1: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
2: %% Proceeding of the talk given at SPIN02 %%
3: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
4: %% This comes from source file `template-6s.tex', %%
5: %% generated with the docstrip utility. %%
6: %% %%
7: %% The original source files were: %%
8: %% %%
9: %% template.raw (with options: `6s') %%
10: %% %%
11: %% Template for the LaTeX class aipproc. %%
12: %% %%
13: %% (C) 1998,2000,2001 American Institute of Physics %%
14: %% and Frank Mittelbach %%
15: %% All rights reserved %%
16: %% %%
17: %% %%
18: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
19: %% Please remove the next line of code if you
20: %% are satisfied that your installation is
21: %% complete and working.
22: %%
23: %% It is only there to help you in detecting
24: %% potential problems.
25: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
26:
27: %\input{aipcheck}
28:
29: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30: %% SELECT THE LAYOUT
31: %%
32: %% The class supports further options.
33: %% See aipguide.pdf for details.
34: %%
35: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
36: \documentclass[
37: ,final % use final for the camera ready runs
38: %% ,draft % use draft while you are working on the paper
39: ,numberedheadings % uncomment this option for numbered sections
40: %% , % add further options here if necessary
41: ]
42: {aipproc}
43: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
44: %% STYLE REQUESTED BY AIP
45: \layoutstyle{6x9}
46: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
47: %% Useful new commends to save space %%
48: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
49: \newcommand{\be}{\begin{eqnarray}}
50: \newcommand{\ee}{\end{eqnarray}}
51: \newcommand{\nn}{\nonumber}
52: \newcommand{\pp}{\pi^+ \pi^-}
53: \newcommand{\cosine}{$\langle cos\theta \rangle$ \,}
54: \newcommand{\hermes}{{\it HERMES \,}}
55: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
56: %% FRONTMATTER
57: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
58: \title{Hard Exclusive Electroproduction of Two Pions off
59: Proton and Deuteron at HERMES}
60:
61: \author{Pasquale di Nezza$^1$, \underline{Riccardo Fabbri}$^2$\\
62: ({\it on behalf of the HERMES Collaboration})}{
63: address={$^1$INFN, Laboratori Nazionali di Frascati, Via Enrico Fermi 40,
64: 00044 Frascati, Italy\\
65: $^1$E\_mail: Pasquale.DiNezza@lnf.infn.it\\
66: $^2$Department of Physics, Ferrara University,
67: Via delle Scienze 12, 44100 Ferrara, Italy\\
68: $^2$E\_mail: rfabbri@fe.infn.it}
69: }
70:
71: \begin{document}
72: \begin{abstract}
73: Exclusive electroproduction of $\pi^+\pi^-$
74: pairs off hydrogen and deuterium targets
75: has been studied with the HERMES experiment.
76: The angular distribution of the $\pi^+$ in
77: the $\pi^+\pi^-$ rest system has been studied
78: in the invariant mass range
79: $0.3 < m_{\pi\pi} <1.5$ GeV.
80: %
81: Theoretical models derived in the framework of the
82: Generalized Parton Distributions show that
83: this angular distribution receives
84: only contributions from the
85: interference between the isoscalar channel
86: $I=0$ and the isovector channel $I=1$.
87: %
88: \end{abstract}
89:
90: \maketitle
91:
92: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
93: %% INTRODUCTION
94: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
95: \small
96:
97: \vspace{-1.5cm}
98: \section{Introduction}
99:
100: \vspace{-0.3cm}
101: The analysis of hard exclusive production
102: of $\pi^+\pi^-$ pairs off unpolarized targets
103: of hydrogen and deuterium at HERMES is presented.
104: Recent theoretical studies \cite{collins,polyakov00,polyakov01}
105: have shown that the exclusive process
106: $ e^+ p \longrightarrow e^+ p \; \pi^+\pi^- $
107: can be described in the framework of
108: Generalized Parton Distributions ({\it GPD}s)
109: \cite{radyushkin,ji98,ji97}.
110: %
111: The diagrams relevant for this reaction
112: at leading twist are shown in Fig. \ref{fig:feynman}.
113: %
114: \begin{figure}[b]
115: \includegraphics[height=3.5cm,width=3.8cm]{rho_g.eps}
116: \includegraphics[height=3.5cm,width=3.8cm]{rho_q.eps}
117: \includegraphics[height=3.5cm,width=3.8cm]{f_qq.eps}
118: \includegraphics[height=3.5cm,width=3.8cm]{f_qg.eps}
119: \caption{Leading twist diagrams for the hard exclusive
120: reaction $e^+ p \rightarrow e^+ p \; \pi^+ \pi^-$.
121: The gluon exchange ({\it a)} gives rise to pions
122: in the isovector state only, while
123: the quark exchange ({\it b,c,d}) gives rise to
124: pions both in the isoscalar and in the isovector state.}
125: \label{fig:feynman}
126: \end{figure}
127: %
128: The pion pairs may be produced through gluon or quark
129: exchange with the target, either from quark
130: (Fig. \ref{fig:feynman}a,b,c) or from gluon fragmentation
131: (Fig. \ref{fig:feynman}d).
132: In the range of the considered $\pi^+\pi^-$ invariant mass
133: ($m_{\pi\pi}$),
134: both resonant and non-resonant contributions are present.
135: The particular state describing the
136: $\pi^+\pi^-$ pair, with the angular momentum quantum number
137: {\it l} being odd or even, also defines unequivocally
138: the quantum numbers C-Parity ({\it C}) and Strong Isospin
139: ({\it I}) of this state.
140: %
141: At small values of the Bjorken variable $x_{Bj}$,
142: pions are produced mostly
143: in the isovector state, because the dominant
144: mechanism is two-gluon exchange with
145: positive C-parity (Fig. \ref{fig:feynman}a) \cite{polyakov99}.
146: At large $x_{Bj}$ the production of pion pairs
147: is dominated by $q\bar{q}$ exchange
148: (Fig. \ref{fig:feynman}b,c,d) \cite{diehl}, which leads
149: to a sizable admixture of pion pairs with isospin
150: zero.
151: %
152:
153: In the HERMES kinematics the $\pi^+\pi^-$
154: cross section is dominated by the isovector
155: channel, i.e. $\rho^0$ production.
156: The less copious isoscalar $\pi^+\pi^-$
157: production can be investigated by studying the
158: interference between the odd and even {\it l}
159: wave production.
160: An observable suitable to probe this
161: interference appears to be the {\it intensity density}
162: \cite{polyakov00,polyakov01},
163: defined as the $l^{th}$ Legendre Polynomial $P_l(cos\theta)$
164: moment
165: \be
166: \langle P_l(cos\theta)\rangle^{\pp} =
167: \frac{\int_{-1}^1 d cos\theta \, P_l(cos\theta)\,
168: \frac{d\sigma^{\pp}}{d cos\theta}}
169: {\int_{-1}^1 d cos\theta\,\frac{d \sigma^{\pp}}{d cos\theta}},
170: \ee
171: %
172: where $\theta$ is the scattering angle
173: of $\pi^+$ meson in the $\pi^+\pi^-$ rest frame
174: with respect to the direction of the recoiled
175: system \cite{diehl}.
176: %
177: Some of the above defined intensities
178: are non-zero only if the interference
179: between the isoscalar and the isovector
180: channels is present.
181: %
182: Experimentally they are obtained as average
183: values of $P_l(cos\theta)$.
184: In this paper we describe the measurement of
185: the intensity density for $l=1$ which is
186: calculated by the average $\langle cos\theta \rangle$.
187: %
188:
189: %======================
190: %== DATA ANALYSIS ==
191: %======================
192: %
193: %-------------------Data-analysis---------------%
194: %
195: \vspace{-0.5cm}
196: \section{Data analysis}
197: %
198:
199: \vspace{-0.3cm}
200: The data have been accumulated with the
201: HERMES forward spectrometer \cite{hermes} during the
202: running period 1996-2000 of HERA.
203: The 27.6 GeV positron beam was scattered off
204: hydrogen and deuterium targets, respectively.
205: Events have been selected with one
206: positron track and two oppositely charged hadron
207: tracks ($E_h > 1.0$ GeV) without additional neutral
208: clusters in the calorimeter.
209: The exclusivity of these events was ensured by
210: imposing a further cut on the inelasticity
211: $\Delta E = \frac{M^2_X - M^2_P}{2 M_P}$,
212: where $M_X$ is the invariant mass of the undetected
213: system and $M_p$ is the proton mass.
214: As explained above, to enhance the isoscalar production
215: and consequently the interference between the isovector
216: and isoscalar channel, it was also required
217: that $x_{Bj} > 0.1$.
218: Moreover, in order to enter the {\it hard} regime of the
219: process, constraints on $Q^2$ ($Q^2>1$ GeV$^2$) and
220: $W$ ($W>2$ GeV) were imposed, where $W$ is the invariant mass
221: of the virtual-photon nucleon system.
222:
223: The most important source of background to the exclusive
224: channel comes from fragmentation of partons in
225: semi-inclusive deep inelastic scattering ({\it DIS})
226: at low $\Delta E$.
227: Due to the instrumental resolution and smearing,
228: those events may contaminate the exclusive sample.
229: The shape of the {\it DIS} background is well
230: reproduced by a $\it Monte \,Carlo$ simulation ({\it MC}).
231: It is based on the {\it LEPTO} generator using the
232: {\it LUND} model and the detector response was simulated
233: by a {\it GEANT}-based $\it Monte \,Carlo$ code.
234: %As can be seen in Fig.\ref{fig:norm} (left panel)
235: {\it MC} data were normalized to experimental data at
236: $\Delta E > 2$ GeV and then subtracted.
237: In order to illustrate our procedure, in
238: Fig. \ref{fig:norm} both the unsubtracted (left panel)
239: and full subtracted (right panel) $\pi^+\pi^-$ data
240: are shown as a function of the inelasticity $\Delta E$.
241: Here only data within the $\pi^+\pi^-$
242: invariant mass window $0.6<m_{\pi\pi}<1.0$ GeV,
243: around the $\rho^0$ mass, are shown.
244: %
245: \begin{figure}[t]
246: \includegraphics[height=.3\textheight]{rho_norm_d.eps}
247: \includegraphics[height=.3\textheight]{rho_excl_d.eps}
248: \caption{Left panel: {\it DIS MC} (shaded) normalized
249: to HERMES data (crosses) at $\Delta E > 2$ GeV.
250: Right panel: the exclusive $\pi^+\pi^-$ channel
251: obtained by subtracting the {\it DIS}-MC
252: spectrum from the experimental one.
253: In both panels only the $\rho^0$
254: invariant mass window is considered.}
255: \label{fig:norm}
256: \end{figure}
257: %
258: Exclusive events were then selected for
259: $\Delta E < 0.625$ GeV to maximize the ratio
260: of the exclusive signal over the background
261: ($Sg/Bg$) and minimize its relative statistical error.
262: %
263: For these events the intensity density was calculated
264: in 10 bins of $m_{\pi \pi}$.
265: The intensity density of the background,
266: $\langle cos\theta \rangle_{Bg}$,
267: was evaluated using data for $\Delta E > 2$
268: GeV, assuming its $\Delta E$ independence.
269: This assumption was tested checking the
270: stability of $\langle cos\theta \rangle_{Bg}$
271: in different $\Delta E$ bins.
272: In every bin of $m_{\pi \pi}$, the value of
273: $\langle cos\theta \rangle_{Bg}$, weighted by
274: the ratio $Sg/Bg$ for the chosen $\Delta E$ cut,
275: was subtracted.
276: The background corrections range
277: between $10-70\%$ in the various bins.
278: %
279: %--------------------------Results--------------------------%
280: %
281:
282: \vspace{-0.7cm}
283: \section{Results}
284: %
285:
286: \vspace{-0.3cm}
287: In Fig. \ref{fig:results_mhh} the preliminary HERMES results
288: on the $m_{\pi\pi}$-dependence of the intensity density
289: \cosine are shown, for the proton in the left panel and for
290: the deuteron in the right one.
291: %
292: \begin{figure}[b]
293: \includegraphics[height=6cm, width=7.5cm]{cosvsmhh_h.eps}
294: \includegraphics[height=6cm, width=7.5cm]{cosvsmhh_d.eps}
295: \caption{$m_{\pi\pi}$-dependence of the intensity
296: density \cosine for the proton (left panel)
297: and the deuteron (right panel).
298: Shaded areas in both panels represent
299: the systematic uncertainty.}
300: \label{fig:results_mhh}
301: \end{figure}
302: %
303: The distributions show a clear angular asymmetry
304: whose size changes with $m_{\pi\pi}$.
305: At low $m_{\pi\pi}$ ($m_{\pi\pi}<0.6$ GeV)
306: the asymmetry may be due to an
307: interference between the lower tail of the $\rho^0$ meson
308: and the non-resonant $\pi^+\pi^-$ $S$-wave production
309: ($I=1$ and $I=0$ interference).
310: This interference is present over the entire
311: invariant mass region considered.
312: At large $m_{\pi\pi}$ ($m_{\pi\pi}>1.0$ GeV), additionally,
313: an interference between the upper tail of the
314: $\rho^0$ meson and the $f$-type mesons arises and is
315: superimposed.
316: In particular, the possible change of sign of
317: the asymmetry at $m_{\pi\pi}\approx 1.3$ GeV
318: may be understood as being caused by the interference
319: of the broad $\rho^0$ tail and the $f_2$ resonance
320: ($1.270$ GeV).
321: %
322: Note that no similar behavior is seen at the narrow
323: $f_0$ resonance ($0.980$ GeV), possibly due to the experimental
324: resolution.
325:
326: In Fig. \ref{fig:results_xbj} the $x_{Bj}$-dependence
327: of the intensity density \cosine is shown for the proton
328: within the above defined $\rho^0$ window.
329: %
330: \begin{figure}[t]
331: \includegraphics[height=5.5cm, width=8.cm]{cosvsxbj_h.eps}
332: \caption{$x_{Bj}$-dependence of the intensity
333: density \cosine in the $\rho^0$ invariant mass
334: window for the proton.
335: The shaded area represents
336: the systematic uncertainty.}
337:
338: \vspace{-2.0cm}
339: \label{fig:results_xbj}
340: \end{figure}
341: %
342: The size of the \cosine asymmetry increases with
343: $x_{Bj}$.
344: This experimental finding is in agreement with
345: theoretical expectations according to which
346: at increasing $x_{Bj}$ the $\pi^+\pi^-$ production
347: becomes increasingly dominated by $q\bar{q}$ exchange,
348: leading to a sizable admixture
349: of isoscalar and isovector pion pairs.
350: As explained above this leads to an enhancement of
351: the interference term.
352: %
353:
354: The systematic uncertainties have been evaluated
355: considering all hadrons as pions, using different
356: exclusive cuts and applying various procedures
357: for the normalization of the {\it MC} generated background
358: to the data.
359: The main contribution has been found to originate
360: from using different exclusive cuts.
361: %
362: \begin{figure}[b]
363: \includegraphics[height=5.0cm, width=7.5cm]{mhh_th.eps}
364: \includegraphics[height=5cm, width=7.5cm]{xbj_th.eps}
365: \caption{The experimental $m_{\pi\pi}$-dependence (left panel)
366: and $x_{Bj}$-dependence (right panel) of the
367: intensity density \cosine for the proton
368: are compared with theoretical predictions in
369: \cite{polyakov00,polyakov01}.
370: {\it Triangles} ({\it stars}) show the predictions
371: with the gluon {\it GPD} neglected
372: \cite{polyakov00} (included \cite{polyakov01}).
373: In the left panel, {\it Triangles} have been
374: slightly shifted for a better visibility.}
375: \label{fig:theory}
376: \end{figure}
377: %
378: Using the diffractive {\it DIPSI MC} generator \cite{dipsi},
379: effects due to the acceptance were found to be negligible.
380: Radiative corrections were shown in \cite{akushevich} to be
381: smaller than 1\% in the H1 and ZEUS kinematics.
382: At larger $x_{Bj}$, where the HERMES analysis is performed,
383: they are even smaller, and for that reason have been
384: neglected.
385:
386: The comparison with the theoretical predictions available
387: so far is promising.
388: In the left (right) panel of Fig. \ref{fig:theory}
389: the experimental dependence of the intensity density
390: \cosine on $m_{\pi\pi}$($x_{Bj}$) for the proton is compared
391: with predictions performed at leading twist
392: \cite{polyakov00,polyakov01}.
393: The shape of the theoretical distribution is nicely
394: reproducing the data, in particular at the $f_2$ meson
395: mass where the asymmetry changes sign.
396: %
397: The reasonable agreement of the leading twist predictions
398: with data may be understood as to arise from the cancellation
399: of higher twist effects in this kind of asymmetry \cite{garcon}.
400:
401:
402: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
403: %% CONCLUSIONS %%
404: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
405: \vspace{-0.6cm}
406: \section{Conclusions}
407:
408: \vspace{-0.3cm}
409: The $l=1$ intensity density in $\pi^+\pi^-$
410: hard exclusive electroproduction was measured
411: for the first time, for both proton and deuteron
412: at $x_{Bj} > 0.1$.
413: %
414: The quantity $\langle cos\theta \rangle$
415: is sensitive to the interference between the
416: isoscalar channel ($I=0$) and the isovector channel
417: ($I=1$).
418: The absolute value of the asymmetry measured in
419: the $l=1$ intensity density \cosine shows a minimum at
420: $m_{\pi\pi}=m_{\rho^0}$.
421: At smaller invariant mass the asymmetry may be
422: interpreted as originating from the interference
423: of the $\rho^0$ lower tail with the non-resonant
424: $\pi^+\pi^-$ production ($S$-wave), and at larger
425: invariant mass from the interference of the
426: $\rho^0$ upper tail ($P$-wave) with $\pi^+\pi^-$
427: production from the $f_2$ ($D$-wave).
428: %
429: The interference signal in the $\rho^0$
430: invariant mass window was shown to increase in size
431: with $x_{Bj}$. This behavior may be understood
432: as to arise from the increased dominance of $q\bar{q}$
433: exchange at larger values of $x_{Bj}$, which
434: leads to a sizable admixture of isoscalar
435: and isovector pion pairs.
436:
437: This is the first evidence of an $I=0$ admixture
438: in $\pi^+ \pi^-$ exclusive electroproduction.
439: This fact by itself is interesting in view of
440: the importance of the scalar meson sector.
441:
442: In the {\it GPD} framework, theoretical predictions
443: are available with and without
444: inclusion of the {\it GPD} describing the two-gluon
445: exchange.
446: Both predictions appear to be consistent with the data.
447: Therefore these results on exclusive $\pi^+ \pi^-$
448: production, together with results from
449: different exclusive channels (e.g. Deeply Virtual Compton
450: Scattering \cite{airapetian01}, exclusive $\pi^+$
451: production \cite{airapetian02}) may lead
452: to a better modeling of {\it GPD}s.
453:
454:
455: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
456: %% ACKNOWLEDGMENTS
457: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
458:
459: \vspace{-0.5cm}
460: \begin{theacknowledgments}
461:
462: \vspace{-0.3cm}
463: \small
464: We are deeply grateful to N. Bianchi, A. Borissov,
465: M. Diehl, W.-D. Nowak, B. Pire and M.V. Polyakov for useful
466: discussions and suggestions.
467: \end{theacknowledgments}
468: %%%%%%%%%%%%%%%%%%%%% BIBLIOGRAFY %%%%%%%%%%%%%%%%%%%%%%%
469:
470: \vspace{-0.5cm}
471: \begin{thebibliography}{20}
472:
473: \vspace{-0.3cm}
474: \bibitem{collins} J. C. Collins, L. Frankfurt and M. Strikman,
475: Phys.\ Rev.\ D {\bf 56}, (1997) 2982.
476: \bibitem{polyakov00}
477: B. Lehmann-Dronke et al.,
478: Phys.\ Lett.\ B {\bf 475}, (2000) 147.
479: \bibitem{polyakov01}
480: B. Lehmann-Dronke et al.,
481: Phys.\ Rev. D {\bf 63}, (2001) 114001.
482: \bibitem{radyushkin} A.V. Radyushkin,
483: hep-ph/0101225.
484: \bibitem{ji98} X. Ji, J.\ Phys. G {\bf 24}, (1998) 1181.
485: \bibitem{ji97} X. Ji, Phys.\ Rev. D {\bf 55}, (1997) 7114.
486: \bibitem{polyakov99}
487: M.V. Polyakov,
488: Nucl.\ Phys. B {\bf 555} (1999) 231.
489: \bibitem{diehl} M. Diehl, T. Gousset, B. Pire,
490: hep-ph/9909445, Talk given at 6th INT/JLAB
491: Workshop on Exclusive and Semiexclusive
492: Processes at High Momentm Trasfer,
493: NewPort News, VA, 19-23 May 1999.
494: \bibitem{hermes} K. Ackerstaff et al., (HERMES Collaboration),
495: N.I.M. A {\bf 417}, 230 (1998).
496: \bibitem{dipsi} M. Arneodo, L. Lamberti, M. Ryskin,
497: Comput.\ Phys.\ Commun. {\bf 100} (1997) 195.
498: \bibitem{akushevich} I. Akushevich and P. Kuzhir,
499: Phys.\ Lett. B {\bf 474} (2000) 411.
500: \bibitem{garcon} M. Garcon,
501: hep-ph/0210068.
502: \bibitem{airapetian01}
503: A. Airapetian et al. (HERMES Collaboration),
504: Phys.\ Rev.\ Lett. {\bf 87} (2001) 182001.
505: \bibitem{airapetian02}
506: A. Airapetian et al. (HERMES Collaboration),
507: Phys.\ Lett. B {\bf 535} (2002) 85.
508: \bibitem{pire} Ph. Haegler, B. Pire, L. Szymanowski and O.V. Teryaev,
509: Phys.\ Lett.B {\bf 535} (2002), 117.
510: \end{thebibliography}
511:
512: \end{document}
513: %%
514: %%%%%%%%%%%%%%%% End of file %%%%%%%%%%%%%%%%%%%%%%%%
515: