K + Λ (1405) photoproduction at the BGO-OD experiment
Georg Scheluchin, Stefan Alef, Patrick Bauer, Reinhard Beck, Alessandro Braghieri, Philip Cole, Rachele Di Salvo, Daniel Elsner, Alessia Fantini, Oliver Freyermuth, Francesco Ghio, Anatoly Gridnev, Daniel Hammann, Jürgen Hannappel, Thomas Jude, Katrin Kohl, Nikolay Kozlenko, Alexander Lapik, Paolo Levi Sandri, Valery Lisin, Giuseppe Mandaglio, Roberto Messi, Dario Moricciani, Vladimir Nedorezov, Dmitry Novinsky, Paolo Pedroni, Andrei Polonski, Björn-Eric Reitz, Mariia Romaniuk, Hartmut Schmieden, Victorin Sumachev, Viacheslav Tarakanov, Christian Tillmanns
KK + Λ (1405) photoproduction at the BGO-OD experiment Georg
Scheluchin , ∗ , Stefan
Alef , Patrick
Bauer , Reinhard
Beck , Alessandro
Braghieri , Philip
Cole , Rachele
Di Salvo , Daniel
Elsner , Alessia
Fantini , , Oliver
Freyermuth , Francesco
Ghio , , Anatoly
Gridnev , Daniel
Hammann , Jürgen
Hannappel , Thomas
Jude , Katrin
Kohl , Nikolay
Kozlenko , Alexander
Lapik , Paolo
Levi Sandri , Valery
Lisin , Giuseppe
Mandaglio , , Roberto
Messi , , Dario
Moricciani , Vladimir
Nedorezov , Dmitry
Novinsky , Paolo
Pedroni , Andrei
Polonski , Björn-Eric
Reitz , Mariia
Romaniuk , Hartmut
Schmieden , Victorin
Sumachev , Viacheslav
Tarakanov , and Christian
Tillmanns Rheinische Friedrich-Willhelms-Universität Bonn, Physikalisches Institut, Nußallee 12, 53115 Bonn, Germany Helmholtz-Institut fuer Strahlen- und Kernphysik, Universitaet Bonn, Nussallee 1-16, D-53115 Bonn Germany Lamar University, Department of Physics, Beaumont, Texas, 77710, USA INFN Roma “Tor Vergata", Rome, Italy Università di Roma “Tor Vergata", Via della Ricerca Scientifica 1, 00133 Rome, Italy INFN sezione di Roma La Sapienza, P.le Aldo Moro 2, 00185 Rome, Italy Istituto Superiore di Sanita, Viale Regina Elena 299, 00161 Rome, Italy Petersburg Nuclear Physics Institute, Gatchina, Leningrad District, 188300, Russia Russian Academy of Sciences Institute for Nuclear Research, prospekt 60-letiya Oktyabrya 7a, Moscow 117312, Russia INFN - Laboratori Nazionali di Frascati, Via E. Fermi 40, 00044 Frascati, Italy INFN sezione Catania, 95129 Catania, Italy Universita degli Studi di Messina, Via Consolato del Mare 41, 98121 Messina, Italy INFN sezione di Pavia, Via Agostino Bassi, 6 - 27100 Pavia, Italy
Abstract.
Since the discovery of the Λ (1405), it remains poorly described by conventional constituent quarkmodels, and it is a candidate for having an "exotic" meson-baryon or "penta-quark" structure, similar to statesrecently reported in the hidden charm sector.The Λ (1405) can be produced in the reaction γ p → K + Λ (1405). The pure I = Σ π isprohibited for the mass-overlapping Σ (1385). Combining a large aperture forward magnetic spectrometer anda central BGO crystal calorimeter, the BGO-OD experiment is ideally suited to measure this decay with the K + in the forward direction. Preliminary results are presented.*Supported by DFG (PN 388979758, 405882627). For decades hadron spectroscopy has been used to in-vestigate the strong interaction. Experiments on protonsshowed that baryons consist of three valance quarks andsea-quarks. While three quark models show a good agree-ment between calculated and measured for most of thestates, the Λ (1405) deviates from the predictions. It islighter in mass than its non-strange counterpart, N (1535),even thought it has a strange quark in its composition.Furthermore the mass distribution, also called line shape,does not follow a Breit-Wigner distribution. The initial in-consistencies elevated Λ (1405) as an unconventional statecandidate since the discovery more than 50 years ago.Nowadays there are more theoretical models for Λ (1405) having a molecule like structure of N ¯ K . LatticeQCD calculations give more support for the molecule likestructure compared to a genuine three quark state [1][2].A study using a chiral unitarity model, where the reso-nance is generated dynamically from N ¯ K interactions with ∗ e-mail: [email protected] other channels constructed from the octets of baryons andmesons, shows that the line shape would depend on the Σπ Invariant Mass (GeV/c ) d σ / d m ( µ b / G e V ) Figure 1.
Line shape results from the CLAS experiment [4][5]for di ff erent decay modes. Colored lines show model predictionsof Nacher et al. [3]. Figure taken from [4]. a r X i v : . [ nu c l - e x ] A ug ime γ p K + Λ ( ) π Σ N K K Figure 2.
Possible photoproduction scheme of the Λ (1405). decay mode [3]. The results from the CLAS experimentare seen in figure 1 [4]. The measurements show that theline shape does di ff er for the decay modes. On close in-spection it is visible that the measurement and predictionsfor the charged decays di ff er from the predictions and asecond experiment could help to resolve this discrepancy. Λ (1405) can be produced via γ p → K + Λ (1405) asseen in figure 2. With the assumption of a molecular likestate, one can assume that the cross section for Λ (1405)is increased if the transferred momentum to the baryon islow. This means that the K + needs to take most of the pho-ton momentum, which corresponds to extreme forward an-gles in a fixed target experiment, while the Λ (1405) decaysalmost at rest isotropically. The BGO-OD experimentwith its central calorimeter and forward spectrometer isideally suited to measure such kinematics, which were notyet explored by other experiments for the Σ π decay. Inthe following chapters the experimental setup and particleidentification are described in more detail. The experimentis explained in section 2 and the reaction identification insection 3. Preliminary results are shown in section 4. B ismuth G ermanum O xide calorimeter with a O pen D ipole magnetspectrometer The BGO-OD experiment is located at the ELSA facil-ity in Bonn. ELSA is an electron accelerator with a quasicontinuous beam up to 3 . . . . / TAPS[8] or theBGO-OD experiment each revolution. This results in aquasi continuous electron beam for the duration of 5 s to12 s until the ring is empty and the process starts anew.In figure 4 the BGO-OD experiment is shown in moredetail. The electron beam hits a radiator inside the go-niometer tank, which creates a real photon beam viabremsstrahlung. The energy of the photons is determinedby measurement of the bremsstrahlung electron with theTagger detector. The photon beam interacts with the targetcell inside the BGO ball, which is filled with liquid hydro-gen or deuterium. The final state particles of the reactionare detected with the BGO calorimeter using bismuth ger-minate oxide crystals between polar angles 25 ◦ to 155 ◦ .Charged particles traveling in θ < ◦ are detected in theforward detector. The track trajectory before the open-dipole magnet is measured with the MOMO and SciFi2scintillating fibre detectors, and the drift chambers mea-sure the trajectory after the magnet. The measured trackcurvature is used to determine the momentum, while theToF walls at the end of the experiment measure the veloc- The E lectron S tretcher A ccelerator in Bonn(Germany) Figure 3.
Overview of the Electron Stretcher Accelerator [7] in Bonn, showing the main components. GO ball O pen D ipole magnet fl ux monitoring e - -beamgoniometerTagger magnetTagger ARGUSbeam dumpMOMOSciFi2 f o r w a r d s p e c t r o m e t e r c e n t r a l d e t e c t o r ToF driftchambers
Figure 4.
Overview of the BGO-OD experiment. ity via time of flight. Both measured quantities can be usedfor particle identification as seen in figure 5. momentum / MeV be t a π K p
Figure 5.
Particle velocity β against momentum p detected inthe forward spectrometer. Due to the mass di ff erence the chargedparticles π , K and p can be distinguished indicated by lines. Λ (1405) identification Λ (1405) was identified via the Σ π decay, which is pro-hibited for Σ (1385). The complete reactions is γ p → K + Λ (1405) → K + Σ π . The reaction can be identified bydetecting K + and π , while the Σ is identified via missingmass techniques. This can be achieved with the K + de-tected in the forward spectrometer and is described in sec-tion 3.1. An additional technique is used by detecting thefull final state increasing the K + polar angle acceptance inexchange for lower statistics, which is described in section3.2. K + at extreme forward angles The K + can be identified in the forward spectrometer us-ing the measured velocity and momentum of the chargedparticles, as seen in figure 5. A π can be identified bycombining two measured photons in the central calorime-ter and select events with invariant mass close to the π mass. This leaves the Σ particle to be identified throughmissing mass techniques. Figure 6 shows the missing massto K + and K + π systems. The red lines indicate the pos-sible missing hyperons in the reaction. The events with a Λ mass originate from the prominent Σ (1385) → Λ π de-cay. While it looks like there are some events with missing Σ , studies showed that most of these events are combina-torial background from Σ (1385) as seen in figure 7. Firstpreliminary results for the di ff erential cross section can beseen in section 4.2. Further analysis is needed to improvethe signal to background ratio. / MeV + missing mass to K / M e V π + m i ss i ng m a ss t o K ΣΛ Figure 6.
Missing mass to K + π against missing mass to K + ,while the K + was detected in the forward spectrometer. Measuring all final state particles in the reaction: γ p → Λ (1405) K + (1) → Σ π K + → Λ γπ K + (2) → p π − γγγ K + , (3)allows identification with K + polar angles outside the for-ward spectrometer acceptance. However the lack of par-ticle identification increases the number of combinatorialbackground. Therefore after creating all possible com-binatorics, a kinematics fit is used to improve the mass
400 1500 1600/ MeV + miss. mass to K020406080100120140160180200 c oun t s real data(1405) Λ + K (1520) Λ + K (1385) Σ + K π Λ + K π Σ + K - π + Σ + Ksimulation sum
Figure 7.
Missing mass to K + . This is a projection of the twodimensional histogram in figure 6. The colored lines show thecontributions of di ff erent simulated channels fitted to the data. resolution and reduce background via confidence level se-lection cuts. In figure 8 the angular distribution of thephotons is shown for signal and background simulationafter the kinematic fit. The combinatorial backgroundpasses the selection cuts predominantly with low photonangles. Events with low photon angles are removed fromthe analysis to improve the signal to noise ratio at the costof statistics. The γ Λ against γ Λ π invariant mass distribu-tion is plotted in figure 9. The signal and background canbe distinguished, as only the signal shows a correlation tothe Σ mass in the Λ γ invariant mass. A peak for Λ (1405)and Λ (1520) is visible as both decay to Σ π . To sub-tract the background, the mass outside the signal region,marked by the red lines is used to modulate the amplitudeof di ff erent simulated reactions, for example K + Σ (1385).In figure 10 the projection of the Λ γπ mass is shown,where the signal with Λ γ mass close to Σ is subtracted.With this fit the background distribution is modeled andcan be extrapolated to the region of the signal. In figure11, the Λ γ projection of the two dimensional fit is shown.The signal region was not included in the fit. Subtractingthe fitted background distribution from the data yields theinvariant mass distribution for pure Σ π reactions, whichis plotted in figure 12. The contribution of non Λ (1405)events is estimated by the green line. At this point theline shape and di ff erential cross section can be extractedby subtraction of the simulated contributions. The prelim-inary results are shown in section 4. / MeV γ direction of θ p r obab ili t y SignalBackground
Figure 8.
Angular distribution of photon compared between sig-nal in red and background in black. inv. mass / MeV πγΛ i n v . m a ss / M e V γ Λ Figure 9.
Invariant mass of γ Λ against γ Λ π in the reaction γ p → Σ π K + . inv. mass / MeV Λγ π c oun t s real data(1385) Σ + K π Λ + K - π + Σ + Ksimulation sum
Figure 10. Λ γπ mass projection of figure 9, without eventswhere Λ γ mass is close to Σ mass. The colored lines show ex-ample reactions contributing to the background shape, while thered dotted line shows the total fit including also reactions notshown in figure. Λ (1405) → Σ π line shape and differential cross section Using the analysis steps from the previous section the lineshape and di ff erential cross section was determined. Fornow only the Σ π decay was analyzed, as the BGO-ODsetup is well suited to measure the decay particles. In sec- nv. mass / MeV Λγ c oun t s real data(1385) Σ + K π Λ + K - π + Σ + Ksimulation sum
Figure 11. Λ γ mass projection of figure 9. The colored linesshow example reactions contributing to the background shape,while the red dotted line shows the total fit including also reac-tions not shown in figure. )=-1.00..0.80) θ =1500..2300 MeV,cos( γ (E real data (1520) Λ + simulated K π Σ + simulated K Σ + simulated K*background sum π Σ Figure 12.
Invariant mass of Σ π after all analysis steps. Thecolored lines show the estimation for not Λ (1405) reactions. tion 4.1 the line shape results are shown, while section 4.2shows the di ff erential cross section. The preliminary results on the line shape are comparedto other experimental results in figure 13. The Λ (1520)signal was not subtracted for consistency checks. Theresults are statistically comparable to the CLAS experi-ment, which derives from the better acceptance for thisdecay mode compared to the CLAS setup. Comparing theANKE and BGO-OD results indicates a double peak struc-ture in the line shape, which agree to the two pole structureof Λ (1405)[10]. This seems not to be present in the CLASdata and the statistical error is too big to make a definitestatement. All experimental results agree statistically tothe predicted line shape in the figure. As mentioned in theintroduction the deviations to the predictions show up inthe charged decay modes, which will be analyzed in thefuture. The results on the cross section are plotted in figure14. In general the cross section agrees to the CLAS c o un t s / a . u . Σ π invariant mass / MeV ANKE 2007BGO-OD CLAS 2013 p r e li m i n a r y Figure 13.
Preliminary line shape of Λ (1405) for E γ = results. The forward spectrometer analysis extends theCLAS results to extreme forward angles, which corre-lates to minimum transfer momentum. While the resultsagree for higher energies, it suggest the photon energy bin E γ = .. The BGO-OD experiment is ideally suited to investigatethe formation of unconventional states. Since more than50 years Λ (1405) is a potential candidate for such astate. Models describe Λ (1405) as a N ¯ K molecule-likestructure, which is formed below the free N ¯ K produc-tion threshold. This is also used as an explanation ofthe sudden cut-o ff in the invariant mass distribution (lineshape) above 1426 MeV as this marks the free produc-tion threshold. Thus the line shape is of great interest forthe study of Λ (1405). Preliminary results for the decaymode Λ (1405) → Σ π line shape and di ff erential crosssection agree within statistics to the CLAS experiment,while the di ff erential cross section could be extended toextreme forward angles thanks to the forwards spectrome-ter of the BGO-OD experiment. The charged decay modesof Λ (1405) will be investigated in the future. Acknowledgements
I thank the ELSA group for operating and maintainingof the electron accelerator. For the strong support on themaintenance and improvement of our experimental setup Ithank the technical sta ff of the contributing institutions.This work was supported by the DeutscheForschungsgemeinschaft project numbers 388979758and 405882627. Our Russian collaborators thank the CMS θ cos( b / s r c µ / Ω / d σ d =1550..1750 MeV γ E =1950..2150 MeV γ E − =1750..1950 MeV γ E =2150..2350 MeV γ E − CLASBGO-OD full top.BGO-OD f.spec p r e li m i n a r y Figure 14.
Preliminary di ff erential cross section against angle compared to the CLAS experiment[5] in blue. The black points markthe full topology reconstruction while magenta shows the K + in the forward spectrometer analysis. Russian Scientific Foundation (grant RSF number19-42-04132) for financial support.
References [1] J.M.M. Hall et al., Physical Review Letters ,doi:10.1103 / PhysRevLett.114.132002 (2015)[2] R. Molina and M. Döring, Physical Review Letters D ,056010 (2016)[3] J.C.Nacher et al., Physical Review Letters B ,55(1999)[4] K.Moriya et al.(CLAS Collaboration), Physical Re-view Letters C ,035206 (2013) [5] K.Moriya et al.(CLAS Collaboration), Physical Re-view Letters C ,045201 (2013)[6] W. Hillert, Europ. Phys. Jour. A , 139 (2006)[7] On basis of website. url: http: // / index_en.html[8] M.Nanova et al. (CB-ELSA Collaboration), Phys.Rev. C, (2016)[9] I.Zychor et al., Physical Review Letters B , 167-171 (2008)[10] J.A. Oller and U.-G. Meißner, Phys. Lett. B500