14 November 1996

PHYSICS LETTERS 6

Physics Letters B 388 (1996) 450-456

New data on OZI rule violation in pp annihilation at rest

OBELIX Collaboration

A. Bertina, M. Bruschi a, S. De Castro a, A. Ferretti”, D. Galli a, B. Giacobbe a, U. Marconi ‘, M. Polia,‘, M. Piccinini a, N. Semprini-Cesari ‘, R. Spighi a, S. Vecchi a,

A. Vezzani a, F. Vigotti a, M. Villaa, A. Vitale a, A. Zoccoli a, G. Belli b, M. Corradini b, A. Donzellab, E. Lodi Rizzini b, L. Venturelli b, A. Zenoni ‘, C. Cicalod, A. Masoni d, G. Puddu d, S. Serci d, P.P. Temnikov d, G. Usai d, O.Yu. Denisov e, O.E. Gorchakov e,

V.P. Nomokonov e, S.N. Prakhov”, A.M. Rozhdestvensky e, M.G. Sapozhnikov e, V.I. Tretyak e, I? Gianotti . f, C. Guaraldo f, A. Lanarof, V. Lucherini f, F. Nichitiu f,2,

C. Petrascu f,2, A. Rosca f,2, V.G. Ableev s3, C. Cavion a, U. Gastaldi s, M. Lombardi a, A. Andrighetto h, M. Morando h, G. Bendiscioli i, V. Filippini’, A. Fontana’, P. Montagna’,

A. Rotondi i, A. Saino’, P. Salvini’, F. Balestraj, E. Bottaj, T. Bressani j, MP. Bussaj, L. Bussoj, D. Calvoj, P. Cerelloj, S. Costaj, L. Favaj, A. Felicielloj, L. Ferreroj, A. Filippij, R. Garfagnini’ J, A. Grassoj, D. D’Isepj, A. Maggioraj, S. Marcelloj,

D. Panzierij, D. Parenaj, E. Rossettoj, F. Toselloj, G. Zosij, M. Agnellok, F. Iazzi k, B. Minetti k, G.V. Margagliotti e, G. Paulie, S. Tessaroe, L. Santi”

a Dipartimento di Fisica, Universita di Bologna and INFN, Sezione di Bologna, Bologna, Italy b Dipartimento di Chimica e Fisica per i Materiali, Universita di Brescia and INFN, Sezione di Torino, Turin, Italy ’ Dipartimento di Chimica e Fisica per i Materiali, Universita di Brescia and INFN, Sezione di Pavia, Pavia, Italy

d Dipartimento di Scienze Fisiche, Universita di Cagliari and INFN, Sezione di Cagliari, Cagliari, Italy e Joint Institute for Nuclear Research, Dubna, Russia

’ Laboratori Nazionali di Frascati dell’INFN, Frascati, Italy s Laboratori Nazionali di Legnaro dell’lNFN, Legnaro, Italy

b Dipartimento di Fisica. Universita di Padova and INFN, Sezione di Padova, Padua. Italy i Dipartimento di Fisica Nucleare e Teorica, Universita di Pavia. and INFN Sezione di Pavia, Pavia, Italy

.i Dipartimento di Fisica, Universita di Torino and INFN, Sezione di Torino, Turin, Italy k Politecnico di Torino and INFN, Sezione di Torino, Turin, Italy

r Istituto di Fisica. Universita di Trieste and INFN, Sezione di Trieste, Trieste, Italy m Istituto di Fisica, Universita di Udine and INFN, Sezione di Trieste, Trieste, Italy

Received 1 I July 1996; revised manuscript received 2 September 1996 Editor: L. Montanet

0370-2693/96/$12.00 Copyright 0 1996 Published by Elsevier Science B.V. All rights reserved.

PII SO370-2693(96)0 1198-7

OBELIX Collaboration/Physics Letters B 388 (1996) 450-456 451

Abstract

The results of measuring the ratio R = Y( &r+a- ) /Y( WP’T-) for antiproton annihilation at rest in a gaseous and a liquid hydrogen target are presented. It was found that the value of this ratio increases with the decrease of the dipion mass, which demonstrates the difference between the 4 and o production mechanisms. An indication of the momentum transfer dependence of the apparent 021 rule violation for the (b production from the ‘St initial state was found.

In the recent experiments with stopped antipro- tons at LEAR (CERN) [ l-51 strong violation of the Okubo-Zweig-Iizuka rule [6] was found. The ratio of the yields of 4 and o mesons R(4X/wX) turned out to be about 30-50 times larger than the OZI pre- diction R( 4X/wX) M 4 x 10F3. In particular, the degree of the OZI rule violation was found [ 1,5] to depend strongly on the quantum numbers of the initial state.

A number of theoretical models were developed to explain these data. Approaches based on the traditional ideas [7,8] fail to reproduce all the features of the 4 production observed now (for a review, see [9]). It has been suggested [ lo-121 that the nucleon wave function contains some sS pairs already at small mo- mentum transfer. Then the abundant 4 production in NN annihilation should not be considered as a viola- tion of the OZI rule, because it does not involve dis- connecting quarks diagram. In this model the 4 pro- duction is described in terms of rearrangement of the sS pairs already stored in the nucleons.

This is different from the standard view that the 4, which is practically a pure SS state, should be produced in the interactions of the non-strange hadrons only due to a small admixture of the light quarks in its wave function and the production amplitudes of Cp and w mesons should be rather similar.

In this paper, we study the production mechanisms of 4 and w mesons in the reactions

(1)

(2)

’ Dipartimento di Energetica, Universiti di Firenze. ’ On leave of absence from Department of High Energy Physics,

Institute of Atomic Physics, Bucharest, Romania. 3 On leave of absence from Joint Institute for Nuclear Research,

Dubna, Russia.

of antiproton annihilation at rest. The ratio R = Y( &r+r-) /Y( WT+T-) was measured for different invariant masses of the dipion system. It was found that the value of this ratio increases with the decrease of the dipion mass, which demonstrates the difference between the 4 and w production mechanisms.

Antiproton annihilations at rest were obtained us- ing the OBELIX spectrometer on the M2 beam line of LEAR. A detailed description of the spectrometer can be found elsewhere [ 131. The experimental setup consists of a number of subdetectors arranged around the Open Axial Field Magnet. The Time Of Flight (TOF) system contains two coaxial barrels of plas- tic scintillators for charged particle identification and trigger. The Jet Drift Chamber (JDC) provides geom- etry reconstruction of the event and particle identifica- tion by energy loss measurement. The High Angular Resolution Gamma Detector (HARGD) reconstructs the multiplicity, the direction and the energy of the detected gammas.

In order to investigate reactions ( 1) and (2), events with 4 charged particles were analyzed. To study w- meson production, the trigger based on the TOF sys- tem was used. It selected events characterized by 4 hits in the inner barrel of the scintillators and 3-4 hits is the outer one. To study &meson production, a sam- ple of data was collected using also a “slow particle” condition - at least one time difference between hits in the inner and outer barrels had to be greater than 8 ns. This condition was used to enrich the sample by charged kaons.

Data were taken using two hydrogen targets: liquid and gas at a pressure of 3 atm. In the liquid target, annihilations occur mainly from S-wave initial states, whilst in the gas target annihilations from S- and P- waves are possible with an approximately equal prob- ability. A sample of 16 x lo6 annihilations in the liquid target with the “slow particle” condition was collected and 1 x lo6 events without it were taken. For the gas

452 OBELIX Collaboration/Physics Lefters B 388 (1996) 450-456

Liquid target Gas target

M(K’K-), GeV/c’ M(K’K-), GeV/c’

- cl 2 I N

2 “0 rnL 2

0.5 0.75 1 1.25 1.5

M(n’neno), GeV/c’

25 d)

20

15

10

5

0 ~ 0.5 0.75 1 1.25 1.5

M(n’n-no), GeV/d

Fig. 1. The K+K- and P+P-& invariant mass spectra for the Fig. 2. The background-free ‘IT+Z- invariant mass distributions reactions pp + K+K-?r+r- (a,b) and pp -+ 27r+27r-7r0 (c,d) for the reactions pp + @-+r- (a,b) and pp -+ o&6 (c,d) of antiproton annihilation at rest in the liquid target (a,c) and gas of antiproton annihilation at rest in the liquid target (a,~) and gas target (b,d) (uncorrected for apparatus acceptance). target (b,d) (uncorrected for apparatus acceptance).

target, 1.5 x lo6 and 3 x lo6 events were taken, re-

spectively.

The following criteria were used for event selec- tion. The event should have 4 tracks with the zero to-

tal charge. Data samples with the “slow particle” trig-

ger were used to select events with the K+K-r+r- final state. Two kaons of opposite charge should be

recognized by dE/dx and TOE Finally, a 4C kine- matic fit to the hypothesis pp --+ K+K-IT+T- was

applied by imposing energy and momentum conser- vation. This leaves 59 299 KKmr events for the liquid

target sample and 70 15 events for the gas target sample which satisfy the fit with a confidence level of more than 5%. A 1C kinematic fit to the hypothesis ~7p -+

&v-r+s--r” was applied to select the correspond- ing channel. This leaves 127904 and 230769 events

for liquid and gas sample, respectively, which satisfy the fit with a confidence level of more than 10%.

The KfK- and 7r’r-r” invariant mass spectra for the selected events are shown in Fig. 1. Clear signals from +J and w mesons are apparent. Their approxi- mation by the sum of a Breit-Wigner function with the fixed 4(w) width smeared by a Gaussian distri-

< 160 3 140

5: 120

ii?100

80

60

40

20

Liquid target Gas target

M(n’rrs), GeV/c’

[*$, , , , , +)’ 0.4 0.6 0.8 1

0.6 0.8 M(n’n-), GeV/c”

M(n’n-), GeV/c’ M(n’n-), GeV/c’

bution gives the following values for the mass and experimental resolution u (for the gas target sam- ple): rn+ = (1019.1 f 0.5) MeV/c*, ~4 = (5.3 *

0.6) MeV/c’ and m, = (781 + 1) MeV/c*, CT, =

(18.9 f 0.1) MeV/c*.

Background free 7r+7~ - invariant mass distributions

(uncorrected for apparatus acceptance) for reactions (1) and (2) are shown in Figs. 2a-d. For reaction

( 1) (4 production) the spectra were obtained after

subtraction of the events from the bins nearby the (p peak.

For reaction (2) ( w production), the so-called A-

subtraction procedure [ 141 was used. The w decay amplitude is proportional to the parameter

A = I& x P?r+ I2 A’

where p are the three-momenta of the charged pions from the w decay in its rest frame, and A’ is the maxi- mum possible value of Ipr- x p,+ /*. The distribution of this parameter for events coming from o decay is proportional to 1, while for the background events the corresponding &distribution should be flat. This was verified for the analyzed sample. The events were sep-

OBELIX Collaboration/ Physics Letters B 388 (19%) 450-456 453

arated into two samples: one with A < 0.5 and the other with A > 0.5. Subtracting the spectrum with A < 0.5 from that with A > 0.5 one can obtain a background-free distribution.

In the &rTT- invariant mass distributions associated with 4 production shown in Figs. 2a,b, there is no evidence for a p-meson peak. This is consequence of the apparatus acceptance. Charged kaons with a momentum less than 200 MeV/c cannot arrive at the JDC. As a result, it is not possible to observe large values of M,,.

The dipion mass distributions associated with w production are shown in Figs. 2c,d. They are dom- inated by the p-meson peak, which turns out to be especially pronounced for annihilation in gas. In the case of the liquid target (Fig. 2c), remnants of the j’z( 1270) peak, truncated by the phase space, can be seen. These distributions are rather similar to those measured in the previous experiments of w production for annihilation in the liquid [ 141 and gas [ 151 tar- gets. From Figs. 2c,d it turns out that at small masses, 300 MeV/c* < M ?rT < 500 MeV/c*, the invariant mass distributions are rather flat. This interval is also suitable for detection of pions produced together with 4 mesons in reaction ( 1) .

The ratio R = Y(&~TT)/Y(~~T~) (where Y is the yield of 4 or w mesons) was determined using the numbers of 4 and w obtained from the KfK- and p+n-7r” invariant mass distributions for the dipion mass within the interval 300 MeV/c* < M,, < 500 MeV/c* and compared with the same ratio ob- tained for all events. To avoid a loss of useful events, we did not use the background-free distributions. To estimate the number of&(w) mesons, the correspond- ing part of the invariant mass spectrum was approxi- mated by the sum of a Breit-Wigner function with the fixed 4(w) width, smeared by a Gaussian distribution and a polynomial function to describe the background.

To evaluate the detection efficiency of reactions ( 1) and (2), the Monte Carlo simulated events were re- constructed and analysed using the same selection cri- teria and kinematic cuts as for the real data. The phase space distribution of the dipion mass was assumed both for 4 and w production.

In Table 1, the number of events Nev selected for each reaction, the detection efficiencies 8, the yields Y of 4 and w mesons, and the corresponding ra- tios R are reported. The yield is calculated as Y =

Nev/hL,~ B), where B is the fraction of 4 decays into K+K-. The total number of annihilations in the target Naon was measured using the TOF inner barrel of scintillator counters. Nann is equal to 33.6 x lo6 for gas and 10.6 x lo6 for liquid for the samples obtained without the “slow particle” condition and 130.8 x lo6 for gas and 494.3 x lo6 for liquid for the samples ob- tained with the “slow particle” trigger. The yields are affected by a systematic uncertainty of about 6%. It is due to uncertainties of different selection procedures, particle identification, evaluation of the apparatus ef- ficiency, and the total number of annihilated antipro- tons. In addition, the total yields are underestimated due to low acceptance of the $p channel.

The total yields obtained for the w&r- channel Y = (719174) x 10m4for liquidand Y = (628f34) x 10m4 for gas are in general agreement with the results of the previous measurements Y = (660 f 60) x 10e4 for liquid [ 141 and Y = (682 f 74) x 10v4 for gas [15].Theyieldsof&r+r-channelY= (3.5&0.4)x 10e4 for liquid and Y = (3.7 f 0.5) x 10e4 for gas are lower than those measured by other authors, Y = (4.6 f 0.9) x 10e4 for liquid [ 141 and Y = (5.4 f 1 .O) x 10e4 for gas [ 11. The reason is the lack of the events from the &J channel discussed above. It is important to stress that this lack of @p events is not relevant to our subsequent analysis, which is mainly based on the measurements of the &r7r yield at small dipion masses.

As one can see from Table 1, for all events, without any cut on dipion invariant mass, the ratio R( #I/O) is about 5 x 10m3 for annihilation both in liquid and in gas. This is in agreement with the OZI rule prediction R(qb/w) x 4 x 10m3. However, for small masses of the dipion system R( t#~/o) increases by a factor 3-4, which leads to a substantial deviation from the OZI rule prediction.

This effect has been observed for the first time and it is important to demonstrate that it is not an artefact due to a difference in phase space for #J and w production. Different forms of this correction have been discussed (e.g, [ 161 and references therein). It was argued [ 161 that the two-body phase space factor til+‘, where 1 is the angular momentum between the two mesons and q is the c.m.s. momentum, is unrealistic and the phe- nomenological factor q. exp( -A (s - sii) I/*) adopted by Vandermeulen [ 171 is more correct. Here s is the invariant mass squared of the fip system, sij = (mi +

454

Table 1

OBELIX Collaboration /Physics Letters B 388 (1996) 450-456

Yields of 4 and w mesons in reactions ( 1) and (2) of annihilation of stopped antiprotons in liquid and gas targets (only statistical errors are taken into account). R( 4/o) is the ratio of the yields, Rv(qf~/o) is the same ratio corrected by the Vandermeulen scaling factor and R,-( 4/o) is the ratio corrected for the difference in the three-body phase space

Liquid HZ

x!” E, % YX 104 R(+/w) x IO3 Rv(t~/o) x IO3 Rc(d~/w) x IO3

All events

&r+?r- 853 f 94 1.08 f 0.03 3.5 f Cl.4 4.9 * 0.8

wr+7r- 10.3 f 1.6

32042 & 1541 4.60 f 0.41 719It74

300 MeV/c2 < M,, < 500 MeV/c’

&l+n- 408 & 67 2.29 f 0.09 0.7fO.l

> 16.5 + 3.5 15.4 It 3.3 I’= 4.3

w7r+w 20.4

19181t244 4.67 f 0.40 42.4 f 6.5

Hz at 3 atm

All events

4 7r+w 292 f 43 1.24 & 0.02 3.7 f 0.5

> 5.9 f 0.9 12.5 f 2.0

w?i-+?r- 71640& 1011 3.80 f 0.20 628 zt 34

300 MeV/c’ < M,, < 500 MeV/c2

@r+n- 104* 23 2.41 f 0.06 0.7 f 0.2 29.3 f 8.6 27.4 f 8.0 36.2 f 10.6

w-r+G?- 2701 f 578 3.74 f 0.26 23.9 f 5.4

mj) ‘, mi, mj are the masses of the mesons in the final substantial deviation from the 021 rule prediction,

state and A = - 1.2 GeV-’ is the phenomenological observed at small masses of the dipion system, could

parameter obtained from a successful fit of different not be attributed to the difference in the &rr and

two-body annihilation channels. wrr phase spaces.

The corrected values of R are shown in Table 1.

Here Rv( 4/w) denotes the ratio corrected by the Van-

dermeulen scaling factor, calculated for the quasi-two- body final state r,A( w) - M,, with the effective mass M,, = 400 MeV/c2. We have verified that with this

value of M,, it is possible to approximate the three-

body phase space of 4( W)~T~T at low dipion masses

300MeV/c2 < M,, < 500 MeV/c2 by the two-body

phase space of 4(w) - M,,. The last column of Table 1 contains the ratios

Rc(cJ/w) corrected for the difference in the three-

body phase space for the reactions of @TIT and wrrrr, It turns out that after this correction the value of R only increases. The Vandermeulen scaling factor gives only a small correction to the R. Therefore, a

The increase of R at small dipion masses for anni-

hilation in liquid could be explained as a manifesta- tion of the difference in the amplitudes of # and w production. The conservation of C- and P-parities in

strong interactions unambiguously couples the spin- triplet 3St initial state with the +(w)rr final state,

when two pions are in the S-wave relative to each other. The spin-singlet initial ‘So state couples with the Ql(w)zr7r system when the two pions are in P-wave, i.e. the final state is +( w)p. The partial wave analysis of the &r7r channel demonstrates [ 181 that in the S- wave the production of 4 mesons is completely dom- inated by the spin-triplet 3Si initial state, whereas for the w-meson production both the 3St and ‘SO states are important [ 141. Thus, neglecting the annihilation

OBELIX Collaboration/Physics Letters B 388 (1996) 450-456 455

-5ot,O”““““““““” 200 400 600 800 1000

M (MeV/c’)

Fig. 3. The ratio RC = 4X/0x for different reactions of pp annihilation at rest as a function of the mass M of the system X. The crosses indicate the result of this experiment.

from the P-waves, we will get that the ratio R for an- nihilation in liquid is

R = y(h-+r-) M y(4d~+~->s) Y ( w7r+77- ) Y(w(?r+r->s> + Y(q) .

(3)

It is clear that at small dipion masses, far from the p peak, this ratio should increase.

Analogous consideration for annihilation in gas is more complicated due to the substantial contribution of the annihilation from the P-wave. However, it is im- portant to stress that the ratio R at small dipion masses in gas is high not due to the increase of the &r+rr- yield. As one could see from Table 1, Y( &r+?r-) is the same for annihilation in liquid and in gas at small dipion mass, but the w&r- yield for 300 < M,, < 500 MeV/c* mass interval in gas is less than in liq- uid. Therefore again, as in the case of annihilation in liquid, we observed the difference in the C$ and u pro- duction dynamics.

In Fig. 3 the values of RC are compared with the results of other measurements of two-body final states for antiproton annihilation at rest. Some apparent de- pendence of the degree of the OZ.&rule violation on the mass of the system produced with 4 is seen. For anni- hilation at rest NN -+ 4X, a decrease in the mass of

X means an increase in the momentum transferred to 4. However, one should be cautious, because different two-body reactions were measured for different initial states. Thus, the 4y, +p and 40 final states for anni- hilation from the S-wave produced from the ‘&-J initial state, whereas the &r and C$JQ channels come from the 3St initial state. As discussed above, in liquid at small dipion masses both &r+r- and W&C chan- nels come predominantly from the 3St state. Therefore the corresponding ratio R could be directly compared with the values of R( @r/am) and R( q5~/or]). The comparison shows that for the # production from the 3St initial state the degree of the OZI rule violation in- creases with decreasing mass of the system produced with 9, i.e. with increasing momentum transfer.

It would be interesting to investigate further this possible dependence of the degree of the OZI rule violation on the momentum transfer. The same effect was found in the 4 production in the IT* N --+ c#N interaction [ 191 where the da/& distribution of the 4 production at large t considerably differs from the one for w meson, which leads to a larger 4/w ratio at large t. Direct measurements of the t-dependence of the differential cross sections for the &r and wr channels of ~7p annihilation in flight should clarify the problem.

In conclusion, the measurement of the ratio R = Y ( &rtn--) /Y ( WTT+GT-) for annihilation of stopped antiprotons in gaseous and liquid hydrogen targets was performed. It was found that the value of this ratio increases with decreasing dipion mass, which demon- strates the difference between the 4 and o production mechanisms. An indication of the momentum transfer dependence of the apparent OZI rule violation for the C$ production from the 3St initial state was found.

We would like to thank the technical staff of the LEAP machine group for their support during the runs.

We are very grateful to J. Ellis and D. Kharzeev for interesting discussions.

The JINR group acknowledges the support from the International Science Foundation, grant No. ML9300.

References

[ 1 J I. Reifenrother et al., Phys. Lea. B 267 ( 1991) 299. [2] C. Amsler et al., Phys. Lett. B 346 (1995) 363. [3] V.G. Ableev et al., Phys. L&t. B 334 (1994) 237.

456 OBELIX Collaboration /Physics Letters B 388 (1996) 450-456

[4] V.G. Ableev et al., Nucl. Phys. A 585 (1995) 577. [5] V.G. Ableev et al., Nucl. Phys. A 594 (1995) 375. [6] S. Okubo, Phys. Lett. B 5 (1963) 165;

G. Zweig, CERN Report No. 8419/TH 412 (1964); I. lizuka, Prog. Theor. Phys. Suppl. 37 ( 1966) 21.

]7] D. Buzatu and E Lev, Phys. Lett. B 329 (1994) 143. [8] MP Lecher, Y. Lu and B.-S. Zou, Z. Phys. A 347 (1994)

281; MP Lecher and Y. Lu, Z. Phys. A 351 (1995) 83.

[9] M.G. Sapozhnikov, JINR preprint E 15-95-544, Dubna ( 1995); Lecture at the XXIX St. Petersburg Winter School on Nuclear Physics and Elementary Particles, Zelenogorsk, 1995, p. 252.

[IO] J. Ellis, E. Gabathuler and M. Karliner, Phys. Len. B 217 (1989) 173.

[ 111 J. Ellis et al., Phys. Lett. B 353 ( 1995) 319. [ 121 M. Alberg, J. Ellis and D. Kharzeev, Phys. Lett. B 356

(1995) 113. [ 131 A. Adamo et al., Sov. J. Nucl. Phys. 55 (1992) 1732. [ 141 R. Bizzarri et al., Nucl. Phys. B 14 (1969) 169. [ 151 I? Weidenauer et al., Z. Phys. C 59 (1993) 387. [ 161 C.B. Dover et al., Prog. Part. Nucl. Phys. B 29 ( 1992) 2 19. [ 171 J. Vandermeulen, Z. Phys. C 37 (1988) 563. [ 181 A. Bertin et al., Proc. NAN-95 Conference, Moscow, 1995. [ 191 D. Cohen et al., Phys. Rev. L&t. 38 (1977) 269.