PHYSICAL REVIEW VOLUME 176, NUMBER 5 25 DECEMBER 1968

Search for the Electroproduction of the Nf(1470)Resonance from Deuterium* J. L. ALBERI,J. A. ATPEL,? R. J. BUDNITZ,~J. CHEN, J. R. DUNNING,JR., M. GOITEIN,~ K. HANSON,D. C. IMRIE,~~C. A. MISTBETTA,** AND RICHARDWILSON IIarvard University, Cambridge, Massaclzusetts 01451 (Received 30 July 1968)

The energy spectrum of electrons inelastically scattered at 20' lab from deuterium has been measured for an incident electron energy of 1578 MeV. The virtual photon cross section has two peaks at mass values corresponding to the A(1236) and N1(1525) resonances. No peak was observed in the mass region of the N'(1470) Roper resonance. The upper limit for the contribution of the Roper resonance to the virtual photon cross section is estimated to be 120 pb. I. INTRODUCTION for the electro~roductionof this resonance suggests-- a large excitation at high four-momentum transfers; RESONANCE with the same quantum numbers other models suggest small effects.14 as the nucleon, (I,Jp)= and a mass of A (+,if), The above experiments have investigated scattering approximately 1470 MeV, has been predicted in pion- from protons. The aim of the experiment reported nucleon phase-shift analyses by Roper et al,' and by here was to determine whether the electroproduction other^.^^^ A peak at the appropriate mass value has of the Roper resonance from neutrons was s~ciently been observed in small-angle p-p and n-p inelastic large to produce a clear peak in the energy spectrum scattering at several energie~."~ The total photon- of electrons inelastically scattered from deuterium at a proton cross section does not exhibit an obvious bump four-momentum transfer near 0.1 (B~V/C)~.For com- in the vicinity of 1470 MeV; the presence of the reso- parison purposes, a limited amount of less precise data nance must be inferred from a complicated analysis of was obtained with the target filled with hydrogen. the photoproduction angular distribution. Two recent The reaction was studied at a low four-momentum analyses have predicted small Roper resonance con- transfer since the hadron scattering experiments noted tributions?JO Similarly, no obvious peak has been ob- above have indicated that the A(1236) and N'(1470) served at the appropriate mass value in the energy resonance cross sections decrease more rapidly with spectrum of electrons scattered inelastically from pro- four-momentum transfer than those of the higher pion- tons at four-momentum transfers between 0.1 and nucleon resonances. The application of static threshold 4 (B~V/C)~."-'~One model (relativistic N/D model14) relations to the electroproduction of nucleon reso- * Work supported by the U. S. Atomic Energy Commission. nances" also indicates that the relative contribution Partially supported by a Danforth Foundation Graduate Fello~vship. of the Roper resonance may be larger at small four- $ Presently at the Lawrence Radiation Laboratory, University momentum transfers. of California, Berkeley, Calif. Several alternative SU (3) classifications of the Roper $ Partially supported by a Frank Knox Memorial Fellowship and by an IBM Fellowship. resonance have been suggested, but the existing evi- 11 Permanent address: University College London, London, dence appears to favor the assignment of the resonance England. ** Presently at the University of Wisconsin, Madison, Wisc. to the {ib).'j LipkinlGhas observed that if the reso- L. D. Roper, Phys. Rev. Letters 12,340 (1964) ; L. D. Roper, nance is a member of the {io),then its photoproduction R. M. Wright, and B. T. Feld, Phys. Rev. 138, B190 (1965). from a proton is forbidden by U-spin conservation, 2B. H. Bransden, P. J. O'Donnell, and R. G. Moorhouse, Phys. Rev. 139, B1566 (1965). whereas photoproduction from a neutron is allowed, P. Auvil, A. Donnachie, A. T. Lea, and C. Lovelace, Phys. l.e., Letters 12, 76 (1964). -+ (forbidden) K. J. Foley, R. S. Jones, S. J. Lindenbaum, W. A. Love, S. y+p N+(+,++), Ozaki, E. D. Platner, C. A. Quarles, and E. H. Willen, Phys. Rev. U=# U=$ Letters 19, 397 (1967). I. M. Blair, A. E. Taylor, W. S. Chapman, P. I. P. Kalmus, y+n -+ No(+,$+), (allowed) . (1) J. Litt, M. C. Miller, D. B. Scott, H. J. Sherman, A. Astbury, U=l U=l and T. G. Walker, Phys. Rev. Letters 17, 789 (1966). 0 E. Gellert, G. A. Smith, S. Wojcicki, E. Colton, P. E. Schlein, Similarly, Lipkin noted that decay selection rules and H. K. Ticho, Phys. Rev. Letters 17, 884 (1966). 7 E. W. Anderson, E. J. Bleser, G. B. Collins, T. Fujii, J. Menes, N. F. Ramsey, J. K. Walker, and Richard Wilson, Phys. Rev. F. Turkot, R. A. Carrigan, Jr., R. M. Edelstein, N. C. Hein, 156, 1490 (1967). T. J. McMahon, and I. Nadelhaft, Phys. Rev. Letters 16, 855 l2 H. L. Lynch, J. V. Allaby, and D. M. Ritson, Phys. Rev. (1966). 164, 1635 (1967). 8G. Bellettini, A. Cocconi, A. N. Diddens, E. Lillethun, J. 18 F. W. Brasse, J. Engler, E. Ganssauge, and M. Schweizer, P. Scanlon, A. M. Shapiro, and A. M. Wetherell, Phys. Letters DESY Report No. 67/34, 1967 (unpublished) ; Nuovo Cinlento 18, 167 (1965). 55A, 679 (1968). 9 F. A. Berends, A. Donnachie, and D. L. Weaver, Nucl. Phys. 14P. L. Pritchett, N. S. Thornber, J. D. Walecka, P. A. B4, 1 (1967). Zucher, Stanford Linear Accelerator Report No. TP-297, 1968 lo Y. C. Chau, R. G. Moorhouse, and N. Dombey, Phys. Rev. (unpublished). 163, 1632 (1967). 16 A. Donnachie, Phys. Letters 24B, 420 (1967). l1 A. A. Cone, K. W. Chen, J. R. Dunning, Jr., G. Hartwig, 16 H. J. Lipkin, Phys. Letters 12, 154 (1964). 1632 Al2I3ER1~etat. 176

FIG. 1. Side view of electron spectrometer. occur if the Roper resonance can be assigned to the { io}. For example, a member of the { 10) cannot decay into a member of the (10) or the (81, so that the I f decay of the Roper resonance into the A(1236) and a pion is forbidden. It has been observed that the in- elastic decay of the N'(1470) is not A(1236)+a,15 so that this part of Lipkin's prediction appears to be con- firmed. Recently, Donnachie15 has suggested, on the basis of fixed-t dispersion relations, that the photo- .60 .80 1.33 1.20 1.40 production of the Roper resonance from neutrons is I'IB~VI orders of magnitude greater than that FIG.2. Doubly diBerential inelastic scattering cross sections protons, a result which tends to confirm Lipkin's for hydrogen and deuterium at 19.98' lab, 1.578-MeV incident assignment. electron energy. The dashed line is the radiative correction applied to the deuterium clata.

11. APPARATUS AND DATA REDUCTION 111. RESULTS AND DISCUSSION The apparatus, which is shown in Fig. 1, has been Figure 2 shows the doubly differential cross sections described in detail by Goitein17 and by Budnitz et a1.18 d2u/dE'dQ for hydrogen and deuterium as a function Only the scattered electron was detected in this ex- of the final electron energy. The dashed curve is the periment. The apparatus comprised a half-quadrupole radiative correction applied to the deuterium data. In magnetic spectroineter at 20' to the incident beam, the deuterium measurements, peaks are observed followed by a threshold gas Cerenlrov counter and a for both A(1236) and Jf(1525) resonances at the lead-Lucite shower counter. The momentum resolution appropriate final electron energy, while the less exten- of the spectrometer was approximately 2.070 (full sive hydrogen data indicate the presence of a peak width at half-maximun~j.The electron solid angle was at 1525 h2eV. The 117'(1470) is not strongly excited 0.8 msr. An external electron beam from the Cambridge from either hydrogen or deuterium. The deuterium-to- electron accelerator, with an energy of 1578 MeV, was hydrogen cross-section ratio is approximately 1.7 over focussed to forni a 2-mm-dianl spot on a 5-cm-thick the region studied. liquid deuterium (or hydrogen) target. The horizontal position of the electron beam after the target was monitored by a tuned rf cavity. A Faraday cup, whose stability mas monitored by a secondary emission moni- tor, measured the incident beam flux. The data have been corrected for the effects of counter efficiencies, target-produced background, and electron radiation. The method used to make the radiative correction has been described by Mistrettalg and is similar to the procedure of Bj~rlien.~~The data were normalized to a concurrent n~eas~irementof the elastic electro-proton cross section at the same incident energy and scattering angle. The over-all normalization error is estimated to be 5%.

l7 M. Goitein, Ph.D. thesis, Harvard University,. . 1968 (un- published). l8 R. J. Budnitz, J. A. Appel, L. Carroll, J. Chen, J. R. Dunning, FIG.3. (dZu/dE'dO)/F~for deuterium as a function of isobar Jr., M. Goitein, K. Hanson, D. C. lmrie, C. Mistretta, J. K. mass. Curve (e) is the best fit to the data using three nonrela- Walker, and Richard L\'ilson, Phys. Rev. 173, 1357 (1968). tivistic Breit-Wigner resonance curves plus a straight-line back- I9 C. Mistretta. Ph.D. thesis, Harvard Universitv. 1968 ground. Curves (a)-(c) are the individual contrit~utionsof the (unpublished). A(1236), Roper, and iV1(1525) resonances, respectively; (d) is an J. D. Bjorken, Ann. Phys. (N. Y.) 24, 201 (1963). estimate of the background. 176 SEARCH FOR ELECTROPRODUCTION OF N1(1470) RESONANCE 1633

TABLEI. (d2u/dE'dQ)/r~for deuterium. The tail of the quasi- elastic peak has been subtracted from the five cross-section values with W<1.12 BeV. The errors shown are relative only; there is an additional scale error of 5%

PER (d2u/dE'dil)/r~ Error E' (BeV) W (BeV) (fib) (fib) 0.703 0.717 0.73 1 0.747 0.758 0.775 0.789 0.804 0.818 0.831 0.846 0.861 0.875 0.888 0.903 0.917 0.934 0.945 0.961 0.975 0.990 1.001 1.018 1.032 1.050 FIG.4. (a) X2 of fit as a function of the assumed mass of the 1.061 Roper resonance. (6)Peak value of fitted Roper resonance con- 1.077 tribution as a function of assumed mass. 1.090 1.103 1.118 Following the virtual photon cross section 1.134 may be written as 1.146 1.162 1.253 1.266 1.278 1.291 where I 1.305 and a straight-line background. The masses and widths of ~=[1+2(l+q0~1q2) tan2(40)]-I. the A(1236) and Nf(1525) were constrained to known values. The Roper resonance was assumed to have a E and E' are the incident and scattered electron width of 210 MeV and several fits were attempted with energies, respectively, e is the electron scattering angle, the mass varying from 1370 to 1500 MeV, since the q2 is the invariant square of the four-momentum trans- precise peak position expected in electroproduction is fer, and qo= (E-El). K is the equivalent photon energy, uncertain. The experimental deuterium quasi-elastic (W2- M2)/2M, where W is the pion-nucleon c.m. energy peak was folded into the Breit-Wigner shapes to ap- and M is the proton mass. UT and uoare the transverse proximate the broadening of the resonances due to the and longitudinal parts of the virtual photon cross nucleon motion in the deuteron and the experimental section. r~is the number of transverse virtual photons energy resolution. per unit energy and angle of the scattered electron, and Curve (e) in Fig. 3 is the best fit; curves (a)-(c) are is the transverse polarization of the virtual photon. the contributions of the individual resonances and (d) Figure 3 and Table I present the virtual photon cross section of deuterium as a function of W. The tail is an estimate of the background. Figure 4(a) shows of the quasi-elastic peak has been subtracted from the the X2 value of the fits and Fig. 4(b) the peak value of the Roper resonance contribution as a function of the five cross-section values with ?V<1.12 BeV. For the mass assumed for the resonance. The errors shown in remainder of the data, its effect was negligible. To obtain an upper limit for the contribution of the Roper the figure are purely statistical. X2 has a broad minimum at the appropriate mass and there is a large region over resonance, the deuterium data were fitted, using the sum of three nonrelativistic Breit-Wigner curves plus which the fitted value of the Roper resonance con- tribution remains approximately constant at 60 pb.

21 L. N. Hand, Phys Rev. 129, 1834 (1963). However, the dominant errors in the estimation of the 1634 ALBERI et al. 176

Roper resonance contribution are clearly not statistical, but are due to uncertainties in the detailed behavior of the tails of the other resonances and the background. We estimate that the uncertainty in these factors is of the same order as the fitted Roper resonance con- tribution. Therefore, our upper limit for the contribu- tion of the Roper resonance to the virtual photon cross section of deuterium is 120 pb at W=1430 MeV, q2 = 0.16 (B~V/C)~,E= 0.80. The photoproduction analysis of Ref. 9 predicts a Roper resonance contribution at W=1350 MeV of approxinlately 12% of the cross section at the A(1236) peak. Extrapolating this result to FV= 1470 MeV, assuming a Breit-Wigner shape for the Roper resonance, the predicted contribution at the resonance peak is approximately 50y0 larger than the experimental upper limit. RG.5. (l/r~)(d2u/dE'dS1)[G~~(0)/G~v(qz)]2 as a function of Although the hydrogen cross-section values measured IqlZ, the square of the three-momentum transfer to the isobar in the present work are of limited precision, some useful system, at W = 1525 MeV. comparisons can be made with previous data. Since the deuterium-to-hydrogen cross-section ratio is observed approximated by the dipole fit relationz2: to be practically independent of W in this energy region, an estimate of the virtual photon cross section of hydrogen can be obtained by scaling the best fit The total photoproduction cross section shown in Fig. to the deuterium data by the measured cross-section 5 is that obtained by Brasse et al.13 from an analysis of ratio. bubble-chamber data. At W=1236 MeV, this method predicts a virtual ACKNOWLEDGMENTS photoncross section of 520f 41 pb for q2=0.22 (B~V/C)~, ~=0.9.Lynch et a1.12 have obtained o,=444&13 pb, It is a pleasure to thank the staffs of the Ilarvard ao=88f 23 ,ub at q2=0.2 (B~V/G)~,implying, from Eq. University Cyclotron Laboratory and the Cambridge (I), a virtual photon cross section of 520=t24 pb at Electron Accelerator for their cooperation during the E= 0.9, in excellent agreement with the present estimate. preparation and execution of this experiment. G. Similarly, at W= 1525 MeV, the virtual photon cross Thompson aided in the data taking and A. Litke con- section of hydrogen is estimated to be 184f20 pb. tributed valuable programming assistance. Finally, we Figure 5 shows that this result is in good agreement should like to thank A. Donnachie and N. Dombey for with electroproduction results at higher four-momen- several interesting discussions. tum transfers."-l3 For the purposes of this comparison, 22 M. Goitein, J. R. Dunning, Jr., and Richard Wilson, Phys. the nucleon isovector form factor G,wv(q2) has been Rev. Letters 18, 1018 (1967).