1: \documentclass[a4paper]{jpconf}
2: \usepackage{graphicx}
3: \begin{document}
4: \title{Evidence for $h_{c}$ Production from $\psi^{\prime}$ at CLEO}
5:
6: \author{Amiran Tomaradze}
7:
8: \address{Northwestern University, USA. Representing the CLEO Collaboration}
9:
10: \ead{amiran@bartok.phys.northwesetrn.edu}
11:
12: \begin{abstract}
13: Using the $\sim 6$ pb$^{-1}$ of $e^{+}e^{-}$ annihilation data
14: taken at $\psi^{\prime}(3686)$ with CLEO III and CLEO-c detectors
15: with estimated $\sim 3.0 \times 10^{6}$ $\psi^{\prime}$ events, we have
16: searched for the $h_c (1^1P_1)$ state of charmonium in the reaction
17: $\psi^{\prime}(3686) \to \pi^{0} h_{c} \to (\gamma \gamma)(\gamma \eta_{c})$.
18: The preliminary results are reported.
19: \end{abstract}
20:
21: \section{Introduction}
22:
23: Charmonium spectroscopy has played a crucial role in the understanding of the quark-gluon structure of hadrons and the underlying theory of Quantum Chronodynamics (QCD). This is primarily due to the fact that the charmonium system is expected to be far less sensitive to the problems associated with relativistic effects and the large value of the strong coupling constant, $\alpha_s$, than the light quark ($u$,$d$,$s$) systems. Formation cross-sections for charmonium states, their masses and widths are also favorable for precision measurements. The existing experimental data have defined the spin-independent one-gluon exchange part of the $q\bar{q}$ interaction quite well, however,
24: the spin dependence of the $q\bar{q}$ potential is not very well understood.
25: In particular, the $\vec{s_1}\cdot\vec{s_2}$ spin--spin, or
26: { hyperfine interaction is not well understood}, because there is
27: little experimental data to provide the required constraints for theory.
28: The primary experimental data required for understanding the $q\bar{q}$ hyperfine interaction
29: is hyperfine, or spin-singlet/spin-triplet splitting:
30: $\Delta M_{hf}(nL) \equiv \left< M(n^3L_J) \right> - M(n^1L_{J=L})$.
31:
32: For nearly 20 years, the only hyperfine splitting known was that for the $1S$ states of charmonium,
33: $\Delta M_{hf}(1S) = M(J/\psi) - M(\eta_c) = 116\pm2 \; \mathrm{MeV.}$
34: Very recently, Belle, CLEO and BaBar succeeded in identifying
35: $\eta_c'(2S)$,
36: with the rather surprising result that
37: $\Delta M_{hf}(2S) = M(\psi') - M(\eta_c') = 48\pm4 \; \mathrm{MeV.}$
38: Potential model and quenched lattice calculations predicted
39: a larger $\Delta M_{hf}(2S)$ [1].
40:
41: It is of great importance to find out how the hyperfine
42: interaction manifests itself in $P$ states, i.e., to find
43: $\Delta M_{hf}(1P)\equiv M(< ^{3}P_{J} >)-M(^{1}P_{1})$.
44: With scalar confinement, $\Delta M_{hf}(1P)=0$ is expected.
45: It is necessary to determine if this is true. The c.o.g. of $^3P$ states, $M(< ^{3}P_{J} >)$,
46: is well measured, $M(< ^{3}P_{J} >)$=3525.3$\pm$0.1 MeV.
47: What is needed is to identify $h_{c}$ and make a precision measurement
48: of its mass.
49:
50:
51: \section{Prior Experimental Searches for $h_c$}
52:
53: %Before we present the results of our investigations,
54: %we briefly describe the results of earlier attempts.
55: The Crystal Ball experiment at SLAC made a search for $h_c$ in 1982 [2].
56: The search was unsuccessful and they
57: reported 95\% confidence limits of
58: $B(\psi' \to \pi^0 h_c \: , \: h_c \to \gamma \eta_c) <$ 0.32\%
59: in the range $M_{h_c} = 3440 - 3543$ MeV. The next
60: search for $h_c$ was made by the Fermilab experiment E760 [3] in
61: the reaction
62: %$$p\bar{p} \to (h_c) \to \pi^0 J/\psi, \$
63: $p\bar{p} \to h_c \to \pi^0 J/\psi$.
64: % The mass region, 3522.6--3527.2 MeV was scanned with a total invested luminosity of $16 \, pb^{-1}$.
65: It was claimed that a statistically significant enhancement was observed and that the data indicated
66: $M(h_c) = 3526.2\pm 0.15\pm 0.2$ MeV.
67: However, such an enhancement has not been
68: confirmed by the successor Fermilab E835 experiment, with significantly
69: higher statistics [4,5].
70: The E835 experiment also searched for $h_c$ in the reaction
71: $p\bar{p} \to h_c \to \gamma \eta_{c} $. Preliminary evidence
72: at the $\sim$3$\sigma$ significance level has been recently reported with
73: $M(h_c) = 3525.8\pm 0.2\pm 0.2$ MeV [5]. No positive evidence
74: has been reported yet by Belle and BaBar Collaborations.
75:
76: It is fair to say that at present there is no convincing experimental evidence for $h_c$ observation.
77: % Needless to iterate, $h_c$ is the only undiscovered bound state of charmonium. Its discovery will be a crowning achievement of charmonium spectroscopy.
78:
79: \section{CLEO Searches and Results}
80:
81: The above considerations have motivated us to search for $h_c$ in the $\sim 6 pb^{-1}$ data taken at CLEO with estimated $\sim 3.0 \times 10^{6}$
82: $\psi^{\prime}$ events, in the reaction
83:
84: $$ \psi' \to \pi^0 h_c \; , \; h_c\to\gamma \eta_c.$$
85:
86: We search for this channel: (a) without using $\eta_{c}$ decays
87: (INCLUSIVE approach, see Section 3.1), and (b) using six dominant $\eta_{c}$ decay modes
88: (EXCLUSIVE approach, see Section 3.2).
89: In both methods we search for $h_{c}$ in the mass recoiling against
90: $\pi^{0}$ from decay $\psi^{\prime} \to \pi^{0} h_{c}$.
91: This method benefits from the excellent resolution of the CLEO calorimeter.
92:
93: \subsection{Inclusive Analyses}
94:
95: Two independent analyses have been performed, and results from the
96: two are consistent. I will describe
97: one of them in detail, and will later mention the differences
98: between the two analyses.
99: We use the following selection criteria: $N_{shower}\ge$3,
100: $N_{track}\ge$2.
101: The selection of the showers and charged particles
102: are done using the standard CLEO quality cuts.
103:
104: We reconstruct $\pi^{0}$'s by requiring that the two photon
105: invariant mass be in the range $M_{\gamma\gamma}$=135$\pm$15 MeV,
106: and that the two photons have been succesfully fitted to $\pi^0$.
107: We require that there be only
108: one $\pi^{0}$ in the event with a recoil mass in the expected
109: $h_{c}$ mass range of 3526$\pm$30 MeV.
110:
111: The $\psi^{\prime} \to \pi^{+}\pi^{-} J/\psi$ and
112: $\psi^{\prime} \to \pi^{0}\pi^{0} J/\psi$ events are
113: removed by cutting on the recoil mass of $\pi^{+}\pi^{-}$ and
114: $\pi^{0}\pi^{0}$, respectively.
115:
116: We define hard $\gamma$'s, the possible candidates from
117: $h_{c} \to \gamma \eta_{c}$ decays, by $E_\gamma > 400$ MeV.
118: We reject such $\gamma$'s which make a $\pi^{0}$
119: or $\eta$ with any other $\gamma$'s. We then require that the
120: energy of hard $\gamma$ should be in the range
121: $E_{\gamma}$=503$\pm$40 MeV.
122:
123: The background in data has been fitted in three ways: (a) ARGUS shape,
124: $y=x\times\sqrt{1-(x/a)^{2}}\times exp(b\times[1-(x/a)^{2}])$,
125: (b) second--order polynomial shape, (c) background shape from Monte Carlo.
126: The significance levels are obtained as
127: $\sigma\equiv\sqrt{-2\ln(L_0/L_{max})}$, where $L_{max}$ and $L_0$ are
128: the likelihoods of the fits with and without the $h_c$ resonance.
129:
130: The analysis on the Monte Carlo samples has been performed.
131: The event selection criteria applied to the Monte Carlo samples
132: were identical to those applied to the data. 10,000 signal Monte Carlo
133: events for the channel
134: $\psi^{\prime} \to \pi^{0} h_c \to (\gamma\gamma) (\gamma\eta_c)$
135: were simulated. The recoil mass distribution against $\pi^{0}$
136: in signal Monte Carlo, for input $\Gamma(h_{c})$=0 MeV is well
137: fitted with a double Gaussian with parameters $\sigma_{1}$=1.3 MeV,
138: $\sigma_{2}$=3.7 MeV, and the fraction of second Gaussian was 0.43.
139: These parameters, which represent the $\pi^{0}$ recoil mass
140: resolution at $h_{c}$, are used to fit the signal in the data.
141: The selection efficiency was about 16\%.
142: We also analyzed a sample of $\sim12\times 10^6$ generic $\psi'$
143: Monte Carlo events
144: (events containing all measured $\psi'$ decays except those via $h_c$)
145: in four separate samples, each with approximately the same size ($\sim3\times 10^6$) as the data. The signal Monte Carlo events were added in
146: to the generic Monte Carlo.
147: % assuming
148: %$B(\psi^{\prime} \to \pi^{0} h_c) \times B(h_c \to \gamma\eta_c)$
149: %=5.0$\times 10^{-4}$, as the theoretical predications for this
150: %product branching fractions is in the range $(1-10)\times 10^{-4}$.
151: The study of these Monte Carlo events yielded good agreement
152: between input and output values for both, $M(h_{c})$ and
153: $B(\psi^{\prime} \to \pi^{0} h_c) \times B(h_c \to \gamma\eta_c)$,
154: and showed that the analysis is sensitive to $h_{c}$ production.
155:
156: Figure 1 shows recoil mass distribution against $\pi^{0}$ in data.
157: The results of the fit are: $M(h_c)=3524.4\pm 0.7$ MeV,
158: N$(h_c$) = $156\pm48$, significance($h_c$) = 3.3 $\sigma$.
159:
160: An independent alternative analysis has been done.
161: The main difference is that in this analysis instead of constraining
162: the energy of the hard photon, the constraint is put
163: in terms of recoil against $\pi^{0}\gamma$ ($\eta_{c}$ mass).
164: The results are consistent with those shown above.
165: Thus our preliminary CLEO results from two inclusive analyses are:\\
166: $\bullet$ $M(h_{c})$=3524.8$\pm$0.7(stat)$\pm\sim$1(syst) MeV, \\
167: $\bullet$ $B(\psi^{\prime} \to \pi^{0} h_c) \times B(h_c \to \gamma\eta_c)$
168: =(2--6)$\times 10^{-4}$, \\
169: $\bullet$ The significance of $h_{c}$ detection $>$ 3 $\sigma$.
170:
171: Estimates of systematic errors in $M(h_{c})$ have been made by
172: studying the following: $\pi^{0}$ energy scale, background shapes,
173: Monte Carlo input/output differences, non-resonant background,
174: assumed $h_{c}$ width, binning effects, cut variations, and finally,
175: the difference in $M(h_{c})$ in the two inclusive analyses.
176:
177: \begin{figure}[h]
178: \includegraphics[width=18pc]{fig1.eps}\hspace{2pc}%
179: \begin{minipage}[b]{18pc}\caption{\label{label}
180: Distribution of the recoiling mass against $\pi^{0}$ in
181: data (inclusive analysis).
182: The curves are the results of the fit.
183: The shape of the signal is assumed as Double Gaussian, and the shape of the
184: background is assumed as ARGUS shape (see text).}
185: \end{minipage}
186: \end{figure}
187:
188:
189: \subsection{Exclusive Analysis}
190:
191: Six $\eta_{c}$ decay modes which have reasonably high PDG04 branching
192: ratios have been studied:
193: $K_{s}K^{\pm}\pi^{\mp}$,
194: $K^{+}K^{-}\pi^{0}$,
195: $K^{+}K^{-}\pi^{+}\pi^{-}$,
196: $2\pi^{+}2\pi^{-}$,
197: $\pi^{+}\pi^{-}\eta$ ($\eta \to \gamma\gamma$), and
198: $\pi^{+}\pi^{-}\eta$ ($\eta \to \pi^{+}\pi^{-}\pi^{0}$).
199:
200: Standard CLEO selections are used for showers, tracks, and particle
201: identification.
202: The total energy--momentum conservation of the event has been required,
203: and the invariant mass of the $\eta_c$ decay candidates
204: are required to be close to the nominal $\eta_c$ mass (within 50 MeV).
205: Figure 2(upper plot) shows the $\pi^0$ recoil mass distribution
206: for the sum of the six exclusive channels. The fit results are:\\
207: $\bullet$ $M(h_{c})$=3524.4$\pm$0.9(stat) MeV, \\
208: $\bullet$ N$(h_{c})$=15.0$\pm$4.2, \\
209: $\bullet$ The significance of $h_{c}$ detection $\sim$~5~$\sigma$.\\
210: Note that the significance is calculated using likelihood differences.
211: The background estimation by using $\eta_{c}$
212: sidebands(closed circles in Figure 2, lower plot), or by using
213: generic Monte Carlo events(open squares in Figure 2, lower plot),
214: yield consistent results.
215: No estimate of the systematic uncertainty in $M(h_{c})$ has been made so far.
216:
217: \begin{figure}[h]
218: \begin{center}
219: \rotatebox{270}{\includegraphics[width=2.5in]{fig2.eps}}
220: \caption{\label{label}
221: Distribution of the recoiling mass against $\pi^{0}$ (exclusive analysis).}
222: \end{center}
223: \end{figure}
224:
225: \section{Summary}
226:
227: We have analyzed $\sim 3.0 \times 10^{6}$ $\psi^{\prime}$ from
228: CLEO III and CLEO-c
229: to search for $h_{c}(^1P_{1})$ production in the reaction
230: $\psi^{\prime} \to \pi^{0} h_{c}$, $h_{c} \to \gamma \eta_{c}$
231: by two methods 1. INCLUSIVE -- which does not use $\eta_{c}$ decay modes,
232: 2. EXCLUSIVE -- which uses six hadronic decay modes of $\eta_{c}$.\\
233: In the recoil mass spectrum of $\pi^0$, we see an
234: enhancement in both analyses.\\
235: $\bullet$ In the inclusive analysis we obtain
236:
237: $M(h_{c})$=3524.8$\pm$0.7(stat)$\pm\sim$1(syst) MeV,
238:
239: $B(\psi^{\prime} \to \pi^{0} h_c) \times B(h_c \to \gamma\eta_c)$
240: =(2--6)$\times 10^{-4}$,
241:
242: significance of $h_{c}$ detection $>$3 $\sigma$.
243:
244: Thus, $\Delta M_{hf} \equiv \left< M(\chi_J) \right> - M(^1P_1)$ = 0.5$\pm$0.7(stat)$\pm\sim$1(syst) MeV.\\
245: $\bullet$ In the exclusive analysis we obtain
246:
247: $M(h_{c})$=3524.4$\pm$0.9(stat) MeV,
248:
249: significance of $h_{c}$ detection $\sim$ 5 $\sigma$. \\
250: $\bullet$ The inclusive and exclusive results for $M(h_c)$ are in excellent agreement.
251:
252:
253: \section*{Acknowledgments}
254: I would like to thank Kam Seth, Sean Dobbs, Zaza Metreveli (Northwestern
255: University), Jon Rosner (University of Chicago), Hajime Muramatsu
256: (Syracuse University), Datao Gong, Yuichi Kubota (University of Minnesota),
257: for their contributions to this analysis. I thank David Asner,
258: Gocha Tatishvili, Hanna Mahlke-Kruger, Helmut Vogel, Roy Briere, Steven Dytman,
259: Todd Pedlar, for their helpful comments.
260: Special thanks goes to Rich Galik for his useful comments and
261: continuous help during preparation of these results.
262:
263: This work was supported by U.S. Department of Energy.
264:
265: \section*{References}
266:
267: \medskip
268: \begin{thebibliography}{9}
269: \item Z. Metreveli, Presented at Heavy Quarks and Leptons Workshop 2004, San Juan, Puerto Rico, 1-5 Jun 2004. hep-ex/0408001.
270: \item F. C. Porter, et. al., $17^{\mathrm{th}}$ Recontre de Moriond Workshop on New Flavors, Les Arcs, France, (1982) p. 27. E. D. Bloom and C. W. Peck, Ann. Rev. Nucl. Part. Sci. \textbf{33} (1983) 143.
271: \item E760 Collaboration, T. A. Armstrong, et. al., Phys. Rev. Lett.
272: \textbf{69} (1992) 2337.
273: \item D. Joffe., Ph. D. dissertation, Northwestern University, 2004.
274: \item C. Patrignani, Presented at BEACH2004, Chicago, June 27--July 3, 2004,
275: and at QWGIII workshop, Beijing, Oct. 12--15, 2004.
276: \end{thebibliography}
277: \smallskip
278:
279: \end{document}
280:
281: