1: \documentstyle[aps,epsf,preprint]{revtex}
2: \begin{document}
3: \draft
4: \title{Radiative decays of \boldmath{$\phi$}-meson and
5: nature of light scalar resonances}
6: \author{N.N. Achasov}
7: \address{Laboratory of Theoretical Physics,
8: Sobolev Institute for Mathematics,
9: Novosibirsk, 630090, Russia}
10: \date{\today}
11: \maketitle
12: \begin{abstract}
13: Based on gauge invariance, we show that the new threshold
14: phenomenon is discovered in the $\phi$ radiative decays
15: $\phi\to\gamma a_0\to\gamma\pi^0\eta$ and
16: $\phi\to\gamma f_0\to\gamma\pi^0\pi^0$. This enables to conclude
17: that production of the lightest scalar mesons $a_0(980)$ and
18: $f_0(980)$ in these decays
19: is caused by the four-quark transitions, resulting in strong restrictions on the
20: large $N_C$ expansions of the decay amplitudes. The analysis
21: shows that these constraints give new evidences in favor
22: of the four-quark nature of $a_0(980)$ and $f_0(980)$ mesons.
23: \end{abstract}
24: \pacs{ PACS number(s): 12.39.-x, 13.40.Hq, 13.65.+i}
25: The discovered more than thirty years ago the lightest scalar mesons
26: $a_0(980)$ and $f_0(980)$ became the hard problem for the naive
27: quark-antiquark ($q\bar q$) model from the outset. Really, on the
28: one hand the almost exact degeneration of the masses of the
29: isovector $a_0(980)$ and isoscalar $f_0(980)$ states revealed
30: seemingly the structure similar to the structure of the vector
31: $\rho$ and $\omega$ mesons , and on the other hand the strong
32: coupling of $f_0(980)$ with the $K\bar K$ channel pointed
33: unambiguously to a considerable part of the strange quark pair
34: $s\bar s$ in the wave function of $f_0(980)$.
35:
36: In 1977 R.L. Jaffe noted that in the MIT bag model, which
37: incorporates confinement phenomenologically, there are light
38: four-quark scalar states \cite{jaffe}. He suggested that
39: $a_0(980)$ and $f_0(980)$ might be these states with symbolic
40: structures $a^0_0(980)=(us\bar u\bar s - ds\bar d\bar s)/\sqrt{2}$
41: and $f_0(980)=(us\bar u\bar s + ds\bar d\bar s)/\sqrt{2}$. From
42: that time $a_0(980)$ and $f_0(980)$ resonances came into beloved
43: children of the light quark spectroscopy, see, for example,
44: reviews \cite{montanet,achasov-84}.
45:
46: Ten years later we showed \cite{achasov-89} that the study of the
47: radiative decays $\phi\to\gamma a_0\to\gamma\pi\eta$ and
48: $\phi\to\gamma f_0\to \gamma\pi\pi$ can shed light on the problem
49: of $a_0(980)$ and $f_0(980)$ mesons. Over the next ten years the
50: question was considered from different points of view
51: \cite{bramon,achasov-97,achasov-97a,achasov-97b}.
52:
53: Now these decays have been studied not only theoretically but also
54: experimentally, so that the time is ripe to sum up.
55:
56: Present time data have already been obtained from Novosibirsk with
57: the detectors SND
58: \cite{snd-fit,snd-ivan} and CMD-2 \cite{cmd},
59: which give the following
60: branching ratios :
61: $BR(\phi\to\gamma\pi^0\eta)=(0.88\pm0.14\pm0.09)\cdot10^{-4}$
62: \cite{snd-fit} (2000), $BR(\phi\to\gamma\pi^0\pi^0)=
63: (1.221\pm0.098\pm0.061)\cdot10^{-4}$ \cite{snd-ivan} (2000) and
64: $BR(\phi\to\gamma\pi^0\eta)=(0.9\pm0.24\pm0.1)\cdot10^{-4}$,
65: $BR(\phi\to\gamma\pi^0\pi^0)=(0.92\pm0.08\pm0.06)\cdot10^{-4}$
66: \cite{cmd}. DA$\Phi$NE also confirms the Novosibirsk results
67: \cite{barbara}.
68:
69: These data give evidence in favor of the four-quark $(q^2\bar
70: q^2)$ \cite{jaffe,achasov-84,black,josef} nature of $a_0(980)$
71: and $f_0(980)$ mesons \cite{achasov-89,achasov-97,achasov-01}.
72: Note that the isovector $a_0(980)$ meson is produced in the
73: radiative $\phi$ meson decay
74: as intensively as the well-studied $\eta '(958)$ meson containing $\approx 66\%$ of $s\bar s$,
75: responsible for the decay .
76: It is a clear qualitative argument
77: for the presence of the $s\bar s$ pair in the isovector $a_0(980)$
78: state, i.e., for its four-quark nature.
79:
80: Since the one-loop model $\phi\to K^+K^-\to\gamma a_0$ and
81: $\phi\to K^+K^-\to\gamma f_0$, suggested at basing the
82: experimental investigations \cite{achasov-89}, was used in the
83: data treatment from the outset, the question on the mechanism of
84: the scalar meson production in the $\phi$ radiative decays was put
85: into the shade. We show below that the present data give the
86: conclusive arguments in favor of the $K^+K^-$ loop mechanism of
87: $a_0(980)$ and $f_0(980)$ mesons production in the $\phi$
88: radiative decays. The knowledge of this mechanism allows to
89: conclude that the production of $a_0(980)$ and $f_0(980)$ in the
90: $\phi$ radiative decays is caused by the four-quark transitions.
91: This constrains the large
92: $N_C$ expansions of the decay amplitudes and gives new impressive evidences in favor
93: of the four-quark nature of $a_0(980)$ and $f_0(980)$.
94:
95: In Figs. \ref{figa0} and \ref{figf0} are shown the SND data on
96: $\phi\to\gamma\pi^0\eta$ \cite{snd-fit} (2000) and
97: $\phi\to\gamma\pi^0\pi^0$ \cite{snd-ivan} (2000). The data are
98: described in the following model $\phi\to(\gamma
99: a_0+\pi^0\rho)\to\gamma\pi^0\eta$ and $\phi\to(\gamma
100: f_0+\pi^0\rho)\to\gamma\pi^0\pi^0$, see details in Ref.
101: \cite{achasov-01}. It is important for us now nothing but the fact
102: that the $\phi\to\gamma a_0\to\gamma\pi^0\eta$ process dominates
103: everywhere over the region of the $\pi^0\eta$ invariant mass
104: $m_{\pi^0\eta}= m$ and the $\phi\to\gamma f_0\to\gamma\pi^0\pi^0$
105: process dominates in the resonance region of the $\pi^0\pi^0$
106: system, the $\pi^0\pi^0$ invariant mass $m_{\pi^0\pi^0}=m > 670$
107: MeV \cite{cut}.
108:
109: The resonance mass spectrum is of the form \cite{mixing}
110:
111: \begin{eqnarray}
112: &&S_R(m)=d\Gamma(\phi\to\gamma R\to\gamma ab\,,\, m)/dm\nonumber\\
113: &&=\frac{2}{\pi}\frac{m^2\Gamma(\phi\to\gamma R\,,\,m)\Gamma(R\to
114: ab\,,\,m)}{|D_R(m)|^2} = \frac{4|g_R(m)|^2\omega (m)
115: p_{ab}(m)}{3(4\pi)^3m_{\phi}^2}\left |\frac{g_{Rab}}{D_R(m)}\right
116: |^2,
117: \label{spectrumR}
118: \end{eqnarray}
119: where $R = a_0\ \mbox{or}\ f_0$ and $ab=\pi^0\eta\ \mbox{or}\
120: \pi^0\pi^0$ respectively, $\omega (m)=(m_{\phi}^2-m^2)/2m_{\phi}$
121: is the photon energy in the $\phi$ meson rest frame, $p_{ab}(m)$
122: is the modulus of the $a$ or $b$ particle momentum in the $a$ and
123: $b$ mass center frame, $g_{Rab}$ is the coupling constant,
124: $g_{f_0\pi^0\pi^0}=g_{f_0\pi^+\pi^-}/\sqrt{2}$, $D_R(m)$ is the
125: $R$ resonance propagator the form of which everywhere over the $m$
126: region can be find in \cite{achasov-89,achasov-80,achasov-01a},
127: $g_R(m)$ is the invariant amplitude that describes the vertex of
128: the $\phi (p)\to\gamma (k) R(q)$ transition with $q^2=m^2$. This
129: is precisely the function which is the subject of our
130: investigation.
131:
132: By gauge invariance, the transition amplitude is proportional to
133: the electromagnetic field strength tensor $F_{\mu\nu}$ (in our
134: case to the electric field in the $\phi$ meson rest frame):
135: \begin{eqnarray}
136: && A\left [\phi(p)\to\gamma (k) R(q)\right ]= G_R(m)\left [p_\mu
137: e_\nu(\phi) - p_\nu e_\mu(\phi)\right]\left [k_\mu e_\nu(\gamma) -
138: k_\nu e_\mu(\gamma)\right],
139: \label{gauge}
140: \end{eqnarray}
141: where $e(\phi)$ and
142: $e(\gamma)$ are the $\phi$ meson and $\gamma$ quantum polarization
143: four-vectors, $G_R(m)$ is the invariant amplitude free from
144: kinematical singularities. Since there are no charge particles or
145: particles with magnetic moments in the process, there is no pole
146: in $G_R(m)$. Consequently, the function
147: \begin{eqnarray}
148: g_R(m)= - 2(pk)G_R(m) = - 2\omega (m) m_\phi G_R(m) \label{gRm}
149: \end{eqnarray}
150: is proportional to $\omega (m)$ (at least!) in the soft photon
151: region.
152:
153: To describe the experimental spectra in Figs. \ref{figa0} and
154: \ref{figf0}, the function $|g_R(m)|^2$ should be smooth (almost
155: constant) in the range $m\leq 0.99$ GeV, see Eq.
156: (\ref{spectrumR}). Stopping the function $(\omega (m))^2$ at
157: $\omega (990\,\mbox{MeV})=29$ MeV with the help of the form-factor
158: $1/\left [1+(R\omega (m))^2\right ]$ requires $R\approx 100$
159: GeV$^{-1}$. It seems to be incredible to explain the formation of
160: such a huge radius in hadron physics. Based on the large, by
161: hadron physics standard, $R\approx10$ GeV$^{-1}$, one can obtain
162: an effective maximum of the mass spectrum under discussion only
163: near 900 MeV. To exemplify this trouble let us consider the
164: contribution of the isolated R resonance: $g_R(m)=-2\omega (m)
165: m_\phi G_R\left (m_R\right )$. Let also the mass and the width of
166: the R resonance equal 980 MeV and 60 MeV, then
167: $S_R(920\,\mbox{MeV}):S_R(950\,\mbox{MeV}):S_R(970\,\mbox{MeV}):S_R(980\,\mbox{MeV})
168: =3:2.7:1.8:1$.
169:
170: So stopping the $g_R(m)$ function is the crucial point in
171: understanding the mechanism of the production of $a_0(980)$ and
172: $f_0(980)$ resonances in the $\phi$ radiative decays.
173:
174: The $K^+K^-$ loop model $\phi\to K^+K^-\to\gamma R$
175: \cite{achasov-89} solves this problem in the elegant way: the fine
176: threshold phenomenon is discovered, see Fig. \ref{g}, where the
177: universal in $K^+K^-$ loop model function $|g(m)|^2=\left
178: |g_R(m)/g_{Rab}\right |^2$ is shown \cite{gm}.
179:
180: To demonstrate the threshold character of this effect we present
181: Fig. \ref{gg} and Fig. \ref{ggg} in which the function $|g(m)|^2$
182: is shown in the case of $K^+$ meson mass is 25 MeV and 50 MeV less
183: than in reality.
184:
185: One can see from Figs. \ref{gg} and \ref{ggg} that the function
186: $|g(m)|^2$ is suppressed by the $(\omega (m))^2$ law in the region
187: 950-1020 MeV and 900-1020 Mev respectively \cite{phone}.
188:
189: In the mass spectrum this suppression is increased by one more
190: power of $\omega (m)$, see Eq. (\ref{spectrumR}), so that we
191: cannot see the resonance in the region 980-995 MeV. The maximum in
192: the spectrum is effectively shifted to the region 935-950 MeV and
193: 880-900 MeV respectively.
194:
195: In truth this means that $a_0(980)$ and $f_0(980)$ resonances are
196: seen in the radiative decays of $\phi$ meson owing to the $K^+K^-$
197: intermediate state, otherwise the maxima in the spectra would be
198: shifted to 900 MeV.
199:
200: So the mechanism of production of $a_0(980)$ and $f_0(980)$
201: mesons in the $\phi$ radiative decays is established.
202:
203:
204: Both the real part and the imaginary one of the $\phi\to\gamma R$
205: amplitude are caused by the $K^+K^-$ intermediate state. The
206: imaginary part is caused by the real $K^+K^-$ intermediate state
207: while the real part is caused by the virtual compact $K^+K^-$
208: intermediate state, i.e., we are dealing here with the four-quark
209: transition \cite{im} . Needless to say, radiative four-quark
210: transitions can happen between two $q\bar q$ states as well as
211: between $q\bar q$ and $q^2\bar q^2$ states but their intensities
212: depend strongly on a type of the transitions. A radiative
213: four-quark transition between two $q\bar q$ states requires
214: creation and annihilation of an additional $q\bar q$ pair, i.e.,
215: such a transition is forbidden according to the Okuba-Zweig-Izuka
216: (OZI) rule, while a radiative four-quark transition between $q\bar
217: q$ and $q^2\bar q^2$ states requires only creation of an
218: additional $q\bar q$ pair, i.e., such a transition is allowed
219: according to the OZI rule.
220:
221: Let us consider this problem from the large $N_C$ expansion
222: standpoint. Let us begin with the $q\bar q$ model.
223:
224: In the two-quark model $a^0_0(980)=(u\bar u - d\bar d)/\sqrt{2}$,
225: the large $N_C$ expansion of the $\phi = s\bar s\to\gamma
226: a_0(980)$ amplitude starts with the OZI forbidden transition of
227: the $1/N_C$ order ( $s\bar s$ annihilation and $u\bar u,\, d\bar
228: d$ creation). But its weight is bound to be small, because this
229: term does not contain the $K^+K^-$ intermediate state emerging
230: only in the next to leading term of the $\left (1/N_C\right )^2$
231: (!) order for creation and annihilation of additional $q\bar q$
232: pairs. It will be recalled that $\phi\to\gamma\eta ' (958)$
233: transition does not require creation of an additional $q\bar q$
234: pair at all (the OZI superallowed transition) and has the $\left
235: (N_C\right )^0$ order.
236:
237: In the two-quark model $f_0(980)=(u\bar u + d\bar d)/\sqrt{2}$,
238: which involves the $a_0$\,-$f_0$ mass degeneration, the large
239: $N_C$ expansion of the $\phi = s\bar s\to\gamma f_0(980)$
240: amplitude starts also with the OZI forbidden transition of the
241: $1/N_C$ order, whose weight also is bound to be small, because
242: this term does not contain the $K^+K^-$ intermediate state
243: emerging only in the next to leading term of the $\left
244: (1/N_C\right )^2$ order.
245:
246: In the two-quark model $f_0(980)=s\bar s$, which has the serious
247: trouble with the $a_0$\,-$f_0$ mass degeneration, the $\left
248: (N_C\right )^0$ order transition without creation of an additional
249: $q\bar q$ pair ( similar in this regard to $\phi\to\gamma\eta '
250: (958)$ ) is bound to have a small weight in the large $N_C$
251: expansion of the $\phi = s\bar s\to\gamma f_0(980)$ amplitude,
252: because this term does not contain the $K^+K^-$ intermediate state
253: emerging only in the next to leading term of the $1/N_C$ order,
254: i.e., in the OZI forbidden transition. Emphasize that the
255: mechanism without creation and annihilation of the additional
256: $u\bar u$ pair cannot explain the $S_{f_0}(m)$ spectrum because
257: it does not contain the $K^+K^-$ intermediate state!
258:
259: But if $a^0_0(980)$ and $f_0(980)$ mesons are compact $K\bar K$
260: states $(u\bar ss\bar u - d\bar ss\bar d)/\sqrt{2}$ and $(u\bar
261: ss\bar u + d\bar ss\bar d)/\sqrt{2}$
262: respectively, i.e., four-quark states similar to states of
263: the MIT-bag model \cite{molecule}, the large $N_C$ expansions of
264: the $\phi = s\bar s\to\gamma a_0(980)$ and $\phi = s\bar
265: s\to\gamma f_0(980)$ amplitudes start with the OZI allowed
266: transitions of the $\left (1/N_C\right )^{-1/2}$ order, which
267: require only creation the additional $u\bar u$ pair for the
268: $K^+K^-$ intermediate state. It will be recalled that a OZI
269: allowed hadronic decay amplitude, for example, the $\rho\to\pi\pi$
270: amplitude, has the $\left (1/ N_C\right ) ^{-1/2}$ order.
271:
272:
273: In summary the fine threshold phenomenon is discovered, which is
274: to say that the $K^+K^-$ loop mechanism of
275: the $a_0(980)$ and $f_0(980)$ scalar meson production in the
276: $\phi$ radiative decays is established at a physical level of
277: proof. The case is rarest in hadron physics. This production
278: mechanism is the four-quark transition what constrains the large
279: $N_C$ expansion of the $\phi\to\gamma a_0(980)$ and $\phi\to\gamma
280: f_0(980)$ amplitudes and gives the new strong (if not crucial)
281: evidences in favor of the four-quark nature of $a_0(980)$ and
282: $f_0(980)$ mesons.
283:
284: \begin{references}
285: \bibitem{jaffe}
286: R.L. Jaffe, Phys. Rev. D {\bf 15}, 267, 281 (1977).
287: \bibitem{montanet}
288: L. Montanet, Rep. Prog. Phys. {\bf 46}, 337 (1983); F.E. Close,
289: Rep. Prog. Phys. {\bf 51}, 833 (1988); L.G. Landsberg, Usp. Fiz.
290: Nauk. {\bf 160}, 3 (1990) [Sov. Phys. Usp. {\bf 33}, 1 (1990)];
291: M.R. Pennington, HADRON '95, Proceedings of the 6h International
292: Conference on Hadron Spectroscopy, Manchester, UK, 10th-14th July
293: 1995, Eds. M.C. Birse, G.D. Lafferty, and J.A. McGovern, page 3,
294: World Scientific, Singapore; T. Barnes, HADRON SPECTROSCOPY, VII
295: International Conference, Upton, NY August 1997, Eds. S.U. Chung,
296: H.J. Willutzki, AIP CONFERENCE PROCEEDINGS 432, page 3; C. Amsler,
297: Rev.Mod.Phys. {\bf 70}, 1293 (1998); Muneyuki Ishida, Shin Ishida,
298: and Taku Ishida, Prog.Theor.Phys. {\bf 99}, 1031 (1998); R.
299: Delbourgo and M.D. Scadron, Int. J. Mod. Phys. A {\bf 13}, 657
300: (1998); S. Godfrey and J. Napolitano, Rev. Mod. Phys. {\bf 71},
301: 1411 (1999); P. Minkowski and W. Ochs, Eur. Phys. J. C {\bf 9},
302: 283 (1999); S. Spanier and N.A. T\"{o}rnqvist, Eur. Phys. J. C
303: {\bf 15}, 437 (2000); E. van Beveren, hep-ph/0201006.
304: \bibitem{achasov-84}
305: N.N. Achasov, S.A. Devyanin and G.N. Shestakov, Usp. Fiz. Nauk.
306: {\bf 142}, 361 (1984) [Sov. Phys. Usp. {\bf 27}, 161 (1984)]; N.N.
307: Achasov and G.N. Shestakov, Usp. Fiz. Nauk. {\bf 161} (6), 53
308: (1991)[Sov. Phys. Usp. {\bf 34} (6), 471 (1991)]; N.N. Achasov,
309: Nucl. Phys. B (Proc. Suppl.) {\bf 21}, 189 (1991); N.N. Achasov,
310: Usp. Fiz. Nauk. {\bf 168}, 1257 (1998) [ Phys. Usp. {\bf 41}, 1149
311: (1998)]; N.N. Achasov, Nucl. Phys. A {\bf 675}, 279c (2000).
312: \bibitem{achasov-89}
313: N.N. Achasov and V.N. Ivanchenko, Nucl. Phys. B {\bf 315}, 465
314: (1989); Preprint INP 87-129 (1987) Novosibirsk.
315: \bibitem{bramon}
316: A. Bramon, A. Grau, G. Pancheri, Phys.Lett. B {\bf 289}, 97
317: (1992); F.E. Close, N. Isgur, and S. Kumano, Nucl. Phys. B {\bf
318: 389}, 513 (1993); E. Marco, S. Hirenzaki, E. Oset, and H. Toki,
319: Phys. Lett. B {\bf 470}, 20 (1999); J.L. Lucio, M. Napsuciale,
320: Contribution to 3rd Workshop on Physics and Detectors for DAPHNE
321: (DAPHNE 99), Frascati, Italy, 1999.
322: \bibitem{achasov-97}
323: N.N. Achasov and V.V. Gubin, Phys. Rev. D {\bf 56}, 4084 (1997).
324: \bibitem{achasov-97a}
325: N.N. Achasov, V.V. Gubin, and V.I. Shevchenko, Phys. Rev. D {\bf
326: 56}, 203 (1997).
327: \bibitem{achasov-97b}
328: N.N. Achasov, V.V. Gubin, and E.P. Solodov, Phys. Rev. D {\bf 55},
329: 2672 (1997);
330: N.N. Achasov and V.V. Gubin, Phys. Rev. D {\bf 57}, 1987
331: (1998).
332: \bibitem{snd-fit}
333: M.N. Achasov et al., Phys. Lett. B {\bf 438}, 441 (1998); {\bf
334: 479}, 53 (2000).
335: \bibitem{snd-ivan}
336: M.N. Achasov et al., Phys. Lett. B {\bf 440}, 442 (1998); {\bf
337: 485}, 349 (2000).
338: \bibitem{cmd}
339: R.R. Akhmetshin et al., Phys. Lett. B {\bf 462}, 380 (1999).
340: \bibitem{barbara}
341: Barbara Valeriani, Parallel session talk at HADRON 2001,
342: Proceedings.
343: \bibitem{black}
344: O. Black, A. Fariborz, F. Sannino, and J. Schechter, Phys.Rev. D
345: {\bf 59}, 074026 (1999).
346: \bibitem{josef}
347: HADRON 2002, Proceedings: J. Schechter (Plenary session talk),
348: hep-ph/0110356; S.F. Tuan ( Parallel session talk),
349: hep-ph/0109191; T. Teshima (Parallel session talk),
350: hep-ph/0105107.
351: \bibitem{achasov-01}
352: N.N. Achasov and V.V. Gubin, Phys. Rev. D {\bf 63}, 094007 (2001).
353: \bibitem{cut}
354: The difference from the experimental data, observed in the region
355: $m_{\pi\pi}<670$ MeV, is due to the fact that in the experimental
356: processing of the $e^+e^-\to\gamma\pi^0\pi^0$ events the
357: background events $e^+e^-\to\omega\pi^0\to\gamma\pi^0\pi^0$ are
358: excluded with the help of the invariant mass cutting, in so doing
359: the part of the $e^+e^-\to\phi\to\rho\pi^0\to\gamma\pi^0\pi^0$
360: events is excluded as well.
361: \bibitem{mixing}
362: Notice that in Ref. \cite{achasov-01} we took into account the
363: mixing of $f_0(980)$ meson with another scalar isoscalar
364: resonance, see also Ref. \cite{achasov-97}, but such a
365: complication is not essential for the present investigation.
366: \bibitem{achasov-80}
367: N.N. Achasov, S.A. Devyanin, and G.N. Shestakov, Yad. Fiz. {\bf
368: 32}, 1098 (1980)[Sov. J. Nucl. Phys. {\bf 32}, 566 (1980)].
369: \bibitem{achasov-01a}
370: N.N. Achasov and V.V. Gubin, Phys. Rev. D {\bf 64}, 094016 (2001).
371: \bibitem{gm}
372: The forms of $g_R(m)$ and $g(m)=g_R(m)/g_{Rab}$ everywhere over
373: the $m$ region are in Refs. \cite{achasov-89} and
374: \cite{achasov-01a} respectively.
375: \bibitem{phone}
376: The actual absence of a background at a soft photon energy region
377: $\omega (m)> 112$ MeV ($m > 900$ MeV) owes to gauge invariance
378: also.
379: \bibitem{im}
380: It will be recalled that the imaginary part of every hadronic
381: amplitude describes a multi-quark transition.
382: \bibitem{molecule}
383: In the case of the $K\bar K$ bound states with the binding energy
384: close to 20 MeV , i.e.,in the extended molecule case, the
385: contribution of the virtual intermediate $K^+K^-$ states in the
386: $K^+K^-$ loop is suppressed by the momentum distribution in the
387: molecule, and the real part of the loop amplitudes are negligible
388: \cite{achasov-97a}. It leads to the branching ratios much less
389: than the experimental ones. In addition, the $S_R(m)$ spectra in
390: the $K\bar K$ molecule case are much narrower than the
391: experimental ones, see the behavior of the imaginary part
392: contribution in Fig. \ref{g}.
393: \end{references}
394:
395:
396:
397: \begin{figure}
398: \centerline{\epsfxsize=14cm \epsfysize=9cm \epsfbox{figa0.eps}}
399: \caption{ Fitting of $dBR(\phi\to\gamma\pi^0\eta\,,\,m)/dm\times 10^4\mbox{GeV}^{-1}$ is shown with the solid line,
400: the signal contribution $\phi\to\gamma a_0\to\gamma\pi^0\eta$ is shown with the dashed line.} \label{figa0}
401: \end{figure}
402:
403: \begin{figure}
404: \centerline{\epsfxsize=14cm \epsfysize=9cm
405: \epsfbox{figf0sndi.eps}}
406: \caption{
407: Fitting of $dBR(\phi\to\gamma\pi^0\pi^0\,,\,m)/dm\times
408: 10^{4}\mbox{GeV}^{-1}$ is shown with the solid line, the signal
409: contribution $\phi\to\gamma f_0\to\gamma\pi^0\pi^0$ is shown with
410: the dashed line. The dots show the interference.} \label{figf0}
411: \end{figure}
412:
413:
414: \begin{figure}
415: \centerline{\epsfxsize=14cm \epsfysize=8.5cm \epsfbox{g.eps}}
416: \caption{ The function $|g(m)|^2$ is drawn with the solid line. The contribution of the
417: imaginary part is drawn with the dashed line. The contribution of the real part
418: is drawn with the dotted line.} \label{g}
419: \end{figure}
420:
421:
422: \begin{figure}
423: \centerline{\epsfxsize=14cm \epsfysize=8.5cm \epsfbox{gg.eps}}
424: \caption{ The function $|g(m)|^2$ for $m_{K^+}=469$ MeV is drawn
425: with the solid line. The contributions of the imaginary and real
426: parts are drawn with the dashed and dotted lines respectively.
427: }
428: \label{gg}
429: \end{figure}
430:
431: \begin{figure}
432: \centerline{ \epsfxsize=14cm \epsfysize=8.5cm \epsfbox{ggg.eps}}
433: \caption{ The function $|g(m)|^2$ for $m_{K^+}=444$ MeV is drawn
434: with the solid line. The contributions of the imaginary and real
435: parts are drawn with the dashed and dotted lines respectively.
436: } \label{ggg}
437: \end{figure}
438:
439:
440:
441:
442: \end{document}
443: