arX

iv:1

107.

0879

v3 [

hep-

ex]

14 F

eb 2

014

π0π0 Production in Proton-Proton Collisions atTp=1.4 GeV

The WASA-at-COSY Collaboration

P. Adlarsona, C. Adolphb, W. Augustyniakc, M. Bashkanovd, T. Bednarskie, F.S. Bergmannf, M. Berłowskig, H. Bhatth,K.–T. Brinkmanni, M. Büscherj,k, H. Caléna, H. Clementd, D. Coderrej,k,l, E. Czerwinskie,1, E. Doroshkevichd, R. Engelsj,k,

W. Ervenm,k, W. Eyrichb, P. Fedoretsn, K. Föhlo, K. Franssona, F. Goldenbaumj,k, P. Goslawskif, K. Grigoryevj,k,p,C.–O. Gullströma, L. Heijkenskjölda, J. Heimlichb, V. Hejnyj,k, F. Hinterbergeri, M. Hodanae,j,k, B. Höistada, M. Jacewicza,M. Janusze, A. Janye, B.R. Janye, L. Jarczyke, T. Johanssona, B. Kamyse, G. Kemmerlingm,k, O. Khakimovad, A. Khoukazf,S. Kistryne, J. Klajae,j,k, H. Kleinesm,k, B. Kłosq, F. Krend, W. Krzemiene, P. Kulessar, A. Kupsca, K. Lalwanih, B. Lorentzj,k,A. Magierae, R. Maierj,k, B. Marianskic, P. Marciniewskia, M. Mikirtychiantsj,k,p, H.–P. Morschc, P. Moskale, B.K. Nandih,

S. Niedzwieckie, H. Ohmj,k, A. Passfeldf, C. Paulyj,k,2, E. Perez del Riod, Y. Petukhovs, N. Piskunovs, P. Plucinskia, P. Podkopałe,A. Povtoreykos, D. Prasuhnj,k, A. Prickingd, K. Pyszr, T. Rausmannf, C.F. Redmera, J. Ritmanj,k,l, Z. Rudye, S. Sawanth,

S. Schadmandj,k, A. Schmidtb, T. Sefzickj,k, V. Serdyukj,k,t, N. Shahh, M. Siemaszkoq, T. Skorodkod, M. Skurzoke, J. Smyrskie,V. Sopovn, R. Stassenj,k, J. Stepaniakg, G. Sterzenbachj,k, H. Stockhorstj,k, H. Ströherj,k, A. Szczurekr, A. Täschnerf, T. Tolbaj,k,∗,

A. Trzcinskic, R. Varmah, P. Vlasovi, G.J. Wagnerd, W. Weglorzq, U. Wiednerl, A. Winnemöllerf, M. Wolkea, A. Wronskae,P. Wüstnerm,k, P. Wurmj,k, X. Yuanu, L. Yurevj,k,t, J. Zabierowskiv, C. Zhengu, M.J. Zielinskie, W. Zipperq, J. Złomanczuka,

P. Zupranskic

aDivision of Nuclear Physics, Department of Physics and Astronomy, Uppsala University, Box 516, 75120 Uppsala, SwedenbPhysikalisches Institut, Friedrich–Alexander–Universität Erlangen–Nürnberg, Erwin–Rommel-Str. 1, 91058 Erlangen, Germany

cDepartment of Nuclear Reactions, The Andrzej Soltan Institute for Nuclear Studies, ul. Hoza 69, 00-681, Warsaw, PolanddPhysikalisches Institut, Eberhard–Karls–Universität Tübingen, Auf der Morgenstelle 14, 72076 Tübingen, Germany

eInstitute of Physics, Jagiellonian University, ul. Reymonta 4, 30-059 Kraków, PolandfInstitut für Kernphysik, Westfälische Wilhelms–Universität Münster, Wilhelm–Klemm–Str. 9, 48149 Münster, Germany

gHigh Energy Physics Department, The Andrzej Soltan Institute for Nuclear Studies, ul. Hoza 69, 00-681, Warsaw, PolandhDepartment of Physics, Indian Institute of Technology Bombay, Powai, Mumbai–400076, Maharashtra, India

iHelmholtz–Institut für Strahlen– und Kernphysik, Rheinische Friedrich–Wilhelms–Universität Bonn, Nußallee 14–16, 53115 Bonn, Germanyj Institut für Kernphysik, Forschungszentrum Jülich, 52425Jülich, Germany

kJülich Center for Hadron Physics, Forschungszentrum Jülich, 52425 Jülich, Germanyl Institut für Experimentalphysik I, Ruhr–Universität Bochum, Universitätsstr. 150, 44780 Bochum, Germany

mZentralinstitut für Elektronik, Forschungszentrum Jülich, 52425 Jülich, GermanynInstitute for Theoretical and Experimental Physics, StateScientific Center of the Russian Federation, Bolshaya Cheremushkinskaya 25, 117218 Moscow, Russia

oII. Physikalisches Institut, Justus–Liebig–UniversitätGießen, Heinrich–Buff–Ring 16, 35392 Giessen, GermanypHigh Energy Physics Division, Petersburg Nuclear Physics Institute, Orlova Rosha 2, 188300 Gatchina, Russia

qAugust Chełkowski Institute of Physics, University of Silesia, Uniwersytecka 4, 40-007, Katowice, PolandrThe Henryk Niewodniczanski Institute of Nuclear Physics,Polish Academy of Sciences, 152 Radzikowskiego St, 31-342 Kraków, Poland

sVeksler and Baldin Laboratory of High Energiy Physics, Joint Institute for Nuclear Physics, Joliot–Curie 6, 141980 Dubna, RussiatDzhelepov Laboratory of Nuclear Problems, Joint Institutefor Nuclear Physics, Joliot–Curie 6, 141980 Dubna, Russia

uInstitute of Modern Physics, Chinese Academy of Sciences, 509 Nanchang Rd., 730000 Lanzhou, ChinavDepartment of Cosmic Ray Physics, The Andrzej Soltan Institute for Nuclear Studies, ul. Uniwersytecka 5, 90-950 Lodz, Poland

Abstract

The reactionpp→ppπ0π0 has been investigated at a beam energy of 1.4 GeV using the WASA-at-COSY facility. The total crosssection is found to be (324± 21systematic± 58normalization) µb. In order to study the production mechanism, differential kinematicaldistributions have been evaluated. The differential distributions indicate that both initial state protons are excited into intermediate∆(1232) resonances, each decaying into a proton and a single pion, thereby producing the pion pair in the final state. No significantcontribution of the Roper resonanceN∗(1440) via its decay into a proton and two pions is found.

∗Corresponding authorEmail address:t.tolba@fz-juelich.de (T. Tolba )

1present address: INFN, Laboratori Nazionali di Frascati, Via E. Fermi 40,00044 Frascati (Roma), Italy

2present address: Fachbereich Physik, Bergische Universität Wuppertal,Gaußstr. 20, 42119 Wuppertal, Germany

1. Introduction

Investigations of the two-pion decay of mesons and baryonshave been extensively carried out in pion-inducedπN→ππN [1]and photon-inducedγN→ππN [2, 3, 4, 5] reactions. Doublepion production in nucleon-nucleon (NN) collisions is of par-ticular interest in view of studying the simultaneous excitation

Preprint submitted to Physics Letters B October 28, 2018

of the two baryons and their subsequent decays. Here, the sim-plest case is considered: the excitation of the two nucleonsintothe∆(1232) resonance.The reaction reported on here pro-vides the unique possibility to study this∆∆ process exclu-sively in very detail at its optimal energy of Tp = 1.4 GeV,which corresponds to

√s= 2.48 GeV≈ 2m∆.

Several theoretical models for double pion production havebeen suggested in the energy range from the production thresh-old up to several GeV [6, 7]. A full reaction model describ-ing the double pion production inNN collisions has been de-veloped recently by Alvarez-Rusoet al.(Valencia model)[8].More recent calculations by Cao, Zou and Xu include rela-tivistic correctionsnot taken into account by the Valenciamodel, however, neglect interference between different re-action amplitudes [9]. These models include both resonantand non-resonant terms ofππ-productionand predictthe two-pion production process to be dominated by resonance ex-citation: At energies near threshold it is dominated by theexcitation of one of the nucleons into the Roper resonanceN∗(1440)P11 via σ-exchange, followed by its s-wave decayN∗→N(ππ)s-wave

I=0 (where I indicates the isospin of theππ sys-tem). As the beam energy increases (i.e.Tp>1 GeV), thep-wave decayN∗→∆(1232)π→N(ππ) gives an increasinglygrowing contribution to the cross section. At higher ener-gies (Tp>1.1 GeV) the double∆(1232) excitation is expectedto become the dominant reaction mechanism.

First measurements of two-pion production in NN-collisions stem from low-statistics bubble chamber mea-surements [10, 11]. Morerecently, exclusive high-statisticsmeasurements have become available from near threshold(Tp=650 MeV) up toTp=1.3 GeV from the PROMICE/WASA[12, 13, 14], CELSIUS/WASA [15, 16, 17, 18, 19], COSY-TOF [20], WASA-at-COSY [21] andCOSY-ANKE [22] ex-periments. The analysis of the data obtained from theseexperiments indicate that,indeed as predicted, in case ofpp collisions (isovector channel) only twot-channel reactionmechanisms dominate: the excitation of the Roper resonanceN∗(1440) at energies close to threshold [13, 16], and the exci-tation of the∆∆ system at energiesTp > 1.1 GeV [18].In fact,the pp→ppπ0π0 reaction, which due to its isospin situationis the most suited reaction for studying these two resonanceexcitations [17], exhibits a distinctive dip in the slope ofthetotal cross section separating the regions of dominance forRoper and∆∆ processes.

Model predictions are found to be in good agreement withthe experimental results at energies close to threshold,if thebranching ratio for the decay N∗→∆π→Nππ is agjusted tothe experimental findings [5, 13, 14, 20, 18, 23]. At ener-giesTp≥1 GeV, the Roper resonance contribution to the to-tal cross section is strongly over-predicted in the Valenciacalculations due to the too large branching ratio assumedthere [8]. As shown in Ref. [18] the Valencia calculation isalso at variance with the differential data for the∆∆ process.However, if the ρ exchange, which in the Valencia model isthe dominating exchange process interfering destructivelywith the π exchange, is strongly reduced and if also rela-tivistic corrections are taken into account, then reasonable

agreement with the data is obtained. Thus all three changes(modified Valencia model) lead then to a satifactory descrip-tion of all data from threshold up to Tp=1.3 GeV [18].

The most astonishing point in this result is that ρ ex-change obviously plays only a minor role in the∆∆ exci-tation. Though this agrees with the theoretical findings ofCao, Zou and Xu [9], naively one would have expected thatthe∆∆ process is a shorter-range phenomenon and hence isparticularly sensitive to the ρ exchange, since it involves al-ready a considerable momentum-transfer. In order to studythis result in more detail, it is desirable to investigate the∆∆process at its optimal kinematic condition, which is reachedat√

s= 2m∆ corresponding toTp ≈1.4 GeV.In contrast to theexperimental situationat energiesTp≤1.3

GeV, there is little experimental information at higher energies.Only total cross sections are provided atTp=1.36 GeV [24] andTp=1.48 GeV [11].

Here, we report onexclusive and kinematically completehigh-statistics measurements of thepp→ppπ0π0 reaction atTp=1.4 GeV using the WASA at COSY facility [25]. The beamenergy corresponds to a center-of-mass energy of

√s=2.48

GeV, i.e. twice the∆ mass, thereby allowing a stringent testof thet-channel∆∆ mechanism.

2. Experimental Setup

The experimental data were collected using the Wide An-gle Shower Apparatus (WASA). WASA is an internal tar-get experiment at the COoler SYnchrotron (COSY) of theForschungszentrum Jülich, Germany. The detection systemprovides nearly full solid angle coverage for both charged andneutral particles. It allows multi-body final state hadronic inter-actions to be studied with high efficiency. The WASA facilityconsists of a central and a forward detector part and a cryogenicmicrosphere (pellet) target.

The pellet target generator is located above the central de-tector. It provides frozen pure hydrogen or deuteron pelletsof about 25µm diameter (as the targets), thereby minimizingbackground reactions from other materials.

The central detector is built around the interaction point andcovers polar scattering angles between 20◦–169◦. The inner-most detector, the mini drift chamber, is housed within the mag-netic field of a superconducting solenoid and is used in deter-mining the momenta of charged particles. The next layer, theplastic scintillator barrel provides fast signal for first level trig-ger and charged particle identification. As the outermost layer,1012 CsI(Na) crystals of the calorimeter enable the measure-ment of the energy deposited by charged particles as well as thereconstruction of electromagnetic showers. Due to the differentsize of the crystals, it was found that the energy and angle reso-lutions for the photons in the calorimeter are dependent on theirenergies and scattering angles, with average values of 15% and1.5◦ for energy and angular resolutions, respectively.

The forward detection system covers the polar angular rangeof 3◦–18◦. The multi-plane straw tube detector is implementedfor the precise reconstruction of charged particle track coordi-

2

nates. An arrangement of segmented plastic scintillator lay-ers, the forward range hodoscope, is used to reconstruct ki-netic energies of scattered particles by the∆E − E technique.A three-layered thin hodoscope provides fast charged particlediscrimination. The forward detector can provide a tag on me-son production via the missing mass of the reconstructed re-coil particles. The proton energy resolution shows an approx-imately constant value of bout 3% for protons up toTp=360MeV which is the maximum energy for which protons can bestopped by the forward range hodoscope layers. The resolutionworsensfor more energetic protons and reaches about 20%for 1 GeV protons. The angular resolution of the protons inthe forward detectoris 0.15◦. The trigger for the present ex-periment demanded at least one charged particle candidate toreach the first layer of the forward range hodoscope. Figure 1shows a schematic layout of te WASA detector at COSY, formore details about the WASA-at-COSY facility see Ref. [25].

Figure 1: The layout of the WASA detector at COSY.The SuperConductingSolenoid and the iron yoke for the return path of magnetic fluxis shownshaded. Plastic scintillators are situated in the Plastic Scintillator Barrel(PSB), Forward Window Counters (FWC), Forward Trigger Hodoscope(FTH), Forward Range Hodoscope (FRH), Forward Range IntermediateHodoscope (FRI), Forward Veto Hodoscope (FVH). Cesium Iodide scintil-lators are situated in the Scintillator Electromagnetic Calorimeter (SEC).Proportional wire drift tubes, straws, make up the Mini Drif t Chamber(MDC) and the Forward Proportional Chambers (FPC).

3. Data Analysis

Recoil protons from thepp→ppπ0π0 reaction withTp=1.4GeV are detected in the forward detector, while the two neutralpions are reconstructed in the central detector. The main crite-rion to select the event sample demands 1 or 2 charged tracks inthe forward detector and exactly 4 neutral tracks in the centraldetector. With this selection, the geometrical acceptanceof thepp→ppπ0π0 reaction is found to be 45%. Two event samplesare selected: the first includes events with only one proton de-tected in the forward detector while the other proton is scatteredoutside the forward detector. The second contains events whentwo protons were detected in the forward detector. The com-bination of both data samples gives a finite acceptance over allof the avaliable phase space, as shown in Fig. 2. Here, as anexample, two two-dimensional acceptance distributions ofpπ0

pairs (left plot) and ofpπ0π0- versuspπ0-invariant masses (right

plot) are presented. The Monte Carlo plots are based on equallypopulated phase space and show that nearly the full phase spaceis covered.

]2 [GeV/c0πpM1 1.2 1.4

]2 [G

eV/c

0 πpM

1

1.2

1.4

0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

0.18

0.2

0.22

0.24

0πp vs M0πpM

]2 [GeV/c0πpM1 1.2 1.4

]2 [G

eV/c

0 π0 πpM

1.2

1.4

1.6

0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

0.18

0.2

0.22

0.24

0πp vs M0π0πpM

Figure 2: Product of the geometrical acceptance times the detector efficiencyplotted as two-dimensional function of thepπ0-invariant mass pairs (left) andthe invariant mass of thepπ0π0 versuspπ0 (right).

[GeV]all

Edep FRH0 0.2 0.4 0.6

[GeV

]1

Ede

p F

RH

0

0.05

0.1

1

10

210

γγM0 0.1 0.2 0.3

a.u.

0

2000

4000

6000

8000

γγM

Figure 3: Left:∆E − E spectrum used for the identification of protons in theforward detector. Here, the deposited energy in the first layer of the forwardrange hodoscope (FRH) is plotted versus the total depositedenergy in all layersof the forward range hodoscope. Thesolid lines represent the region used toselect protons. Right:spectrum of theγ-pair invariant massMγγ. Thefor eacheventbest combination of 4 photons forming 2π0 (obtained by theχ2 method)is used. Full dots represent data points, whereas the solidline represents theMonte Carlo simulation.

The identification of protons in the forward detector is basedon the∆E − E method, where the difference between the en-ergy deposited in all layers of the detector (represented here bythe FRH) and the energy deposited in a specific detector layer(represented here by the first layer of the FRH) is plotted as afunction of the energy deposited in all layers of the detector,as shown in Fig. 3 (left plot). On one hand this technique isa powerful tool in distinguishing between the different particlespecies that are stopped in the detector. On the other hand, it isalso used in distinguishing between particles stopped in the de-tector elements and those,which punch through. The depictedselection criterion (the red lines) selects not only protons thatare stopped in the forward range hodoscope but also those thatpunch through. The selection helps to reject the contributionresulting from hadronic interactions in detector material. Thekinetic energy of the protons is reconstructed by translating thesummed deposited energy over all the forward detector layers,after they have beencorrectedfor the energy losses in the deadmaterial between the detector layers as well as the quenchingeffect in the plastic scintillator. For more details about the par-ticles identification and energy reconstruction see Ref. [26].

Neutral pions have been reconstructed from the photon pairsdetected in the central detector. The reconstruction procedureis based on the minimumχ2 method which is applied to select

3

the two-photon combinations with invariant masses closesttotheπ0 mass. Figure 3, right, shows the distribution of invariantmasses (Mγγ) for the best combination of the 4 photons formingtwo γγ pairs. The figure shows good agreement between thedata points (full dots) and the Monte Carlo simulations (solidline). The figure also shows that theMγγ distribution peaks attheπ0 mass witha resolution ofσ = 18 MeV.

Furthermore, a kinematic fit with six constraints, four fortotal energy-momentum conservation and two for each of thetwo γγ pair masses being equate to theπ0 mass, is applied inorder to suppress the contribution from background channelsand to recover the information of the unmeasured proton, scat-tered into the central detector or into inactive material. For con-sistency, the kinematic fit routine is always applied with oneunmeasured proton in the final state, with this assumption thenumber of constraints reduced to three. Hence, in the casewhere two protons are registered in the forward detector onlyone proton is selected and the other one is ignored. The pro-ton with the lower energy is found to have better resolution.Therefore, it is chosen as the measured value in the kinematicfit routine while the higher energy one is treated as the unmea-sured variable. In order to suppress events thatdo not satisfy thekinematic fit conditions a cut-off at the 10% confidence levelwas applied. This specific cut was chosen because it has thelargest product of combinatorial purityand reconstructed effi-ciency, where both data and Monte Carlo simulations are in theplateau region – for more details see Ref. [26].

The absolute normalization of the data has been achievedby normalizing to the measuredpp→ppη cross section [27].Two decay modes of theη meson,η→3π0 and η→2γ, werechosen because they have similar final state particles as thepp→ppπ0π0 reaction. These channels have an additional ad-vantage that they are the dominant neutral decay modes of theη meson [28].

The data are corrected for the detector efficiency and accep-tance by a Monte Carlo simulation using a toy model tunedto match the data. The toy modelaccounts for the previousfindings that the t-channel∆∆ mechanism is expected to bethe dominant effect and is constructed by generating a four-body final state phase space distribution of thepp→ppπ0π0 re-action, employing the GEANT phase space generator, based onthe FOWL program [29]. Then, the generated event weight ismodified to describe the 2π0 production mechanism accordingto the production of two∆(1232)P33 resonances in the interme-diate state, each decaying intopπ0. The partial wave amplitudethat describes the decay of∆(1232) intopπ-systemhas beentaken from Ref. [30]. This amplitude together with correctionterms for the measured proton and pion angular distributions inthe center-of-mass system, as well as for theMπ0π0 and Mpπ0

distributions are multiplied by the generated weights of eachevent. The Monte Carlo simulations are then compared withthe data, and this step is repeated until the data and the simula-tions are in good agreement. The tuned toy model is explainedin detail in Ref. [26].

4. Results

The total cross section of approximately 500 k events of thepp→ppπ0π0 reaction atTp = 1.4 GeV is determined to beσtot=(324± 21systematic± 58normalization) µb. The total cross sec-tion error is evaluated in terms of statistical and systematic un-certainties. The statistical error is found to be<1% and thusnegligible compared to the systimatic contribution. The sys-tematic error is constructed from two terms, systematic effectsand normalization. The systematic contribution is estimatedby observing the variation of the results with different analysisconstraints where the varied parameters are assumed to be inde-pendent of each other. The systematic term is calculated fromthe following main contributions: 1) applying different selec-tion regions to the flat part of the confidence level (probability)distribution of the kinematic fit, the contribution from this termis found to be 5%, 2) the contribution from the correction forthedetector acceptance generated by different Monte Carlo models(the tuned toy model, the model of Ref. [18] and the equallypopulated phase space model) is found to be 4%, and 3) con-straining the reconstructed particles to satisfy the geometricalboundaries of the central and the forward detectors, the contri-bution from this term is found to be 1%. The total error fromthe systematic term is the square root of the quadratic sum ofthe individual terms and found to be 6.5%. The normalizationterm is constructed from two main components: 1) contributionfrom the pp→ppη analysis, found to be 14%, and 2) the un-certainty of the cross section value in Ref. [27] which is foundto be 11%. The total error contribution from the normalizationterm is estimated to be 18%.

Figure 4 compares the cross section from this work (solid cir-cle) with the previous experimental data [10, 11, 12, 17, 18,24]and to the theoretical expectations calculated in Refs. [8,18].The data pointfrom this work is compatible with thepreviousresults [10, 11, 12, 17, 18, 24] and corroborates the stronglyrising trend of the cross section starting at∼ 1170 MeV. As hasbeen verified in Ref. [17], the trend of rising total cross sectionsfrom threshold up toTp∼1 GeV is due to the dominance of theRoper resonance. Above 1 GeV, it levels off and proceeds withonly a slight increase up toTp∼1170 MeV. The rise in the crosssection values at higher energiesTp>1170 MeV is associatedwith the∆∆ excitation, as demonstrated inRef. [18].

In order to study the mechanism of thepp→ppπ0π0 reaction,seven independent kinematic variables are necessary to coverthe available phase space of the reaction. Therefore, differentkinematical variables describing the system have been inves-tigated after the data have been corrected for the detector effi-ciency and acceptance using the tuned toy model. The correcteddata are compared to an uniformly populated phase space dis-tribution and the models according to Refs. [8] and [18]. Alltheoretical models are normalized to the same total cross sec-tion as the data. The differential distributions presented herehave been chosen because they are sensitive to contributionsfrom intermediate∆(1232) and/or theN∗(1440) resonances.

Figure 5, left shows that theπ0π0-invariant mass (Mπ0π0) dis-tribution is closerto the uniformly populated phase space dis-tribution than to the calculations of Ref. [8],which predict two

4

[GeV]pT0.6 0.7 0.8 0.9 1 1.1 1.2 1.3 1.4 1.5

b]µ [

Tot

alσ

−210

−110

1

10

210

310

WASA−at−COSY, this work

Figure 4: Total cross section for thepp→ppπ0π0 reaction as a function ofTp. The result of this work (solid circle), atT p = 1.4 GeV, is compared tothe data from PROMICE/WASA (open triangles) [12], CELSIUS/WASA (filledtriangles) [17] and at 1.36 GeV (square) [24], bubble chamber results (invertedtriangles) [10] and (star) [11], the theoretical calculations of Ref. [8] (dashedline) and of Ref. [18] (solid line), respectively.

]2 [GeV/c0π0πM0.2 0.4 0.6

]−2

[mb/

GeV

.c0 π0 π

/dM

σd

0

0.5

1

1.50π0πM

0π0πCMδcos

−1 −0.5 0 0.5 1

[mb]

0 π0 πCM

δ/d

cos

σd

0

0.2

0.4

0.6

0π0πCMδcos

Figure 5: Comparison of data (black dots) to the theoreticalexpectations cal-culated from Ref. [8] (red-dashed line) and Ref. [18] (blue-line), and withuniformly populated phase space (shaded area). Left: differential distributionof theπ0π0-invariant mass,Mπ0π0 . Right: differential distribution of two pionopening angle in the center-of-mass system,cosδCM

π0π0.

large enhancements at lower and higherMπ0π0 values. The en-hancement at higherMπ0π0 values is due to the dominance oftheρ exchange in the model calculations. In contrast, the dataare well described by the assumption oft-channel∆∆ excitationof Ref. [18] (solid line), where theρ-exchange contribution isstrongly reduced compared to the originalValenciacalculations[8]. The systematic enhancement at lowMπ0π0 values indicatesthe tendency of the two pions to be emittedin parallel with re-spect to each other. This behavior is seen as well in the two pionopening angle distributioncosδCM

π0π0 (right plot of Fig. 5), wherethe data is enhanced atcosδCM

π0π0 = 1 relative to the phase spacespectrum. Here, the data are well described by the modifiedcalculations of Ref. [18],whereas againa large deviation isobserved from the calculations of Ref. [8]. The strong peakingof the latter calculations at an opening angle of 180◦ correlateswith the enhancement at higher values ofMπ0π0 in left frame ofFig. 5.

The upper and lower left plots of Fig. 6 show indicationsfor the∆∆ excitation in the correlation of theMpπ0 pairs (up-per) and in the one-dimensional projection onto theMpπ0-axis(lower). Here, evidence for the∆(1232) resonance can be seenas a strong enhancement atMpπ0 ∼ M∆ = 1.232 GeV/c2. The

]2 [GeV/c0πpM1 1.2 1.4

]2 [G

eV/c

0 πpM

1

1.2

1.40πp vs M0πpM

]2 [GeV/c0πpM1 1.2 1.4

]2 [G

eV/c

0 π0 πpM

1.2

1.4

1.6 0πp vs M0π0πpM

]2 [GeV/c0πpM1 1.2 1.4

]−2

[mb/

GeV

.c0 πp

/dM

σd

0

5

100πpM

]2 [GeV/c0π0πpM1 1.2 1.4 1.6

]−2

[mb/

GeV

.c0 π0 πp

/dM

σd

0

1

2

3

40π0πpM

Figure 6: The upper frame shows two dimensional differential cross sectiondistributions ofpπ0- andpπ0π0-invariant masses divided by the available phasespace volume. This presentation enhances the sensitivity to resonance contri-butions in the production mechanism. The lower frame shows one-dimensionalprojections of the differential cross section onto theMpπ0 axis (left) andMpπ0π0

axis (right). See Fig. 5 for a description of the lines.

uniform phase space distribution of the lower plot shows thestrongest deviation with respect to the data due to the∆ excita-tion in the data points. In contrast, no significant evidenceforthe presence of the Roper resonanceN∗(1440) is observed inthe right-hand plots of Fig. 6. Here, one would expect an en-hancement aroundMpπ0π0 = 1.44 GeV/c2. The small deviationof the data from the solid line at about 1.5 GeV could pos-sibly signal some small contribution from theN∗(1520)D13

resonance. However, since we are here close to the edge ofthe covered phase space, a solid statement on this mattercan not be made.

The upper left plot of Fig. 7 shows thepp-invariant massspectrum,Mpp, which behaves complementary to theMπ0π0

spectrum in Fig. 5 and hencepeaks slightly to the rightwith respect to the uniformly populated phase space distribu-tion. The upper right plot shows the angular distribution ofthe protons in the center-of-mass frame,cosθCM

p . It exhibitsan anisotropic behavior, in agreement with the theoreticalcal-culations. The strong forward-backward peaking of thecosθCM

pspectrum is associated withπ − ρ exchange mediating theppinteraction. The angular distribution of thepπ0-system in thecenter-of-mass frame, cosθCM

pπ0 , (lower right) shows a forward-

backward symmetry. That is similar in shape to thecosθCMp

distribution as expected from the large p/π0 mass ratio. In theppπ0-invariant mass distribution,Mppπ0 , (lower left plot) thedata peak near the sum of the proton and∆masses as expectedfor ∆∆ production at threshold. Here, the modified model [18]and the data points are in good agreement, whereas the calcu-lations of Ref. [8] are shifted towards lowerMppπ0 values. Thedata deviate strongly from thephase space distribution, butare again in favor of the∆∆ excitation process – consistentwith the observations in the other invariant mass distribu-tions.

5

]2 [GeV/cppM1.8 2 2.2

]−2

[mb/

GeV

.cpp

/Mσd

0

0.5

1

1.5

2 ppM

pCMθcos

−1 −0.5 0 0.5 1

[mb]

pCM

θ/d

cos

σd

0

0.2

0.4

0.6pCMθcos

]2 [GeV/c0πppM2 2.2 2.4

]−2

[mb/

GeV

.c0 π

pp/d

Mσd

0

2

40πppM

0πpCMθcos

−1 −0.5 0 0.5 1

[mb]

0 πpCM

θ/d

cos

σd

0

0.5

1

0πpCMθcos

Figure 7: Same as Fig. 5 but for differential cross sections ofMpp (upper left),cosθCM

p (upper right),cosθCMpπ0

(lower right) andMppπ0 (lower left).

5. Conclusions and outlook

The first exclusive and kinematically complete measure-ments atTp=1.4 GeV reveal thet-channel∆∆ excitation tobe the by far dominating process, whereas the Roper ex-citation is found to play no longer any significant role inthe observables at such a high incident energy. The invari-nat mass distributions are characterized by the∆∆ process,which meets an optimal condition with the incident energycorresponding to

√s= 2.48 GeV≈ 2m∆. The modified Va-

lencia model [17] developed for the description of two-pionproduction at lower energies gives a good account for thenew measurements reported here. The most astonishingconclusion from this good greement between data and cal-culations is that counter intuitively and in contrast to theoriginal Valencia calculations [8] theρ exchange does obvi-ously not play a dominant role in thet-channel∆∆ process.

The investigation of the production of charged pions(π+π−) is the next step in the study of the double pion pro-duction in NN collisions with WASA-at-COSY. This chan-nel is of special interest in order to study the production ofρ0(770), which is expected to play an important role in theπ+π− channel at higher energies. Moreover, the extension tohigher proton energies will shed light on the role of heavierresonances.

6. Acknowledgments

This work was in part supported by: the Forschungszen-trum Jülich including the COSY-FFE program, the Euro-pean Community under the FP7-Infrastructure-2008-1, theGerman-BMBF, the German-Indian DAAD-DST exchangeprogram, VIQCD and the German Research Foundation(DFG), the Wallenberg Foundation, the Swedish ResearchCouncil, the Göran Gustafsson Foundation, the Polish Min-istry of Science and Higher Education and the Polish Na-tional Science Center (grant No. 0320/B/H03/2011/40) andFoundation for Polish Science - MPD program.

We also want to thank the technical and administrationstaff at the Forschungszentrum Jülich and at the participat-ing institutes.

This work is part of the PhD thesis of Tamer Tolba.

References

[1] S. Prakhov, et al.,Phys. Rev. C69, (2004) 045202.[2] Y. Assafiri, et al., Phys. Rev. Lett.90, (2003) 222001.[3] J. Ahrens, et al.,Phys. Lett. B624, (2005) 173.[4] U. Thoma, et al.,Phys. Lett. B659, (2008) 87.[5] A. V. Sarantsev, et al.,Phys. Lett. B659, (2008) 94.[6] J. Tejedor, E. Oset,Nucl. Phys. A571, (1994) 667.[7] E. Oset, M. J. Vicente-Vacas,Nucl. Phys. A446, (1985) 584.[8] L. Alvarez-Ruso, et al.,Nucl. Phys. A633, (1998) 519.[9] Xu. Cao, et al.,Phys. Rev. C81, (2010) 065201.

[10] F. Shimizu, et al.,Nucl. Phys. A386, (1982) 571.[11] A. M. Eisner, et al., Phys. Rev138, (1965) B670.[12] J. Johanson et al.,Nucl. Phys. A712, (2002) 75.[13] W. Brodowski, et al.,Phys. Rev. Lett.88, (2002) 192301.[14] J. Pätzold, et al.,Phys. Rev. C67, (2003) 052202.[15] T. Skorodko, et al.,Eur. Phys. J. A35, (2008) 317.[16] M. Bashkanov, et al.,Phys. Rev. Lett.102, (2009) 052301.[17] T. Skorodko, et al.,Phys. Lett. B679, (2009) 30.[18] T. Skorodko, et al.,Phys. Lett. B695, (2011) 115.[19] T. Skorodko, et al.,Eur. Phs. J. a47, (2011) 108.[20] S. Abd El-Bary, et al.,Eur. Phys. J. A37, (2008) 267.[21] P. Adlarson, et al.,Phys. Rev. Lett.106, (2011) 202302.[22] S. Dymov , et al.,Phys. Rev. Lett.102, (2009) 192301.[23] T. Skorodko , et al.,Eur. Phys. J.A35, (2008) 317.[24] I. Koch, PhD Thesis, Uppsala University, (2004).[25] H. H. Adam, et al., arXiv:nucl-ex/0411038, (2004).[26] T. Tolba, PhD Thesis, Ruhr-Universität Bochum, (2010).[27] E. Chiavassa, et al.,Phys. Lett. B322, (1994) 270.[28] K. Nakamura, et al., (Particle Data Group), J. Phys. G37, (2010)

075021.[29] F. James,CERN Program Library Long Writeup W505, (1977).[30] T. Risser, M. D. Shuster,Phys. Lett. B43, (1973) 68.

6