-rudaz}P. G. O. Freund and S. Nandi,
1526: Phys. Rev. Lett. {\bf 32} (1974) 181;
1527: S. Rudaz, Phys. Lett. B {\bf 145} (1984) 281.
1528:
1529: \bibitem{gell-mann}M. Gell-Mann, D. Sharp and W. Wagner,
1530: Phys. Rev. Lett. {\bf 8} (1962) 261.
1531:
1532:
1533: \bibitem{lucio00}J. L. Lucio, M. Napsuciale, M. D. Scadron and
1534: V. M. Villanueva, Phys. Rev. D {\bf 61} (2000) 034013.
1535:
1536: %\cite{Klabucar:1997zi}
1537: \bibitem{Klabucar:1997zi}
1538: D.~Klabu\v{c}ar and D.~Kekez,
1539: %``eta and eta' at the limits of applicability of a coupled Schwinger-Dyson and
1540: %Bethe-Salpeter approach in the ladder approximation,''
1541: Phys.\ Rev.\ D {\bf 58}, 096003 (1998)
1542: [arXiv:hep-ph/9710206].
1543: %%CITATION = HEP-PH 9710206;%%
1544:
1545: %\cite{Klabucar:2000me}
1546: \bibitem{Klabucar:2000me}
1547: D.~Klabu\v{c}ar, D.~Kekez and M.~D.~Scadron,
1548: %``On the eta eta' complex in the SD-BS approach,''
1549: arXiv:hep-ph/0012267.
1550: %%CITATION = HEP-PH 0012267;%%
1551:
1552: %\cite{Klabucar:2001gr}
1553: \bibitem{Klabucar:2001gr}
1554: D.~Kekez, D.~Klabu\v{c}ar and M.~D.~Scadron,
1555: %``Dynamical SU(3) linear sigma model and the mixing of eta' eta and sigma f0
1556: %mesons,''
1557: J.\ Phys.\ G {\bf 27} (2001) 1775
1558: [arXiv:hep-ph/0101324].
1559: %%CITATION = HEP-PH 0101324;%%
1560:
1561:
1562: \bibitem{Choudhury} S. R. Choudhury and M. D. Scadron,
1563: Mod. Phys. Lett. A {\bf 1} (1986) 535.
1564: % [6] S.R. Choudhury???, MDS MPLA \bf{1} (1986) 535.
1565:
1566: %\cite{Yndurain:1999ui}
1567: \bibitem{Yndurain:1999ui}
1568: F.~J.~Yndurain,
1569: %``The theory of quark and gluon interactions,''
1570: %\href{http://www.slac.stanford.edu/spires/find/hep/www?irn=4416570}{SPIRES entry}
1571: ``The theory of quark and gluon interactions,''
1572: New York, Berlin, Heidelberg, Tokyo: Springer-Verlag (1993).
1573: % [ASIN: 3540558039.]
1574:
1575:
1576: \end{thebibliography}
1577:
1578: \end{document}
1579:
1580:
1581:
1582:
1583:
1584:
1585: