J-PARC E27 Experiment

Tomofumi NAGAE (Kyoto University)

2Photo in July of 2009

J-PARC Facility(KEK/JAEA）

South to North

Neutrino Beams　(to Kamioka)

JFY2009 Beams

Hadron Exp. Facility

Materials and Life Experimental

Facility30 GeV Synchrotron

JFY2008 Beams

3 GeV Synchrotron

CY2007 Beams

Linac

Hadron Experimental Hall

K1.8

KL

K1.1BR

High p (under constr.)SKS

K1.8BRK1.1

First beam in Feb. 2009World highest intensity Kaon beams !

30 GeV Primary Beam

Productiontarget (T1)

60m x 56m

World Facilities in the 21st Century

J-PARCJLab

DA NE

GSI/FAIRMainz

(e,e’K+)

(e,e’K+)

(K-,K+), (K-,π-)

(K-,π-)

HI, π, anti-p

For Strangeness Nuclear Physics

Hadron Program History2010: Oct.-Nov.

E19: Penta-quark search in π-p→K-X at 1.92 GeV/c

First physics data taking in Hadron Hall

2012: Feb. , after the Earthquake

E19: π-p→K-X at 2 GeV/c

2012: June

E27: d(π+,K+) for K-pp , a pilot run 5 kW / 270 kW

Hadron Program History2012: Dec. 10 kW

E10: (π-,K+)6ΛH

2013: March - May 20 kW

E15: 3He(K-,n) for K-pp

Radiation Accident

E13: Hypernuclear γ-ray spectroscopy; 4ΛHe, 19ΛF

Beams at K1.8: π± → K-

0

3.5

7

10.5

14

2010.10 2012.2 2012.6 2012.12 2013.1 2015.3 2015.5 2017.4

π x10^6 K- x10^6

3 3.3 5 10 15 20 20

50

100SX Power (kW)

E19E27

E10

E27 : Search for ”K-pp” in d(π

+,K

+)

Spokesperson : T. Nagae (Kyoto)

Y. Ichikawa et al., PTEP (2014) 101D03. Y. Ichikawa et al., PTEP (2015) 021D01.

J-PARC E27 Collaboration

Prog. Theor. Exp. Phys. 2013, 00000 (8 pages)DOI: 10.1093/ptep/0000000000

Inclusive spectrum of the d(π+,K+) reactionat 1.69 GeV/c

Yudai Ichikawa1,2, Tomofumi Nagae1, Hyoungchan Bhang3, Stefania Bufalino4,Hiroyuki Ekawa1,2, Petr Evtoukhovitch5, Alessandro Feliciello4, Hiroyuki Fujioka1,Shoichi Hasegawa2, Shuhei Hayakawa6, Ryotaro Honda7, Kenji Hosomi2,Kenichi Imai2, Shigeru Ishimoto8, Changwoo Joo3, Shunsuke Kanatsuki1,Ryuta Kiuchi2, Takeshi Koike7, Harphool Kumawat9, Yuki Matsumoto7,Koji Miwa7, Manabu Moritsu10, Megumi Naruki1, Masayuki Niiyama1,Yuki Nozawa1, Ryota Ota6, Atsushi Sakaguchi6, Hiroyuki Sako2, Valentin Samoilov5,Susumu Sato2, Kotaro Shirotori10, Hitoshi Sugimura2, Shoji Suzuki8,Toshiyuki Takahashi8, Tomonori Takahashi11, Hirokazu Tamura7,Toshiyuki Tanaka6, Kiyoshi Tanida3, Atsushi Tokiyasu10, Zviadi Tsamalaidze5,Bidyut Roy9, Mifuyu Ukai7, Takeshi Yamamoto7 and Seongbae Yang3

1Department of Physics, Kyoto University, Kyoto 606-8502, Japan2ASRC, Japan Atomic Energy Agency, Ibaraki 319-1195, Japan3Department of Physics and Astronomy, Seoul National University, Seoul 151-747, Korea4INFN, Istituto Nazionale di Fisica Nucleare, Sez. di Torino, I-10125 Torino, Italy5Joint Institute for Nuclear Research, Dubna, Moscow Region 141980, Russia6Department of Physics, Osaka University, Toyonaka 560-0043, Japan7Department of Physics, Tohoku University, Sendai 980-8578, Japan8High Energy Accelerator Research Organization (KEK), Tsukuba, 305-0801, Japan9Nuclear Physics Division, Bhabha Atomic Research Centre, Mumbai, India10Research Center for Nuclear Physics, Osaka 567-0047, Japan11RIKEN, Saitama 351-0198, Japan∗E-mail: yudai@scphys.kyoto-u.ac.jp

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .We have measured an inclusive missing-mass spectrum of the d(π+,K+) reaction at thepion incident momentum of 1.69 GeV/c in the laboratory scattering angles between 2

and 16 with the missing-mass resolution of 2.7 MeV/c2(FWHM). In this letter, we firsttry to understand the spectrum as a simple quasi-free picture based on several knownelementary cross sections considering the neutron/proton Fermi motion in deuteron.While major spectrum structures are well understood in this picture, we have observedtwo distinct deviations; one peculiar enhancement at 2.13 GeV/c2 is due to a thresh-old cusp of the ΛN → ΣN conversion process, and the other notable thing is a shiftof a broad bump structure by about 22 ± 0.4 MeV/c2 toward low mass side mainlycontributed from hyperon resonance productions of Λ(1405) and Σ(1385)+/0.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Subject Index Kaonic nuclei, Λ(1405), Strangeness physics

1. Introduction

The measurement was carried out at the K1.8 beam line [1] of the J-PARC hadron exper-

imental hall by using a π+ beam at 1.69 GeV/c with a typical beam intensity of 3× 106

per 6-seconds spill cycle with a spill length of about 2 seconds. A liquid deuterium target

c⃝ The Author(s) 2012. Published by Oxford University Press on behalf of the Physical Society of Japan.

This is an Open Access article distributed under the terms of the Creative Commons Attribution License

(http://creativecommons.org/licenses/by-nc/3.0), which permits unrestricted use,

distribution, and reproduction in any medium, provided the original work is properly cited.

New type of Strange matterStrange Mesons (K, K-) in nuclei

S =0 NucleiS =-1 NucleiS =-2 Nuclei

0

100

200

300

400

500

Excitation Energy(MeV)

A [Z] A [Z+1] A [Z+2]

π-⊗A [Z+1]Λ ⊗ [Z]A-1

Σ ⊗ [Z+1]A-1-Σ ⊗ [Z]A-10Σ ⊗ [Z-1]A-1+

K ⊗ [Z+1]A-K ⊗ [Z]A0_

Ξ ⊗ [Z+1]A-1-Ξ ⊗ [Z]A-10

ΛΛ⊗ [Z]A-2

K-,K+( )K-, π+( ), π-,K+( )K-, π-( ), π+,K+( )K-,N( )

Kaonic Nuclei

s u

K- Meson

K-ppKN : attraction in Isospin=0

Kaonic hydrogen X-ray ; SIDDHARTA, M.Bazzi et al., NPA 881 (2012) 88-97.

Low-energy scattering measurements

Λ(1405) below the K-p threshold

K-pp : Y=1, I=1/2, Jπ=0-

Experiments on K-ppFirst evidence of K-pp with 6Li+7Li+12C by FINUDA

DISTO data: p+p→K-pp + K+ at 2.85 GeV M=2267±3±5 MeV/c2

Γ= 118±8±10 MeV

B=115+6/-5+3/-4 MeV Γ= 67+14/-11+2/-3 MeV

M. Agnello et al., PRL94, (2005) 212303

T. Yamazaki et al., PRL 104 (2010) 132502. P. Kienle et al., Eur. Phys. J. A 48 (2012) 183.

Mðp!Þ # 2255 MeV=c2 of the K$pp candidate reportedby FINUDA [16].

The X production rate is found to be as much as the!ð1405Þð¼ !&Þ production rate, which is roughly 20% ofthe total ! production rate. Such a large formation istheoretically possible only when the p-p (or !&-p) rmsdistance in X is shorter than 1.7 fm [3,4], whereas theaverage N-N distance in ordinary nuclei is 2.2 fm. Thepp ! !& þ pþ Kþ ! Xþ Kþ reaction produces !&

and p of large momenta, which can match the internalmomenta of the off-shell !& and p particles in the boundstate of X ¼ !&-p, only if X exists as a dense object. Thus,the dominance of the formation of the observed X at high

momentum transfer (#1:6 GeV=c) gives direct evidencefor its compactness of the produced K$pp cluster.As shown in Fig. 4, the peak is located nearly at the "!

emission threshold, below which the N"! decay is notallowed. The expected partial width of K$pp, #N"!, mustbe much smaller than the predicted value of 60 MeV [2],when we take into account the pionic emission thresholdrealistically by a Kapur-Peierls procedure (see [20]). Thus,#non-! ¼ #p! þ #N" ¼ #obs $ #N"! ( 100 MeV, whichis much larger than recently calculated nonpionic widthsfor the normal nuclear density, #non-! # 20–30 MeV[7,21]. The observed enhancement of #non-! roughly by afactor of 3 seems to be understood with the compact natureof K$pp [4].The observed mass of X corresponds to a binding energy

BK ¼ 103) 3ðstatÞ ) 5ðsystÞ MeV for X ¼ K$pp. It islarger than the original prediction [2,8,9]. It could beaccounted for if the $KN interaction is effectively enhancedby 25%, thus suggesting additional effects to be investi-gated [4,22]. On the other hand, the theoretical claims forshallow $K binding [10–12] do not seem to be in agreementwith the observation. We emphasize that the deeply boundand compact K$pp indicated from the present study is animportant gateway toward cold and dense kaonic nuclearmatter [15,23].We are indebted to the stimulating discussion of

Professors Y. Akaishi and R. S. Hayano. This researchwas partly supported by the DFG cluster of excellence‘‘Origin and Structure of the Universe’’ of TechnischeUniversitat Munchen and by Grant-in-Aid for ScientificResearch of Monbu-Kagakusho of Japan. One of us (T. Y.)acknowledges the support of the Alexander von HumboldtFoundation, Germany.

[1] Y. Akaishi and T. Yamazaki, Phys. Rev. C 65, 044005(2002).

[2] T. Yamazaki and Y. Akaishi, Phys. Lett. B 535, 70 (2002).[3] T. Yamazaki and Y. Akaishi, Proc. Jpn. Acad. Ser. B 83,

144 (2007); arXiv:0706.3651v2.[4] T. Yamazaki and Y. Akaishi, Phys. Rev. C 76, 045201

(2007).[5] T. Yamazaki et al., Hyperfine Interact. 193, 181 (2009).[6] M. Maggiora et al., Nucl. Phys. A835, 43 (2010).[7] M. Faber, A. N. Ivanov, P. Kienle, J. Marton, and M.

Pitschmann, arXiv:0912.2084v1.[8] N. V. Shevchenko, A. Gal, and J. Mares, Phys. Rev. Lett.

98, 082301 (2007); N.V. Shevchenko, A. Gal, J. Mares,and J. Revai, Phys. Rev. C 76, 044004 (2007).

[9] Y. Ikeda and T. Sato, Phys. Rev. C 76, 035203 (2007).[10] D. Jido, J. A. Oller, E. Oset, A. Ramos, and U.-G.

Meissner, Nucl. Phys. A725, 181 (2003); V. K. Magas,E. Oset, and A. Ramos, Phys. Rev. Lett. 95, 052301(2005).

[11] T. Hyodo and W. Weise, Phys. Rev. C 77, 035204 (2008).[12] A. Dote, T. Hyodo, and W. Weise, Phys. Rev. C 79,

014003 (2009).

0

0.5

1.0

1.5

2.0

2.5

2150 2200 2250 2300 2350 2400 2450

2150 2200 2250 2300 2350 2400 2450

200 100 0

0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

(a) large-angle proton: high-P (p)

(b) small-angle proton: low-P (p)

Γ = 118 (8)

M = 2267 (2)

M = 2267 (2)

Missing Mass ∆M(K) [MeV/c ]2

Missing Mass ∆M(K) [MeV/c ]2

M(Λ

*+p)

= 2

345

M(Λ

*+p)

= 2

345

M(K

+p+p

) = 2

370

M(K

+p+p

) = 2

370

M(Σ

+π+p

) = 2

267

M(Σ

+π+p

) = 2

267

Dev

iatio

nU

NC

/SIM

(arb

. sca

le)

Dev

iatio

nU

NC

/SIM

(arb

. sca

le)

B (K pp) [MeV]-

T

T

FIG. 4 (color online). (a) Observed DEV spectra of %MðKþÞof events with LAP emission [j cos"cmðpÞj< 0:6] and (b) withSAP emission [j cos"cmðpÞj> 0:6]. Both selected with large-angle Kþ emission [$ 0:2< cos"cmðKþÞ< 0:4].

PRL 104, 132502 (2010) P HY S I CA L R EV I EW LE T T E R Sweek ending2 APRIL 2010

132502-4

When a K! interacts with two protons, one expects thata hyperon-nucleon pair (!" p, "0 " p, or "" " n) isemitted in the opposite direction, ignoring a final stateinteraction inside the nucleus. The angular correlationbetween a ! and a proton from the same point in the target[Fig. 2(b)] clearly indicates the existence of this kind ofreaction. Even for heavy nuclei such as 27Al and 51V thesimilar correlations were observed, which might suggestthe absorption would take place at the surface of a nucleus.

In the following analysis, we use the !-p pairs emittedin the opposite direction ( cos!Lab <!0:8) only from thelight nuclear targets (6Li, 7Li, and 12C).

Since the back-to-back angular correlation between a !and a proton is so clear, it is naturally expected that the twoparticles are emitted from a ‘‘K!pp’’ intermediate system.The angular correlation is smeared out due to the Fermimotions of the two protons at the surface of a nucleus bywhich the K! is absorbed after cascading down the atomicorbits by emitting x rays. If the reaction process weresimply a two-nucleon absorption process, the mass of thesystem should be close to the sum of a kaon and two protonmass, namely, 2:370 GeV=c2. The initial motion of the twoprotons does not affect the invariant-mass distribution.

The invariant-mass distribution of the !-p pairs isshown in Fig. 3. A significant mass decrease of theK!pp system with respect to its expected mass is ob-served. It can be interpreted as a bound state composedof a kaon and two protons, hereafter abbreviated as K!pp.

In the inset of Fig. 3, the acceptance corrected invariant-mass distribution for events with two well-defined long-track protons is shown. Since the trigger and detectionacceptance are monotonically increasing functions ofthe invariant mass in this mass region, the peak fur-ther shifts to a lower mass side. The binding energyBK!pp # 115"6

!5$stat%"3!4$syst% MeV and the width # #

67"14!11$stat%"2

!3$syst% MeV are obtained from the fittingwith a Lorentzian function (folded with a Gaussian with

" # 4 MeV=c2, corresponding to the detector resolution,estimated with a Monte Carlo simulation) in the region of2:22–2:33 GeV=c2. Here, the systematic errors were esti-mated by changing the event selections in the ! invariantmass and the !-p opening angle cut as well as by takingaccount of the detector acceptance change due to possiblesystematic deviations in absolute momentum scale, reac-tion vertex distributions, etc. Although we still have ambi-guities on absolute normalization, a rough estimate on theyield of K!pp ! !" p is of the order of 0.1% perstopped K!. Consistency of the Monte Carlo simulationused for estimations of the acceptance and the resolutionswas examined by producing the K!pp events according tothe obtained mass and width. The same simulation con-ditions were applied to these events; the momentum dis-tributions of !’s and protons, the !-p opening angle

]2 invariant mass [GeV/cΛp-2.1 2.15 2.2 2.25 2.3 2.35 2.4 2.45 2.52.1 2.15 2.2 2.25 2.3 2.35 2.4 2.45 2.5

)2co

unts

/(10

MeV

/c

0

5

10

15

20

25

30

[MeV]pp-KB--250 -200 -150 -100 -50 0-250 -200 -150 -100 -50 0

2.2 2.25 2.3 2.35 2.4

arbi

trar

y un

it

-150 -100 -50 0

FIG. 3. Invariant mass of a ! and a proton in back-to-backcorrelation ( cos!Lab <!0:8) from light targets before the ac-ceptance correction. The inset shows the result after the accep-tance correction for the events which have two protons withwell-defined good tracks. Only the bins between 2.22 and2:33 GeV=c2 are used for the fitting.

]2c invariant mass [MeV/- p-1080 1100 1120 1140 1160 1180 1200

-π1080 1100 1120 1140 1160 1180 1200

)2 cco

unts

/(M

eV/

0

200

400

600

800

1000 (a)

)Λ(pLabθ cos -1 -0.5 0 0.5 1-1 -0.5 0 0.5 1

coun

ts

0

20

40

60

80

100 (b)

FIG. 2. (a) Invariant-mass distribution of a proton and a #! for all the events in which these two particles are observed, fitted by asingle Gaussian together with a linear background in the invariant-mass range of 1100–1130 MeV=c2. (b) Opening angle distributionbetween a ! and a proton: solid line, 6Li, 7Li, and 12C; dashed line, 27Al and 51V. The shaded area ( cos!Lab <!0:8) is selected as theback-to-back event.

PRL 94, 212303 (2005) P H Y S I C A L R E V I E W L E T T E R S week ending3 JUNE 2005

212303-3

Theoretical work on K-ppK-pp does exist !! ...but maybe broad (consistent with EXPs)

(MeV)

ATMS Yamazaki & Akaishi, PLB535

(2002) 70.

Faddeev Shevchenko, Gal, Mares,

PRL98 (2007)

082301.

Faddeev Ikeda & Sato,

PRC79 (2009)

035201.

Variational Wycech &

Green, PRC79 (2009)

014001.

Faddeev, Maeda, Akaishi,

Yamazaki, Proc. Jpn.

Acad., B, 89 (2013) 418.

Variational Dote, Hyodo,

Weise, PRC79 (2009)

014003.

Faddeev Ikeda,

Kamano, Sato,

PTP124 (2010) 533.

Faddeev Barnea,

Gal, Liverts, PLB 712

(2012) 132.

B 48 50-70 60-95 40-80 51.5 17-23 9-16 16

Γ 61 90-110 45-80 40-85 61 40-70 34-46 41

FSI effects ? ; V.K. Magas et al., PRC 74 (2006) 025206.

Λ*N bound state ? ; T. Uchino et al., NPA 868-869 (2011) 53.

E27: d(π+,K+) reaction

Yamazaki & Akaishi, Phys. Rev. C76 (2007) 045201.

1.69 GeV/c

Experimental setup[ d(π+, K+) reaction at pπ = 1.69 GeV/c ]

• K1.8 beam line spectrometer – 1.69 GeV/c π+ beam – Δp/p ~ 2×10-­‐3

• SKS spectrometer – 0.8 – 1.3 GeV/c for K+ – Δp/p ~ 2×10-­‐3 – ΔΩ ~ 100 msr

• Target – Liquid deuterium

15

π+

K+J-­‐PARC

K1.8 beam line

d(π+,K+) inclusive spectrum; in simulation

]2Missing Mass[GeV/c2.05 2.1 2.15 2.2 2.25 2.3 2.35 2.4 2.45 2.5

b/s

r/2M

eV]

µ[(L

ab)

o-1

6o 2

Ω /d

σd

0

2

4

6

8

10

12

(Lab)o-16o2 /dM Ω/d2 σd (Lab)o-16o2 /dM Ω/d2 σd

QFΛ

QFΣ

QFY*+πYNK-pp

Σ+

Σ0 Σ*Λ*

Λ

Range Counter System for E275 layers (1+2+2+5+2cm) of plastic scinti.

39 - 122 deg. (L+R)

50 cm TOF

One-proton tagging

Quasifree Y productions Decays from K-pp

Particle Identification in Range Counter

Spectrometer performance (Calibration) p(π+, K+)Σ+ at 1.58 GeV/c

K+ momentum is almost same as the d(π+, K+)K-­‐pp reaction. ・ Missing mass resolution of Σ+

ΔM = 2.8±0.1 MeV/c2(FWHM)

・ Missing mass resolution of the

d(π+, K+)K-­‐pp reaction ΔM = 2.7±0.1 MeV/c2(FWHM)

20

Σ+

p(π+,K+)1.69 GeV/c

]2Missing Mass [GeV/c1.25 1.3 1.35 1.4 1.45

)]2b/

sr/(4

MeV

/cµ[

(Lab

)o

-16

o 2/d

M

1/d

m2 d

0

1

2

3

4

5

6

7 datatotal

+ K+(1385)Y Ap+/+ K/ R Ap+/+ K/ Y Ap+/

Σ+ production ΔM = 3.2MeV(FWHM) Mass = 1188.92MeV

Σ+(1385) production Yπ production

Coun

ts /

1M

eV

Σ+

Σ*+

Missing Mass[GeV/c2]p(π+,K+)Σ+ @1.69 GeV/c

[deg]Lab eScattering Angle 2 4 6 8 10 12 14 16

b/sr

]µ[

Lab

1/d

md

0

100

200

300

400

500

600

700

800present data

old data

PTEP 2014, 101D03 Y. Ichikawa et al.

(a) (b)

p++pÆK++S(1385)+p++pÆK++L+pp++pÆK++S+p

Fig. 2. (a) The missing-mass spectrum of the p(π+, K+) reaction in the "+(1385) mass region. The experi-mental data are shown by black points with statistical errors. The spectrum was fitted with "(1385)+ (dashedline), #π (dotted line), and "π (dot-dashed line) productions. (b) The differential cross sections of "+ pro-duction at pπ+ = 1.69GeV/c. The present data and the referenced old data are shown by crosses with statisticalerrors and open circles, respectively.

(a) (b)

Fig. 3. (a) The missing-mass spectrum (MMd ) of the d(π+, K+) reaction for the scattering angle from 2 to16 (Lab) per 2 MeV/c2. The crosses and solid line show the experimental data and the simulated spectrum,respectively. The result of the Y ∗ peak fitting is also shown with a dashed red line for the experimental data. (b)The missing-mass spectrum (MMd ) in the 2.09 to 2.17 GeV/c2 region for the forward scattering angle from 2

to 8 (Lab) per 0.5 MeV/c2, which is shown by crosses. The fitting results are shown by solid and dashed lines(χ2/ndf = 1.11). See details in the text.

we can find three major structures in the spectrum: quasi-free # component from the reaction (1),quasi-free " component from the reaction (2), and quasi-free Y ∗ component from the reactions (3),(4). The non-resonant phase space component from the reaction (5) constitutes a broad structureunder the quasi-free Y ∗ bump.

We made an attempt to reproduce the double differential cross section d2σ/d&/dM with a sim-ulation by using the cross sections dσ/d& of each reaction obtained in the past experiments witha smearing by the nucleon Fermi motion in a deuteron. Here, we used a deuteron wave-functionderived from the Bonn potential [20]. For the π+ + “n” reactions, we used the cross sections andangular distributions of π− + p reactions in Ref. [9], assuming isospin symmetry. For the π+ + “p”reactions, we used the values in Ref. [11,12].

5/8

at Library of Research R

eactor Institute, Kyoto U

niversity on February 25, 2015http://ptep.oxfordjournals.org/

Dow

nloaded from

d(π+, K+) @1.69 GeV/c

ΣN-ΛN Cusp @2.13 GeV

Mass shift for Y* by ~30 MeV

K-pp

ΣN→ΛN cuspPeak at 2130.5±0.4±0.9 MeV

Width = 5.3+1.4/-1.2+0.6/-0.3 MeV

PTEP 2014, 101D03 Y. Ichikawa et al.

(a) (b)

p++pÆK++S(1385)+p++pÆK++L+pp++pÆK++S+p

Fig. 2. (a) The missing-mass spectrum of the p(π+, K+) reaction in the "+(1385) mass region. The experi-mental data are shown by black points with statistical errors. The spectrum was fitted with "(1385)+ (dashedline), #π (dotted line), and "π (dot-dashed line) productions. (b) The differential cross sections of "+ pro-duction at pπ+ = 1.69GeV/c. The present data and the referenced old data are shown by crosses with statisticalerrors and open circles, respectively.

(a) (b)

Fig. 3. (a) The missing-mass spectrum (MMd ) of the d(π+, K+) reaction for the scattering angle from 2 to16 (Lab) per 2 MeV/c2. The crosses and solid line show the experimental data and the simulated spectrum,respectively. The result of the Y ∗ peak fitting is also shown with a dashed red line for the experimental data. (b)The missing-mass spectrum (MMd ) in the 2.09 to 2.17 GeV/c2 region for the forward scattering angle from 2

to 8 (Lab) per 0.5 MeV/c2, which is shown by crosses. The fitting results are shown by solid and dashed lines(χ2/ndf = 1.11). See details in the text.

we can find three major structures in the spectrum: quasi-free # component from the reaction (1),quasi-free " component from the reaction (2), and quasi-free Y ∗ component from the reactions (3),(4). The non-resonant phase space component from the reaction (5) constitutes a broad structureunder the quasi-free Y ∗ bump.

We made an attempt to reproduce the double differential cross section d2σ/d&/dM with a sim-ulation by using the cross sections dσ/d& of each reaction obtained in the past experiments witha smearing by the nucleon Fermi motion in a deuteron. Here, we used a deuteron wave-functionderived from the Bonn potential [20]. For the π+ + “n” reactions, we used the cross sections andangular distributions of π− + p reactions in Ref. [9], assuming isospin symmetry. For the π+ + “p”reactions, we used the values in Ref. [11,12].

5/8

at Library of Research R

eactor Institute, Kyoto U

niversity on February 25, 2015http://ptep.oxfordjournals.org/

Dow

nloaded from

Cusp (Tan)

24

θπK dependence (＋data, ―sim)

25

< Peak position >＋ data

＋ simulation

Y* peak positions are shifted to the low mass side for all scattering angles.

Coincidence studyInclusive

Proton Coincidence“proton”=p>250 MeV/c

PTEP 2015, 021D01 Y. Ichikawa et al.

Fig. 2. (a) Inclusive missing-mass spectrum of the d(π+, K +) reaction at 1.69 GeV/c at laboratory scatteringangles from 2 to 14. (b) Missing-mass spectrum of the d(π+, K +) reaction with one proton in the mid-dle of the RCA on each side (Seg2, 5). (c) The coincidence probability of a proton obtained by dividing thecoincidence spectrum (b) with the inclusive spectrum (a). Hatched spectra show the contamination from themisidentification of π± in the RCA.

At this stage, the acceptance of our range counter system is not taken into account. The accep-tance correction needs information on the decay modes of the “K − pp”-like structure. This studywas carried out by requiring coincidence of two protons in the RCAs. All possible combinations oftwo segments were used in this analysis; the combination of Seg1 and 4 gives the largest yield.In such a condition, we can measure the missing mass of X in the d(π+, K + pp)X process bydetecting two protons in the decay of the ppX system, the mass of which is MMd , in three cate-gories: (i) "p, " → pπ−, (ii) #0 p, #0 → "γ → pπ−γ , and (iii) YπN → ppππ . The first twomodes, (i) and (ii), are non-mesonic and the X is one pion (and γ ). The last one, (iii), is mesonicand the X is two pions. Therefore, the missing-mass spectrum of MX should show different dis-tributions for each decay mode. Figure 3 shows such missing-mass square spectra for MX . Threedistributions estimated for each decay mode are shown in the figure by fitting the height of eachtemplate distribution. These templates were made from the simulation, which assumes the reactionof π+d → K +W, W → pY (p(Yπ)) with uniform productions and decays in the center-of-masssystem. The probability of each final state has been estimated from the fitting of M2

X spectra.Thus, we can correct the missing-mass spectrum with the acceptance of the RCAs, which depends

on the decay mode. Each event is given a weight, equal to the probability of belonging to a specificdecay mode, as a function of MMd and M2

X . The RCA acceptance is almost smooth in the missingmass except near the threshold of each decay mode. Figure 4(a) shows a missing-mass distribution for

5/8

at Library of Research R

eactor Institute, Kyoto U

niversity on February 25, 2015http://ptep.oxfordjournals.org/

Dow

nloaded from

Proton Coincidence Rate

“K-pp”-like structureΣN-ΛN Cusp

@2.13 GeV

0.4%

Decay modes classification with two protons

d(π+,K+pp)X ppX = pp-π- for Λp mode, pp-π-γ for Σ0p mode, pp-ππ for Yπp modePTEP 2015, 021D01 Y. Ichikawa et al.

(a) (b) (c)

Fig. 3. Missing-mass square spectra of X obtained in two-proton coincidence events in thed(π+, K + pp)X reaction. Each spectrum shows the mass square of X for a different MMd region: theleft (a) shows the QF" region (MMd < 2.22 GeV/c2), the center (b) shows the “K − pp”-like structureregion (2.22 < MMd < 2.35 GeV/c2), and the right (c) shows the QFY ∗ region (MMd > 2.35 GeV/c2). Thespectra were fitted with three components of #p (dashed line), "0 p (dot-dashed line), and Yπp (dotted line)decay modes.

two-proton coincidence events of the "0 p final state (ii) with the acceptance correction. According tothe simulation, the contribution of the "0 p → "0 p rescattering background is negligible because ofthe two high-momentum proton coincidence. The spectrum was fitted with a relativistic Breit–Wignerfunction:

f (MMd) = (2/π)MMdm0$(q)

(m20 − MM2

d)2 + (m0$(q))2. (2)

The mass-dependent width was $(q) = $0(q/q0), in which q (q0) is the momentum of the"0 and proton in the "0 p rest frame at mass MMd (m0). The obtained mass and width are2275 +17

−18 (stat.) +21−30 (syst.) MeV/c2 and 162 +87

−45 (stat.) +66−78 (syst.) MeV, respectively. This cor-

responds to the binding energy of the K − pp system of 95 +18−17 (stat.) +30

−21 (syst.) MeV and theproduction cross section of the “K − pp”-like structure decaying to "0 p of dσ/d&K − pp→"0 p =3.0 ± 0.3 (stat.) +0.7

−1.1 (syst.) µb/sr. The systematic errors of these values were estimated taking intoaccount uncertainties in the fitting ranges, the binning of the missing-mass spectrum, the detec-tion efficiency of two protons in the RCA, and the Breit–Wigner shape by changing the Lorentzianfunction folded with the missing-mass resolution. The differential cross section of the “K − pp”-like structure of the #p decay mode (i) was also estimated from the fitting assuming the samedistribution of MMd . Thus, a branching fraction of the “K − pp”-like structure was obtained as$#p/$"0 p = 0.92 +0.16

−0.14 (stat.) +0.60−0.42 (syst.). This ratio was discussed from a theoretical point of

view and predicted to be 1.2 using the chiral unitary model in Ref. [27].Next, we try to understand the ratio histogram (Fig. 2(c)) with the obtained K − pp mass distribution

of f (M Md). By using the mass distribution for the “K − pp”-like structure and the double-differentialcross section of the inclusive (π+, K +) process d2σ

d&d M Md(M Md)inclusive, we can obtain the ratio

6/8

at Library of Research R

eactor Institute, Kyoto U

niversity on February 25, 2015http://ptep.oxfordjournals.org/

Dow

nloaded from

“K-pp”-like structureMass :

Width :

Binding Energy : PTEP 2015, 021D01 Y. Ichikawa et al.

Fig. 4. (a) Missing-mass spectrum of the d(π+, K +) reaction for two-proton coincidence and the "0 p decaybranch events. The mass acceptances of the RCAs are corrected. The spectrum was fitted with a relativisticBreit–Wigner function (see text for details). We found a mass 2275 +17

−18 (stat.) +21−30 (syst.) MeV/c2 and a width

162 +87−45 (stat.) +66

−78 (syst.) MeV. (b) The coincidence probability of one proton for the middle segment of theRCA, as in Fig. 2(c), together with the interpretation spectra shown as a colored line. See text for details.

histogram, shown in Fig. 4(b) as a plot colored in pink, which is calculated as

Rp(M Md) =C × f (M Md) × η1p(M Md)

( d2σd%d M Md

(M Md))inclusive, (3)

where C is the normalization constant, and η1p(M Md) is the detection efficiency of a proton in themiddle segments of the RCA (Seg2, 5). The blue line in Fig. 4(b) is an assumed flat component rep-resenting the conversion processes and the contamination from the misidentification of π± in theRCA. Red points with error bars in Fig. 4(b) are the sum of the pink points and blue line. The nor-malization constant C and the amplitude of the flat component (blue line) were adjusted to minimizethe differences between the black and red points. Thus, the obtained one-proton coincidence prob-ability spectrum of the broad enhanced region could be reproduced by the “K − pp” signal and flatbackground.

What is the nature of the “K − pp”-like structure? It should have strangeness −1 and baryon numberB = 2 from the observed reaction mode, so that the hyper charge Y = 1. As for the spin of theK − pp system, a K − is theoretically assumed to couple with a spin-singlet (S = 0) p–p pair inan S-wave (L = 0), so that the J P = 0−, presumably. An alternative view of the system as a &∗ pbound state [28] also predicts the bound-state spin to be 0. There is also a theoretical prediction of a(Y, I, J P) = (1, 3/2, 2+) dibaryon as a π&N–π"N bound state [29].

4. Summary We have observed a “K − pp”-like structure in the d(π+, K +) reaction at1.69 GeV/c with coincidence of high-momentum (>250 MeV/c) proton(s) at large emission angles(39 < θlab. < 122). A broad enhancement in the proton coincidence spectra is observed aroundthe missing mass of 2.27 GeV/c2, which corresponds to a binding energy of the K − pp system of95 +18

−17 (stat.) +30−21 (syst.) MeV and a width of 162 +87

−45 (stat.) +66−78 (syst.) MeV. The branching fraction

between the &p and "0 p decay modes of the “K − pp”-like structure was measured for the first timeas (&p/("0 p = 0.92 +0.16

−0.14 (stat.) +0.60−0.42 (syst.).

7/8

at Library of Research R

eactor Institute, Kyoto U

niversity on February 25, 2015http://ptep.oxfordjournals.org/

Dow

nloaded from

PTEP 2015, 021D01 Y. Ichikawa et al.

(a) (b) (c)

Fig. 3. Missing-mass square spectra of X obtained in two-proton coincidence events in thed(π+, K + pp)X reaction. Each spectrum shows the mass square of X for a different MMd region: theleft (a) shows the QF" region (MMd < 2.22 GeV/c2), the center (b) shows the “K − pp”-like structureregion (2.22 < MMd < 2.35 GeV/c2), and the right (c) shows the QFY ∗ region (MMd > 2.35 GeV/c2). Thespectra were fitted with three components of #p (dashed line), "0 p (dot-dashed line), and Yπp (dotted line)decay modes.

two-proton coincidence events of the "0 p final state (ii) with the acceptance correction. According tothe simulation, the contribution of the "0 p → "0 p rescattering background is negligible because ofthe two high-momentum proton coincidence. The spectrum was fitted with a relativistic Breit–Wignerfunction:

f (MMd) = (2/π)MMdm0$(q)

(m20 − MM2

d)2 + (m0$(q))2. (2)

The mass-dependent width was $(q) = $0(q/q0), in which q (q0) is the momentum of the"0 and proton in the "0 p rest frame at mass MMd (m0). The obtained mass and width are2275 +17

−18 (stat.) +21−30 (syst.) MeV/c2 and 162 +87

−45 (stat.) +66−78 (syst.) MeV, respectively. This cor-

responds to the binding energy of the K − pp system of 95 +18−17 (stat.) +30

−21 (syst.) MeV and theproduction cross section of the “K − pp”-like structure decaying to "0 p of dσ/d&K − pp→"0 p =3.0 ± 0.3 (stat.) +0.7

−1.1 (syst.) µb/sr. The systematic errors of these values were estimated taking intoaccount uncertainties in the fitting ranges, the binning of the missing-mass spectrum, the detec-tion efficiency of two protons in the RCA, and the Breit–Wigner shape by changing the Lorentzianfunction folded with the missing-mass resolution. The differential cross section of the “K − pp”-like structure of the #p decay mode (i) was also estimated from the fitting assuming the samedistribution of MMd . Thus, a branching fraction of the “K − pp”-like structure was obtained as$#p/$"0 p = 0.92 +0.16

−0.14 (stat.) +0.60−0.42 (syst.). This ratio was discussed from a theoretical point of

view and predicted to be 1.2 using the chiral unitary model in Ref. [27].Next, we try to understand the ratio histogram (Fig. 2(c)) with the obtained K − pp mass distribution

of f (M Md). By using the mass distribution for the “K − pp”-like structure and the double-differentialcross section of the inclusive (π+, K +) process d2σ

d&d M Md(M Md)inclusive, we can obtain the ratio

6/8

at Library of Research Reactor Institute, Kyoto U

niversity on February 25, 2015http://ptep.oxfordjournals.org/

Dow

nloaded from

PTEP 2015, 021D01 Y. Ichikawa et al.

(a) (b) (c)

Fig. 3. Missing-mass square spectra of X obtained in two-proton coincidence events in thed(π+, K + pp)X reaction. Each spectrum shows the mass square of X for a different MMd region: theleft (a) shows the QF" region (MMd < 2.22 GeV/c2), the center (b) shows the “K − pp”-like structureregion (2.22 < MMd < 2.35 GeV/c2), and the right (c) shows the QFY ∗ region (MMd > 2.35 GeV/c2). Thespectra were fitted with three components of #p (dashed line), "0 p (dot-dashed line), and Yπp (dotted line)decay modes.

two-proton coincidence events of the "0 p final state (ii) with the acceptance correction. According tothe simulation, the contribution of the "0 p → "0 p rescattering background is negligible because ofthe two high-momentum proton coincidence. The spectrum was fitted with a relativistic Breit–Wignerfunction:

f (MMd) = (2/π)MMdm0$(q)

(m20 − MM2

d)2 + (m0$(q))2. (2)

The mass-dependent width was $(q) = $0(q/q0), in which q (q0) is the momentum of the"0 and proton in the "0 p rest frame at mass MMd (m0). The obtained mass and width are2275 +17

−18 (stat.) +21−30 (syst.) MeV/c2 and 162 +87

−45 (stat.) +66−78 (syst.) MeV, respectively. This cor-

responds to the binding energy of the K − pp system of 95 +18−17 (stat.) +30

−21 (syst.) MeV and theproduction cross section of the “K − pp”-like structure decaying to "0 p of dσ/d&K − pp→"0 p =3.0 ± 0.3 (stat.) +0.7

−1.1 (syst.) µb/sr. The systematic errors of these values were estimated taking intoaccount uncertainties in the fitting ranges, the binning of the missing-mass spectrum, the detec-tion efficiency of two protons in the RCA, and the Breit–Wigner shape by changing the Lorentzianfunction folded with the missing-mass resolution. The differential cross section of the “K − pp”-like structure of the #p decay mode (i) was also estimated from the fitting assuming the samedistribution of MMd . Thus, a branching fraction of the “K − pp”-like structure was obtained as$#p/$"0 p = 0.92 +0.16

−0.14 (stat.) +0.60−0.42 (syst.). This ratio was discussed from a theoretical point of

view and predicted to be 1.2 using the chiral unitary model in Ref. [27].Next, we try to understand the ratio histogram (Fig. 2(c)) with the obtained K − pp mass distribution

of f (M Md). By using the mass distribution for the “K − pp”-like structure and the double-differentialcross section of the inclusive (π+, K +) process d2σ

d&d M Md(M Md)inclusive, we can obtain the ratio

6/8

at Library of Research R

eactor Institute, Kyoto U

niversity on February 25, 2015http://ptep.oxfordjournals.org/

Dow

nloaded from

PTEP 2015, 021D01 Y. Ichikawa et al.

(a) (b) (c)

Fig. 3. Missing-mass square spectra of X obtained in two-proton coincidence events in thed(π+, K + pp)X reaction. Each spectrum shows the mass square of X for a different MMd region: theleft (a) shows the QF" region (MMd < 2.22 GeV/c2), the center (b) shows the “K − pp”-like structureregion (2.22 < MMd < 2.35 GeV/c2), and the right (c) shows the QFY ∗ region (MMd > 2.35 GeV/c2). Thespectra were fitted with three components of #p (dashed line), "0 p (dot-dashed line), and Yπp (dotted line)decay modes.

two-proton coincidence events of the "0 p final state (ii) with the acceptance correction. According tothe simulation, the contribution of the "0 p → "0 p rescattering background is negligible because ofthe two high-momentum proton coincidence. The spectrum was fitted with a relativistic Breit–Wignerfunction:

f (MMd) = (2/π)MMdm0$(q)

(m20 − MM2

d)2 + (m0$(q))2. (2)

The mass-dependent width was $(q) = $0(q/q0), in which q (q0) is the momentum of the"0 and proton in the "0 p rest frame at mass MMd (m0). The obtained mass and width are2275 +17

−18 (stat.) +21−30 (syst.) MeV/c2 and 162 +87

−45 (stat.) +66−78 (syst.) MeV, respectively. This cor-

responds to the binding energy of the K − pp system of 95 +18−17 (stat.) +30

−21 (syst.) MeV and theproduction cross section of the “K − pp”-like structure decaying to "0 p of dσ/d&K − pp→"0 p =3.0 ± 0.3 (stat.) +0.7

−1.1 (syst.) µb/sr. The systematic errors of these values were estimated taking intoaccount uncertainties in the fitting ranges, the binning of the missing-mass spectrum, the detec-tion efficiency of two protons in the RCA, and the Breit–Wigner shape by changing the Lorentzianfunction folded with the missing-mass resolution. The differential cross section of the “K − pp”-like structure of the #p decay mode (i) was also estimated from the fitting assuming the samedistribution of MMd . Thus, a branching fraction of the “K − pp”-like structure was obtained as$#p/$"0 p = 0.92 +0.16

−0.14 (stat.) +0.60−0.42 (syst.). This ratio was discussed from a theoretical point of

view and predicted to be 1.2 using the chiral unitary model in Ref. [27].Next, we try to understand the ratio histogram (Fig. 2(c)) with the obtained K − pp mass distribution

of f (M Md). By using the mass distribution for the “K − pp”-like structure and the double-differentialcross section of the inclusive (π+, K +) process d2σ

d&d M Md(M Md)inclusive, we can obtain the ratio

6/8

at Library of Research R

eactor Institute, Kyoto U

niversity on February 25, 2015http://ptep.oxfordjournals.org/

Dow

nloaded from

Σ0p mode

E27 Summaryd(π+,K+) missing mass spectrum at 1.69 GeV/c

threshold cusp at 2.13 GeV/c2 mass shift of ~30 MeV in Y* region

proton coincidence an enhancement of “K-pp”-like structure

Mass @ 2.275 GeV/c2