1:
2:
3: %version 04/11/00
4: \documentclass[twocolumn,showpacs]{revtex4}
5: \usepackage{psfig}
6: %\documentstyle[prc,twocolumn,aps,epsfig]{revtex4}
7:
8: \begin{document}
9: \draft
10:
11:
12: %\title{\qquad\qquad\qquad\qquad\qquad\qquad\qquad\hfil{NT@UW-00-07,KRL MAP-266}\\
13: \title{\hfill{\tiny FZJ--IKP(TH)--2004--12} \\[1.8em]
14: Comment on 'Photoproduction of $\eta$--Mesic ${^3}$He'}
15:
16: \author{C. Hanhart}
17:
18: \address{IKP(TH), Forschungszentrum J\"ulich, D--52428 J\"ulich, Germany}
19:
20:
21: \begin{abstract}
22: In a recent paper by the TAPS collaboration \cite{exp} a first measurement of
23: a bound system of an $\eta$ meson and a ${^3}$He nucleus was reported.
24: In this comment we critically reexamine the interpretation of
25: the data and show that the data prefers a solution where there is
26: no bound state present. Given the low statistics of the measurement,
27: however,
28: it does not exclude the existence of a bound state.
29: \end{abstract}
30:
31: \pacs{13.75.Cs, 14.20.Gk, 14.40.Aq, 14.40.Cs}
32:
33:
34: \maketitle
35:
36:
37: \vspace{0.8cm}
38: \newcommand{\boldpi}{\mbox{\boldmath $\pi$}}
39: \newcommand{\boldtau}{\mbox{\boldmath $\tau$}}
40: \newcommand{\boldT}{\mbox{\boldmath $T$}}
41: \newcommand{\gaprox}{$ {\raisebox{-.6ex}{{$\stackrel{\textstyle >}{\sim}$}}} $}
42: \newcommand{\saprox}{$ {\raisebox{-.6ex}{{$\stackrel{\textstyle <}{\sim}$}}} $}
43:
44: The interaction of $\eta$ mesons with nucleons is strong and attractive due
45: mainly to the presence of the $S_{11}(1535)$ resonance that strongly couples
46: to this system. Consequently it is expected that the $\eta$ meson should be
47: bound in sufficiently heavy nuclei. So far, however, it is unclear what mass
48: number is sufficient. Some authors predicted a bound state to occur on nuclei
49: as light as ${^3}$He~\cite{Wycech,Belyaev1,Belyaev2,Rakityansky,Fix1,Wilkin1},
50: whereas others expect binding only for heavier nuclei
51: \cite{Haider1,Liu,Chiang,Haider2}. Until recently no direct experimental
52: evidence for the existence of $\eta$--mesic nuclei was available. Only the
53: presence of a strong $\eta$--nucleus interaction was seen experimentally in
54: strong final state interaction effects in reactions like $pn\to \eta
55: d$\cite{Calen}, $pd\to \eta {^3}$He\cite{Berger,Mayer}, and $dd\to \eta
56: {^4}$He\cite{Frascaria}.
57:
58: Thus it was a big step forward from the experimental side when this year the
59: TAPS collaboration reported positive evidence for $\eta$--mesic ${^3}$He.
60: Besides a strong deviation in the angular shape of $\gamma {^3}\mbox{He}\to
61: \eta {^3}$He from the expectation for quasi--free production
62: (the cross section is flat instead of forward peaked), a structure
63: was observed in the cross section $\gamma {^3}\mbox{He}\to
64: \pi^0pX$ just below the $\eta$ production threshold. These signatures were
65: taken as strong evidence for the existence of $\eta$--mesic ${^3}$He.
66:
67: It should be clear that the former evidence---a flat $\eta {^3}$He angular
68: distribution in the close--to--threshold regime is a hint solely for a strong
69: $s$--wave $\eta {^3}$He interaction that leads to a relative suppression of the
70: impulse term with respect to the $s$--wave multiple scattering terms. Thus,
71: given what we already know about the strong $\eta {^3}$He interaction,
72: a flat angular
73: distribution in the close--to--threshold regime should be expected. A closer
74: look at the structure in the cross section $\gamma {^3}\mbox{He}\to
75: \pi^0pX$ is the focus of this comment.
76:
77: \begin{figure}[t]
78: \vspace*{-0mm}\hspace*{-3mm}
79: \psfig{file=onlyres_banb_pap.eps,width=8.cm,height=6.cm}\vspace*{-2mm}
80: \caption{Comparison of the various fits to the data, as a function of the
81: reduced photon energy $W$ defined in Ref. \protect\cite{exp}. Shown are the
82: results in the absence of a background interference ($B=0$ in Eq.
83: (\protect\ref{full})). The left panel corresponds to the calculation using
84: the same binning as the data, whereas the right panel shows the results
85: with no binning. The solid line corresponds to $a=(+4,1)$ fm, the dashed
86: one to $a=(-4,1)$ fm and the dotted one to $a=(0,3.5)$ fm.
87: The vertical line at $W=1486.4$ MeV indicates the position of the $\eta {^3}$He threshold.}
88: \label{resonly}
89: \end{figure}
90:
91: The structure reported by the TAPS collaboration was fitted with a
92: Breit--Wigner function. In its non--relativistic form the scattering
93: amplitude then is
94: \begin{equation}
95: f_{BW} \propto \left(E-E_R+\frac i2 \Gamma\right)^{-1} \ ,
96: \label{BW}
97: \end{equation}
98: where $\Gamma$ is assumed to be constant.
99: The parameters deduced were $(4.4{\pm}4.1)$ MeV and
100: $(25.6{\pm6.1})$ MeV for the binding energy and the width, respectively.
101: However, since the position of the signal
102: coincides with the $\eta$
103: production threshold, one might wonder whether this is more a cusp than
104: the signal of a bound state.
105:
106: Already in 1976, when studying the light scalar mesons $a_0$ and $f_0$,
107: Flatt{\'e} observed that in the presence of thresholds the Breit--Wigner form
108: of Eq. (\ref{BW}) is to be modified to include the momentum--dependence of
109: the elastic width \cite{flatte} (for a more recent discussion of threshold
110: effects in various system we refer to Ref. \cite{Bugg}). Thus Eq. (\ref{BW})
111: should be changed according to
112: $$
113: \Gamma \to \Gamma_{inel}+\Gamma_{el} \ ,
114: $$
115: where $\Gamma_{inel}$ and $\Gamma_{el}$ denote the inelastic and the elastic
116: width (in our case with respect to the $\eta {^3}$He channel)
117: of the resonance respectively. Here $\Gamma_{inel}$ can be assumed constant; however,
118: $\Gamma_{el}$ has to vanish at the elastic threshold! Thus, for an $s$--wave
119: structure, one gets
120: $$
121: \Gamma_{el} = g_{eff}k \ ,
122: $$
123: where $k$ denotes the momentum of the $\eta$ relative to the ${^3}$He nucleus
124: and, above the production threshold, may be written as $k=\sqrt{2\mu E}$. Here
125: $\mu$ denotes the reduced mass of the $\eta$He system and $E$ is its kinetic
126: energy. In the region below threshold, however, $k=i\sqrt{-2\mu E}$.
127: Thus, we find that if a structure is predominantly inelastic, a Breit--Wigner
128: might still be a good approximation, even in the proximity of a threshold;
129: however, if a structure is predominantly elastic, using a Breit--Wigner is not
130: justified.
131:
132: A dynamically generated singularity, like a bound state, also dominates
133: the final state interaction in the $\eta$He system. In addition,
134: if a production reaction is short--ranged (typical
135: momentum transfer significantly larger than any other scale of the problem)
136: the final state interaction is universal (independent of the reaction) and can
137: be related to the elastic scattering of the outgoing particles~\cite{wam}
138: which reads in the effective range approximation
139: \begin{equation}
140: f_{sc} \propto \left(1/a+rk^2/2-ik\right)^{-1} \ .
141: \label{era}
142: \end{equation}
143: Recently
144: the world data set on the reaction $pd\to\eta {^3}$He was analyzed
145: \cite{unserhe}. This study led to quite constrained values for the real and
146: imaginary part of the $\eta {^3}$He scattering length, namely
147: \begin{equation}
148: a=(\pm 4.3 \pm
149: 0.3 \, , \ 0.5\pm 0.5) \ \mbox{fm} \ ,
150: \label{scl}
151: \end{equation}
152: where the first number refers to the real part and the second number to the
153: imaginary part---in the analysis the effective range term of Eq. (\ref{era})
154: was neglected ($r=0$). Note, these numbers
155: where found from a fit to the world data set. However, this dataset is
156: inconsistent and if we use only the newest data in the fit the scattering
157: length is less constrained; see Ref. \cite{unserhe} for details. We come back
158: to this point below.
159: The
160: two signs given in front of the real part indicate that $\eta$ production
161: data can not fix the sign of the real part. A positive sign would point at a
162: virtual state (a singularity on the unphysical sheet), whereas a negative sign
163: would point at the existence of a bound state.
164: In Ref. \cite{mixing} it was
165: stressed that isospin--violating ratios of pion production cross sections taken
166: in the vicinity of the $\eta$ production threshold should be a good tool to
167: fix the sign of the real part.
168: It is
169: important to understand whether or not the TAPS measurement is sufficient to
170: decide on the sign of the
171: real part of the scattering length.
172:
173: The Flatt{\'e} form discussed above can be easily matched to the effective
174: range approximation of Eq. (\ref{era}) \cite{evidence}.
175: One thus finds that neglecting the effective range term in Eq. (\ref{era})
176: is equivalent to assume
177: that $g_{eff}$ is sufficiently large that in the region of interest $E$ can be
178: neglected in Eq. (\ref{BW}).
179: In Ref. \cite{evidence} it was argued that this should be a good
180: approximation if the structure of interest is dynamically generated and the
181: singularity is close to the threshold, as is the case here. In addition, the
182: role played by the effective range term in the $\eta {^3}$He final state
183: interaction is completely unclear and the data for $pd\to \eta {^3}$He could
184: be very well fitted using $r=0$.
185:
186:
187:
188:
189: \begin{figure}[t]
190: \vspace*{-0mm}\hspace*{-3mm}
191: \psfig{file=onlyback_banb_pap.eps,width=8.cm,height=6.cm}\vspace*{-2mm}
192: \caption{Comparison of the various fits to the data, as a function of the
193: reduced photon energy $W$ defined in Ref. \protect\cite{exp}. Shown are the results
194: for a dominance of the interference with the background
195: ($B$ large in Eq. (\protect\ref{full})). The
196: left panel corresponds to the calculation using the same binning as the
197: data, whereas the right panel shows the results with no binning. The
198: solid line corresponds to $a=(+4,1)$ fm, the dashed one to $a=(-4,1)$ fm and
199: the dotted one to $a=(0,3.5)$ fm.
200: The vertical line at $W=1486.4$ MeV indicates the position of the $\eta {^3}$He threshold.}
201: \label{backonly}
202: \end{figure}
203:
204: There is one additional comment necessary before
205: we can apply Eq. (\ref{era}) to the TAPS data: there is in principle some
206: interference with the background. Thus, what was identified as the resonance
207: signal might well have some contribution from an interference term, and the
208: full signal my be written as
209: \begin{equation}
210: N\left(2\mbox{Re}(Bf^{res})+\left|f^{res}\right|^2\right) \ ,
211: \label{full}
212: \end{equation}
213: where $B$ is some complex number parameterizing that part of the background
214: that is allowed to interfere with the resonance signal and $N$ is a measure of
215: the total strength of the signal. Therefore, we performed three different
216: fits: fit 1 included only the pure resonance signal ($B=0$; only $N$ as a free
217: parameter); fit 2 included only the interference term ($B\to \infty$; $N$ and
218: the phase of $B$ as a free parameter); and fit 3 considered the full structure
219: (thus here we have 3 free parameters: $N$, $|B|$ and the phase of $B$). As it
220: turned out, the $\chi ^2$ per degree of freedom for the two scenarios
221: (positive and negative real part of the scattering length) was almost the same
222: in all three cases and thus for illustration in Figs. \ref{resonly} and
223: \ref{backonly} we only show the results of the first two fits, where the left
224: panel corresponds to the results after binning in accordance with
225: that of
226: the experiment and the right panel corresponds to the unbinned results. To
227: keep the numbers of free parameters low we choose $a=(\pm 4,1)$ fm. In both
228: figures the dashed line corresponds to a negative real part (indicating
229: the existence of a bound state) and the solid line corresponds
230: to a positive real part (indicating a virtual state). The fit gave a
231: $\chi ^2$ per degree of freedom of 1 for the latter case, whereas it was worse
232: than 3 in the former. Thus the data prefers the solution that corresponds to
233: a virtual state, although the existence of a bound state can not be excluded,
234: given the quality of the data.
235: Note, already in Ref. \cite{shnm} the interpretation of the TAPS data as a
236: bound state was questioned.
237:
238: There is one important comment to be added: the data set for $pd\to \eta
239: {^3}$He shows some inconsistencies. As discussed in detail in Ref.
240: \cite{unserhe}, a fit to just the most recent data allows for a significantly
241: broader band of scattering lengths: then even the case of a vanishing real
242: part is not excluded (together with Im$(a)$=3.5 fm). To illustrate the impact
243: of this scattering length in Figs. \ref{resonly} and \ref{backonly} we also
244: show the corresponding results as the dotted curve. As can be
245: seen, this fit is almost equally good as that with the positive scattering
246: length ($\chi ^2$ per degree of freedom of about 2).
247:
248: Thus we conclude that the data on $\gamma {^3}\mbox{He}\to \pi p X$ recently
249: measured by the TAPS collaboration does not allow for a conclusion on the
250: existence of a bound system of $\eta$ and ${^3}$He. To improve the situation
251: the measurement should be redone with improved statistics to allow for
252: smaller energy bins. In addition, to permit
253: an unambiguous interpretation of $\gamma {^3}\mbox{He}\to \pi p
254: X$, more refined information on the $\eta {^3}$He scattering length is
255: needed. Fortunately, this will be available soon from measurements performed
256: at COSY \cite{COSY1,COSY2,COSY3}. The present paper clearly shows the usefulness of
257: a combined analysis of data from both electromagnetic and hadronic probes.
258:
259: {\bf Acknowledgement}
260:
261: We thank B. Krusche, J.A. Niskanen, V. Metag, M. Pfeiffer, and C. Wilkin for useful discussions and
262: J. Durso for numerous editorial remarks.
263:
264: \begin{references}
265: \bibitem{exp}
266: M.~Pfeiffer {\it et al.},
267: %``Photoproduction of eta-mesic He-3,''
268: Phys.\ Rev.\ Lett.\ {\bf 92} (2004) 252001
269: %\bibitem{bhalerao}
270: %R.~S.~Bhalerao and L.~C.~Liu,
271: %%``Off-Shell Model For Threshold Pionic Eta Production On A Nucleon And For Eta
272: %%N Scattering,''
273: %Phys.\ Rev.\ Lett.\ {\bf 54} (1985) 865.
274: \bibitem{Wycech}
275: S. Wycech, A.M. Green, and J.A. Niskanen, Phys. Rev.
276: C {\bf 52}, 544 (1995).
277: \bibitem{Belyaev1}
278: S.A. Rakityansky, S.A. Sofianos, W. Sandhas,
279: and V.B. Belyaev, Phys. Lett. B {\bf 359}, 33 (1995).
280: \bibitem{Belyaev2}
281: V.B. Belyaev, S.A. Rakityansky, S.A. Sofianos, M. Braun, and
282: W. Sandhas, Few. Body. Syst. Suppl. {\bf 8}, 309 (1995).
283: \bibitem{Rakityansky}
284: S.A. Rakityansky, S.A. Sofianos, M. Braun, V.B. Belyaev
285: and W. Sandhas, Phys. Rev. C {\bf 53}, R2043 (1996).
286: \bibitem{Fix1}
287: A. Fix and H. Arenh\"ovel, Phys. Rev. C {\bf 66}, 024002 (2002).
288: \bibitem{Wilkin1}
289: C. Wilkin, Phys. Rev. C. {\bf 47}, R938 (1993).
290: \bibitem{Haider1}
291: Q. Haider and L.C. Liu, Phys. Lett. B {\bf 172}, 257 (1986).
292: \bibitem{Liu}
293: L.C. Liu and Q. Haider, Phys. Rev. C {\bf 34}, 1845 (1986).
294: \bibitem{Chiang}
295: H.C. Chiang, E. Oset, and L.C. Liu, Phys. Rev. C {\bf 44}, 738 (1991).
296: \bibitem{Haider2}
297: Q. Haider and L.C. Liu, Phys. Rev. C {\bf 66}, 045208 (2002).
298: %\bibitem{Green}
299: % A.M. Green and S. Wycech, Phys. Rev. C {\bf 68}, 061601 (2003).
300: \bibitem{Calen}
301: H.~Cal{\'e}n {\it et al.},
302: %``Threshold structure of the quasifree p + n $\to$ d + eta reaction,''
303: Phys.\ Rev.\ Lett.\ {\bf 80} (1998) 2069.
304: \bibitem{Berger}
305: J. Berger {\it et al.}, Phys. Rev. Lett. {\bf 61}, 919 (1988).
306: \bibitem{Mayer}
307: B. Mayer {\it et al.}, Phys. Rev. C {\bf 53}, 2068 (1996).
308: \bibitem{Frascaria}
309: R.~Frascaria {\it et al.},
310: %``total d + d $\to$ alpha + eta cross-sections near threshold,''
311: Phys.\ Rev.\ C {\bf 50} (1994) 537.
312: \bibitem{flatte} S. Flatt{\'e}, Phys. Lett. {\bf 63B}, 224 (1976).
313: \bibitem{Bugg}
314: D.~V.~Bugg,
315: %``Reinterpreting several narrow 'resonances' as threshold cusps,''
316: arXiv:hep-ph/0406293.
317: \bibitem{wam}
318: see, e.g., M. Goldberger and K.M. Watson, {\it Collision Theory}
319: (Wiley, New York, 1964).
320: \bibitem{unserhe}
321: A.~Sibirtsev, J.~Haidenbauer, C.~Hanhart and J.~A.~Niskanen,
322: %``The eta He-3 scattering length revisited,''
323: arXiv:nucl-th/0310079; Eur. Phys. J. A., in print.
324: \bibitem{mixing}
325: V.~Baru, J.~Haidenbauer, C.~Hanhart and J.~A.~Niskanen,
326: %``Charge symmetry breaking as a probe for the real part of eta nucleus
327: %scattering lengths,''
328: Phys.\ Rev.\ C {\bf 68}, 035203 (2003)
329: \bibitem{evidence}
330: V.~Baru, J.~Haidenbauer, C.~Hanhart, Y.~Kalashnikova and A.~Kudryavtsev,
331: %``Evidence that the a0(980) and f0(980) are not elementary particles,''
332: Phys.\ Lett.\ B {\bf 586}, 53 (2004)
333: \bibitem{shnm}
334: A.~Sibirtsev, J.~Haidenbauer, J.~A.~Niskanen and U.~G.~Meissner,
335: %``Bounds on the bound \eta ^3He system,''
336: arXiv:nucl-th/0407073.
337: \bibitem{COSY1}
338: H. Machner et al., COSY Proposal no. 50.1, 2000.
339: \bibitem{COSY2}
340: A. Khoukaz et al., COSY Proposal no. 62.2, 2000.
341: \bibitem{COSY3}
342: A. Gillitzer et al., COSY Proposal no. 102.2, 2001.
343:
344:
345:
346:
347: \end{references}
348: \end{document}
349:
350:
351:
352: