A precision measurement of the p(e,e′p)π0 reaction near threshold K. Chirapatpimol,1,2 M.H. Shabestari,1,3 R.A. Lindgren,1 L.C. Smith,1 J.R.M. Annand,4 D.W. Higinbotham,5 B. Moffit,5 V. Nelyubin,1 B.E. Norum,1 K. Allada,6 K. Aniol,7 K. Ardashev,1 D.S. Armstrong,8 R.A. Arndt‡,9 F. Benmokhtar,10 A.M. Bernstein,6 W. Bertozzi,6 W.J. Briscoe,9 L. Bimbot,11 A. Camsonne,5 J.-P. Chen,5 S. Choi,12 E. Chudakov,5 E. Cisbani,13 F. Cusanno,13 M.M. Dalton,5 C. Dutta,14 K. Egiyan‡,15 C. Ferna`ndez-Ram`ırez,5 R. Feuerbach,5 K.G. Fissum,16 S. Frullani,13 F. Garibaldi,13 O. Gayou,6 R. Gilman,17 S. Gilad,6 J. Goity,18 J. Gomez,5 B. Hahn,8 D. Hamilton,4 J.-O. Hansen,5 J. Huang,6 R. Igarashi,19 D. Ireland,4 C.W. de Jager,5,1 X. Jin,1 X. Jiang,17 T. Jinasundera,1 J. Kellie,4 C.E. Keppel,18 N. Kolb,19 J. LeRose,5 5 N. Liyanage,1 K. Livingston,4 D. McNulty,20,21 L. Mercado,20 R. Michaels,5 M. Mihoviloviˇc,22 S. Qian,6 1 X. Qian,23 S. Mailyan,15 V. Mamyan,24 S. Marrone,13 P. Monaghan,6 S. Nanda,5 C.F. Perdrisat,8 0 E. Piasetzky,25 D. Protopopescu,4 V. Punjabi,26 Y. Qiang,6 I.A. Rachek,27 A. Rakhman,28 G. Ron,29,30 2 G. Rosner,4 A. Saha‡,5 B. Sawatzky,31,5 A. Shahinyan,15 S. Sˇirca,32,22 N. Sparveris,6,31 R.R. Subedi,33 r p R. Suleiman,6 I. Strakovsky,9 V. Sulkosky,6,5 J. Moinelo,34 H. Voskanyan,15 K. Wang,1 Y. Wang,17 A J. Watson,33 D. Watts,35 B. Wojtsekhowski,5 R.L. Workman,9 H. Yao,31 X. Zhan,6 and Y. Zhang6 0 (The Hall A Collaboration) 1 1University of Virginia, Charlottesville, VA 22904 2Chiang Mai University, Chiang Mai, Thailand 50200 ] x 3Mississipi State University, Starkville, MS 39762 e 4University of Glasgow, Glasgow, G12 8QQ Scotland UK - 5Thomas Jefferson National Accelerator Facility, Newport News, VA 23606 cl 6Massachusetts Institute of Technology, Cambridge, MA 02139 u 7California State University, Los Angeles, Los Angeles, CA 90032 n 8College of William and Mary, Williamsburg, VA 23187 [ 9The George Washington University, Washington, D.C. 20052 10Duquesne University, Pittsburgh, PA 15282 3 11Institut de Physique Nucleaire, F-91406 Orsay Cedex, France v 7 12Seoul National University, Seoul 151-747, Korea 0 13Istituto Nazionale di Fisica Nucleare, Sezione Sanit`a, I-00161 Rome, Italy 6 14University of Kentucky, Lexington, KY 40506 5 15Yerevan Physics Institute, Yerevan, 0036 Armenia 0 16University of Lund, Box 118, SE-221 00 Lund, Sweden . 17Rutgers University, New Brunswick, NJ 08903 1 18Hampton University, Hampton, VA 23668 0 19University of Saskatchewan, Saskatoon, Canada S7N 0W0 5 1 20University of Massachusetts, Amherst, MA 01003 : 21Idaho State University, Pocatello, ID, 83209 v 22Joˇzef Stefan Institute, SI-1000 Ljubljana, Slovenia Xi 23Duke University, Durham, NC 27708 24Carnegie Mellon University, Pittsburgh, PA 15213 ar 25Tel Aviv University, Tel Aviv 69978, Israel 26Norfolk State University, Norfolk, VA 23504 27Budker Institute, 630090 Novosibirsk, Russia 28Syracuse University, Syracuse, NY 13244 29Lawrence Berkeley National Lab, Berkeley, CA 94720 30Racah Institute of Physics, Hebrew University of Jerusalem, Jerusalem, Israel 91904 31Temple University, Philadelphia, PA 19122 32University of Ljubljana, 1000 Ljubljana, Slovenia 33Kent State University, Kent, OH 44242 34Universidad Complutense de Madrid, Madrid, 98040 Spain 35University of Edinburgh, Edinburgh, EH8 9YL Scotland, UK (Dated: April14, 2015) New results are reported from a measurement of π0 electroproduction near threshold using the p(e,e′p)π0 reaction. The experiment was designed to determine precisely the energy dependence of s− and p−wave electromagnetic multipoles as a stringent test of the predictions of Chiral Per- turbation Theory (ChPT). The data were taken with an electron beam energy of 1192 MeV using a two-spectrometer setup in Hall A at Jefferson Lab. For the first time, complete coverage of the 2 φ∗ and θ∗ angles in the pπ0 center-of-mass was obtained for invariant energies above threshold π π from 0.5 MeV up to 15 MeV. The 4-momentum transfer Q2 coverage ranges from 0.05 to 0.155 (GeV/c)2 in finesteps. A simple phenomenological analysis of our datashows strong disagreement with p−wave predictions from ChPT for Q2 >0.07 (GeV/c)2, while the s−wave predictions are in reasonable agreement. PACSnumbers: 25.30.Rw,13.60.Le,12.39.Fe Neutral pion production from the proton vanishes in TheJLAB/HallAexperimentreportedhereprovidesthe the chiral limit of zero quark masses and pion momenta mostextensive(Q2,W)coverageofπ0 electroproduction p → 0. As a result, the reaction at threshold is partic- to date for testing theories of chiral dynamics substan- π ularly sensitive to non-perturbative mechanisms within tially above threshold. QCDwhichbreakchiralsymmetry. Itisalsoexperimen- Under the one-photon-exchange approximation, the tallythemostchallengingtostudy. Pionphoto-andelec- p(e,e′p)π0 cross section factorizes as follows: troproductionexperimentsarenowproducingdataofun- precedentedprecisiontotestChiralPerturbationTheory d3σ dσ =JΓ (1) (ChPT),thelow-energyeffectivefieldtheoryofQCD[1]. dQ2dWdΩ∗ v dΩ∗ π π ChPT treats the spontaneous and explicit chiral sym- where Γ is the virtual photon flux and the Jacobian metry breaking in terms of a perturbative expansion in v smallmomentaandquarkmasses,andmakespredictions J = ∂(Q2,W)/∂(Ee′,cosθe′) relates the differential vol- ume element of data binned in dQ2dW to the scattered for the s− and p−wavemultipoles for the γN →πN re- action in the near-threshold region. Within ChPT, the electron kinematics dEe′dcosθe′. The pπ0 center-of- mass (C.M.) differential cross section, dσ/dΩ∗, depends internal structure of the pion and nucleon is systemat- π onthetransverseǫandlongitudinalǫ polarizationofthe ically parameterized by Low Energy Constants (LEC), L virtual photon through the response functions R ,R while the long-range external πN dynamics are fixed by T L and their interference terms R and R : the underlying chiral symmetry. Once the LECs are de- LT TT termined by experiment near threshold, the convergence dσ p∗ of the chiral expansion can be tested by comparing pre- dΩ∗ = kπ∗(RT +ǫLRL+ǫRTT cos2φ∗π π γ dictions with data taken at energies above threshold. +p2ǫ (ǫ+1)R cosφ∗). (2) Recently π0 photoproduction cross-section and po- L LT π larized photon beam-asymmetry (Σ) data from the The response functions depend implicitly on Q2,W and MAMI A2/CB-TAPS experiment [2] were used to test θ∗, the π0 C.M. angle, while the angle φ∗ defines the π π two versions of ChPT. The relativistic ChPT calcula- rotation of the pπ0 plane with respect to the electron tion (RChPT/χMAID) [3–5] has been carried out to scattering plane (e,e′). Other definitions are ǫ = L (cOBa(lKcpuπ4Ml)a,0t1iwo)nh[i6(l]eHbBtuhCtehonPnoTlny)-roiefslaOotfi(vpOiπ3s(t)picπf4o)Hrfepoa−rvwyphaBovteaosrpyironondeluCecchttPirooTn- mJ(Qe=2n/tπ|akW∗o|f/2E)tǫhe,eEvΓe′ivmrtup=.alFpαinhEaoeltl′oyknγ∗|Wka∗n/|d2aπnp2dioEnpe∗πmreapsrQpee2tc(ht1eiv−Cel.yMǫ,)w.mahniolde- production (BKM96) [7]. Both the BKM01 and RChPT k∗ =(W2−m2)/2W istherealphotonequivalentenergy. γ p calculations, after fits of LECs to the data, were com- The p(e,e′p)π0 experiment was performed in Hall A patible with the experimental multipoles E0+,E1+,M1+ at Jefferson Lab using the Left High Resolution Spec- and M within an incident photon energy rangeof 7 to 1− trometer (LHRS) [14] to detect the scattered electron 25 MeV above threshold [3, 8]. and the BigBite Spectrometer [15] to detect the coinci- The pionelectroproductionreactionγ∗p→pπ0 allows dent proton. The CEBAF beam was energy-locked to amorestringenttestofChPT,sincethefour-momentum 1192 MeV and delivered to a 6-cm long, 2.54-cm wide transfer Q2 and invariant energy W can be varied in- cylindrical liquid hydrogen (LH ) target. Beam cur- 2 dependently. Chiral πN dynamics naturally involve the rents below 5 µA were used to limit the singles rates in massscaleQ2/m2,whiletheLECsfittedinphotoproduc- both spectrometers. Four angular settings for the LHRS π tion encapsulate higher order processes, involving pos- (θe′ =12.5◦,14.5◦,16.5◦ and20.5◦)coveredanearlycon- sibly N∆ or ρ,ω degrees of freedom. The Q2 depen- tinuous Q2 range of 0.05−0.155 (GeV/c)2 using a 4.4 dencenearthresholdmayrevealtheonsetoftheseshort- msr acceptance cut. The LHRS momentum acceptance ranged mechanisms. Until now, only limited kinematic wascenteredonthe pπ0 thresholdandcoveredthe range coveragefrom γ∗p→pπ0 threshold experiments is avail- −3%<δp/p<+5%. able [9–12]. Several older MAMI experiments showed Three angular settings of the BigBite were used a Q2 dependence of the total cross section near thresh- (θ =43.5◦,48◦ and 54◦) which provided full coverage p oldincompatiblewith HBChPT[11,12], althoughanew (Fig. 1) of the proton cone up to an invariant energy MAMI re-measurement has superceded those data [13]. above threshold of ∆W=15 MeV (at the largest Q2). 3 FIG. 2. Left: Coincidence timing between the LHRS and BigBite. Eventsbelonging to the true coincidence peak were selected using cuts indicated by the vertical lines, while ran- domcoincidenceswereselectedfromtheregionhighlightedin FIG. 1. Left: Overlap between three BigBite Spectrom- red. Right: Missing mass distribution at Q2=0.15 (GeV/c)2 eter proton laboratory angle settings (colored boxes) and pπ0 center-of-massbinsatQ2=0.135(GeV/c)2 and∆W=9.5 fortheinvariantmassrange0<∆W <10MeV.Background ∗ ∗ events from random coincidences (red) and target cell win- MeV.Radialandconcentriclinesseparatebinsofφ andθ , π π ∗ dows (blue) were subtracted from the raw distribution, leav- respectively. Only 5 out of 9 θ bins are shown. The blue line shows φ∗=180. Right: Radiπal and concentric lines sepa- ing theπ0 missing mass peak shown in gray. π ∗ rate bins of θ and ∆W, respectively, projected onto proton π lab momentum pp and θp. Bins to the(left,right) of the blue ∗ ◦ ∗ ◦ line correspond to (φ =180 , φ =0 ). The innermost circle π π represents ∆W = 0.5 MeV. inFig.2beforeandaftersubtractionofbothrandomco- incidences and target-window contributions. The latter background was estimated using cuts on ∆W below the The BigBite momentum acceptance covered the range π0 threshold. (0.25<p <0.5GeV/c),limitedbythetargetenergyloss Before binning the data, both incident and scattered p atlowmomentumandthethresholdsontheE−∆Escin- electron energies were corrected for ionization losses in tillatorcountersathighmomentum. Thelowmomentum the LH2 and target windows, using the calculated en- cutoff was achievedusing a thin (25 µm) Ti exit window trance and exit paths with respect to the measured tar- in the target scattering chamber and a helium bag for get interaction vertex. Proton transport energy losses transportuptoandbetweenthe BigBitedrift chambers. through the target, Ti window and BigBite were also Absolute normalization,energyandanglecalibrationsin corrected for each event. Acceptance corrections were both spectrometers were checked at each kinematic set- derivedfromaMonte-Carlosimulationofbothspectrom- tingusingelasticscatteringrunswithLH andthinsolid eters, using the Dubna-Mainz-Taipei (DMT) model [18] 2 targets. as a physics event generator. Special care was taken to Scintillator hodoscopes provided the primary triggers incorporate into the simulation radiative correction and for both spectrometers. A gas threshold Cˇerenkov de- straggling losses, a fine-mesh magnetic field map for the tector in the LHRS provided electron identification with BigBite,andthemeasuredenergyandangularresolution 99% efficiency. Signals from either E or ∆E scintillator andenergycalibrationdeterminedfromelasticscattering planesattherearofBigBitewereusedinthecoincidence runs,inordertoproperlyaccountfortheirsystematicef- trigger, while signal thresholds in both the hodoscopes fectsnearthreshold. Thedominantsourcesofsystematic andmulti-wire drift chamberswereset to suppressmini- uncertainty are target window background subtraction, mum ionizing tracks from pions. Final proton identifica- accidentalcoincidencecorrectionsandLHRScentralmo- tionwasmadeusingE−∆Ecutsonthehighlysegmented mentum calibration, which combined contribute to the scintillators. The path-lengthcorrectedcoincidencetime overallnormalization error of 20% near threshold at low distribution between the LHRS and BigBite is shown in Q2 decreasing to 7% for data above threshold at higher Fig. 2. A 10 ns wide cut centered on the peak was used Q2. to select true coincidences, while a 30 ns cut (excluding Events were accumulated using (12,30,18,9) bins for the peak) selected random coincidences for subtraction. (Q2,∆W,φ∗,θ∗) respectively,with a cut of±10MeVon π π Selection of the pπ0 final state required calculation of the missing-mass peak. The ∆W bin width was 1 MeV the missing-mass M after reconstruction of the detected andtheLHRSacceptanceextendedupto∆W=30MeV, particle’s 3-momenta: although with reduced C.M. coverage. The average Q2 binwidthwas0.01(GeV/c)2. Figure3showstypicaldif- M2 =(E+mp−Ee′ −Ep)2−(p~e−p~e′ −p~p)2 (3) ferentialcrosssectionsforeachφ∗ andθ∗ binobtainedat π π Theexperimentalmissing-massdistributionisalsoshown Q2=0.135 (GeV/c)2 and ∆W = 9.5 MeV. The curve la- 4 FIG. 3. Differential cross sections for p(e,e′p)π0 from this experiment at Q2 = 0.135 (GeV/c)2 and ∆W = 9.5 MeV FIG.4. Totalcrosssection forp(e,e′p)π0 asafunctionofQ2 binned in pπ0 center-of-mass angles φ∗ and and cos θ∗. See fordifferentbinsin∆W (invariantmassabovethreshold)for π π ((cid:3)) this experiment and (△) MAMI [13]. Units of ∆W are text for description of curves. Units are µb/sr. Errors are MeV. Errors are statistical only. statistical only. beledBKM96istheHBChPTpredictionfromBernardet al. [7], which uses LECs fitted to older photoproduction data from MAMI and electroproduction data at Q2=0.1 (GeV/c)2 from MAMI and NIKHEF. The other curve is an empirical fit to the data which we use to obtain the totalcrosssectionσ . Theempiricalfitusestheformin tot Eq.(2)andexpandstheresponsefunctionswithLegendre polynomials P(x), where x=cosθ∗, l π R +ǫ R =AT+L+AT+LP (x)+AT+LP (x)(4) T L L 0 1 1 2 2 R =ATT (1−x2) (5) TT 0 R =(ALT +ALT P (x))(1−x2)1/2. (6) LT 0 1 1 The total cross section σ is given by 4πp∗π AT+L. tot k∗ 0 γ The Q2 dependence of σtot is shown in Fig. 4 for dif- FIG.5. TheQ2dependenceofa0(left)andb(right)fromthe ferent ∆W bins starting 0.5 MeV above threshold. Two fits of Eq. 7 to the Legendre coefficient AT+L. The theory 0 ChPT calculations are shown (BKM96 [7], χMAID [4]), curves are calculated for the beam energy of our experiment along with the SAID08 solution [16] and phenomeno- (1192 MeV). For the curve labeled REFIT the BKM96 LEC logical models (DMT [18], MAID [17]) which have been bP has been lowered from 13 to 9.3 (GeV)−3 (see text and Fig. 6). Errors are statistical only. fitted to the world data on pion photo- and electropro- duction. Compared to the linear Q2 dependence of the HBChPT/BKM96 curve, our σ measurement shows a tot bending over at higher Q2 similar to the phenomenolog- The b coefficient parameterizes the contribution of ical models and the RChPT/χMAID theory. At lower p−wave multipoles arising from their intrinsic p∗π de- Q2,bothChPTcalculationsareconsistentwithourdata pendence near threshold, while a0 fits the combination over the entire ∆W range shown here. Note that two |E0+|2+ǫL|L0+|2 of s−wave multipoles extrapolated to of the RChPT LECs were fitted to a new MAMI re- threshold. The L0+ multipole dominates a0 over our Q2 measurement [13] (triangles in Fig. 4) of earlier Q2 > 0 range due to a large ǫL factor. The extraction of a0 and experiments,whiletheremainingLECswerefittedtothe b from fitting our data up to ∆W = 9.5 MeV is shown Q2 =0 A2/CB-TAPS data [2]. in Fig. 5, along with fits to the newest MAMI data [13] Near threshold, the s− and p−wave decomposition of up to ∆W = 3.5 MeV (the limit of their measurement) σ can be obtained by fitting the p∗ dependence of andpreviousresultsfromNIKHEF[9,10]. Thereisgood tot π AT+L according to agreement of both a0 and b with the chiral model pre- 0 dictions for our lowest Q2 points. For higher Q2, the AT+L =a +b|p∗|2. (7) HBChPT curve describes a better than RChPT. How- 0 0 π 0 5 by the REFIT b curve in Fig. 5. Moreover, a different adjustment of b is required to match our measurement P of AT+L. 2 The O(p4) RChPT calculation [4] predicts a nearly identical Q2 dependence for the b curve in Fig. 5 as the O(p3) HBChPT theory. At leading-order and next-to- leading order, P is controlled by a single O(p3) LEC 3 d , similarly to HBChPT [3]. However d is highly con- 9 9 strained by the Q2=0 photoproduction fits, and there is almost no room for adjustment. Other O(p4) LECs, which explicitly control Q2 dependent terms, either do not appreciably affect the p−wave multipoles, or effect the same Q2 response as b . P Despite the very different LEC composition of HBChPTandRChPT,itappearsneithercalculationcan be adjusted to agree with the Q2 trend of our p−wave data. Furthermore this discrepancy occurs well within the ∆W range where photoproduction p−waves are well described at O(p4) [8]. Our data therefore suggest that FIG. 6. The ∆W dependence of Legendre coefficients from higher powers of Q2 are needed in the ChPT formalism, fits to our data at Q2 = 0.135 (GeV/c)2. Note that the while the onset of disagreement (Q2 >0.07) implies a ∆W=9.5 MeV bin corresponds to the Legendre fit shown in t−channelenergyscaleabovethepionmass. Similardis- Fig. 3. Errors are statistical only. crepancies in ChPT calculations of nucleon form factors were removed by including vector mesons as dynamical degrees of freedom [21]. Our data could provide strong everthestrongdisagreementofourbcoefficientwithboth constraints to analogous extensions of pion electropro- chiral curves for Q2 > 0.07 GeV2 suggests at least one duction calculations. of the p−wave multipoles is described incorrectly in the calculations. The Q2 dependence of b from fitting the Insummary,aJLAB/HallAexperimenthasmeasured for the first time both the Q2 and extended ∆W depen- MAMI dataisqualitativelysimilar,althoughwithlarger denceofthe thresholdp(e,e′p)π0 reactionwithfullC.M. errors,due to the smaller ∆W range of their data. coverage and fine binning. Our phenomenological fit of Further insight can be obtained from the ∆W de- the data shows reasonable agreement with two leading pendence of the Legendre coefficients in the Q2 > ChPT theories for s−waves, while chiral predictions of 0.07 (GeV/c)2 region. This is shown in Fig. 6 at p−wavecontributions strongly divergefromour data for Q2=0.135(GeV/c)2. Whileallmodelsareingoodagree- Q2 > 0.07 (GeV/c)2. We use a Legendre decomposition ment with our data near threshold, the theory curves of our cross sections to show there is insufficient flexibil- for AT+L, AT+L and ATT show large variations above 0 2 0 ity in the low energy constants available for p−waves to ∆W=3MeV.Thesecoefficientsareparticularlysensitive account for the Q2 discrepancy. tothep−wavemultipolecombinationsP =2M +M 3 1+ 1− The collaboration wishes to acknowledge the Hall A andP =3E −M +M ,whileATT isalsosensitive 2 1+ 1+ 1− 0 to the combination ∆P2 = (P2−P2)/2. Our fit result technicalstaffandtheJeffersonLabAcceleratorDivision 23 2 3 fortheir support. Specialthanks goesto Ulf-G. Meiβner for ATT is close to zero over the ∆W range of our data, 0 which implies P2 ≈ P2 or M /M ≈ −2 (neglecting for his support in the early phases of this experiment. 2 3 1+ 1− Thisworkwassupportedbythe U.S.DepartmentofEn- the weak electric quadrupole E ). 1+ ergy and the U.S. National Science Foundation and in Only the DMT model predicts ATT ≈ 0 for ∆W < 0 particularDOE contractDE-FG02-97ER41025andNSF 15MeV,largelyduetotheircalculationoftheM mul- 1− MRI Award No. 021635. Jefferson Science Associates tipole[19],thevalueofwhichissubstantiallylargerthan operates the Thomas Jefferson National Accelerator Fa- predicted by ChPT. A similar result was obtained from cility under DOE contract DE-AC05-06OR23177. disperson relations [20]. In the BKM96 theory, which usesaO(p3)p−waveexpansion,itisnotpossibletosep- ‡ Deceased. arately adjust M and M , since only P is controlled 1+ 1− 3 by a single LEC b . By reducing b in the calculation P P from 13.0 to 9.3 (GeV)−3, we can improve agreement with both AT+L and ATT as shown in Fig. 6 by the 0 0 [1] V. Bernard, N. Kaiser and Ulf-G. Meiβner, Int.J. Mod. curve labeled REFIT. However this adjustment worsens Phys. E4, 193 (1995). the agreement with p−waves at lower Q2, as indicated [2] D.Hornidgeet al.,Phys.Rev.Lett.111, 062004 (2013). 6 [3] M. Hilt, S. Scherer and L. Tiator, Phys. Rev. C 87, [13] H. Merkel et al., arXiv:1109.5075 (2011). 045204 (2013). [14] J. Alcorn et al., Nucl. Instrum. Methods A 522 294 [4] M. Hilt, B.C. Lehnhart, S. Scherer and L. Tiator, Phys. (2004). Rev.C 88, 055207 (2013). [15] M.Mihovilovicetal.,Nucl.Instrum.MethodsA686,20 [5] MAID07, χMAID and DMT solutions obtained from: (2012). http://wwwkph.kph.uni-mainz.de/MAID/. [16] R.A. Arndtet al., Chinese Phys. C 33, 1063 (2009). [6] V.Bernard et al., Eur.Phys. J. A 11, 209 (2001). [17] D. Drechsel, S.S. Kamalov and L. Tiator, Eur. Phys. J. [7] V.Bernard et al., Nucl.Phys. A 607, 379 (1996). A 34, 69 (2007) [8] C. Ferna`ndez-Ram`ırez and A.M. Bernstein, Phys. Lett. [18] S.S. Kamalov and S.N. Yang, Phys. Rev. Lett. 83, 4494 B 724, 253 (2013). (1999). [9] T.P. Welch et al.,Phys. Rev.Lett 69, 2761 (1992). [19] S.S. Kamalov et al., Phys.Lett. B 522, 27 (2001). [10] H.B. van den Brink et al., Phys. Rev. Lett. 74, 3561 [20] S.S. Kamalov et al., Phys.Rev.C 66, 065206 (2002). (1995). [21] B.Kubis,Ulf-G.Meiβner,Nucl.Phys.A679,698(2001). [11] M.O. Distler et al.,Phys.Rev.Lett. 80, 2294 (1998). [12] H.Merkel et al., Phys.Rev.Lett. 88, 012301 (2002).