Study of the γ decay of high-lying states in 208 Pb via inelastic scattering of 17 O ions

A measurement of the high-lying states in Pb has been made using O beams at 20 MeV/u. The gamma decay following inelastic excitation was measured with the detector system AGATA Demonstrator based on segmented HPGe detectors, coupled to an array of large volume LaBr3:Ce scintillators and to an array of Si detectors. P l m a l com a o w (γ,γ’) a a, o a 5-8 MeV energy interval, are presented.


Introduction
The use of heavy ions inelastic scattering at approximately 20 MeV/u to study highly excited states (up to the region of the Giant Quadrupole Resonance) is a good tool when the measurement of the subsequent gamma decay is also performed with high resolution. Some partial results of the most recent experiments of this type, performed to investigate the electric-dipole (E1) response of nuclei at energies around the particle threshold, are reported in this contribution. The understanding of the electric-dipole response at energy around the binding energy is presently attracting considerable interest since the dipole strength distribution in that region affects considerably the reaction rates in astrophysical scenarios [1,2], where photodisintegration reactions are important. In addition the E1 strength is also interesting because it is expected to provide information on the neutron skin and thus on the symmetry energy of the equation of state [3][4][5][6][7]. The first evidence of an accumulation of low-lying E1 strength in heavy nuclei, larger than that due to the tail of the giant dipole resonance (GDR), dates back to early 70's [8]. However, only in recent years, experimental and theoretical investigations, on both stable and radioactive nuclei, revealed that this is a common phenomenon in most atomic nuclei [9][10][11][12][13][14][15][16][17][18][19][20][21][22][23][24]. The accumulation of E1strength around the particle separation energy is commonly denoted as pygmy dipole resonance (PDR) (see e.g. [5]) due to the much smaller size of its strength in comparison with the giant dipole resonance (GDR). The hydrodynamical model describes this pygmy strength as associated to the vibration of the neutron skin. An interesting feature in the region of the pygmy resonance has been observed [25][26][27][28][29] in a number of different stable nuclei, by comparing results of photonscattering and scattering experiments. In particular, it has been found that one group of states is excited in both type of reactions, while another group of states at higher energies is only excited in the ( ') case. These experimental findings are in qualitative agreement with different phonon models which predict a low-lying isoscalar component dominated by neutron-skin oscillations and a higher-lying group of states with a stronger isovector character associated to the tail of the giant dipole resonance. The use of an additional probe as the inelastic scattering of 17 O at 20 MeV/u which has, similarly to alpha particles, a rather strong isoscalar character is expected to add valuable information on the quest of the nature of these low-lying E1 states. In addition, with the same experiment we intended to study of the gamma and neutron decay of the Isoscalar Giant Quadrupole Resonance (ISGQR) in the 10-13 MeV range. In the past, inelastic scattering of 17 O ions at the energy of 22 MeV/u was used to study the gamma decay of the ISGQR of 208 Pb. The strength of the resonance and its coupling with low-lying collective states [30,31] were obtained. However, the existing measurements were performed using low-resolution gamma-ray detectors. Although the presence of a fine structure superimposed on the broad bump of the ISGQR in 208 Pb has been Study of the γ decay of high-lying states in 208 Pb via inelastic scattering of 17 [32,33]. Different techniques have been used to extract the energy scales associated to the fine structure, [32][33][34][35][36][37][38][39]. Comparison with second-RPA calculations indicates that the energy scales can indeed arise from the first step of the damping mechanism, that is the coupling of the 1p-1h states to the 2p-2h states (see e.g. [32,33]). This calls for additional high resolution investigations which are feasible using germanium detectors and also LaBr 3 :Ce scintillator arrays for the gamma decay measurement.
The experiment described in this paper was made at LNL-INFN laboratory using the Tandem-ALPI accelerator complex [40]. In Sec. 2, a detailed description of the experimental technique and of the gamma and particle detection systems is given, while in Sec. 3 preliminary results of the experiment are discussed for the case of 208 Pb.

Experimental technique and setup
In this experiment an 17 O beam at the energy of 20 MeV/u in the laboratory frame, provided by PIAVE-ALPI accelerator system of the Legnaro National Laboratories [40], was used together with a 208 Pb target.
In the upper panel of figure 1 a schematic illustration of the experimental setup is displayed. The choice of 17 O beam is motivated by the fact that this nucleus has a rather small neutron separation (i.e. 4.1 MeV). This property allows to avoid to have background in the gamma spectra at energy E> 4 MeV due to the projectile when one wants to investigate the gamma emission from the target excitation. In fact, if an excitation energy larger than 4.1 MeV is transferred to the projectile, the neutron emission channel becomes dominant and thus the outgoing nucleus becomes 16 figure 2. From this figure one sees a clear separation of the different oxygen isotopes. A picture of the gamma-ray detection system is given in the bottom panel of figure 1. In this picture the gamma detectors are on the right side. The gamma detectors are part of two separated arrays: i) the AGATA (Advanced GAmma-ray Tracking Array) Demonstrator [42,43], namely the first step of the new generation segmented HPGe gamma-ray spectrometer AGATA, and ii) an array of 8 large volume (3.5" x 8") LaBr 3 :Ce scintillators from the HECTORplus array [44,45]. These scintillators couple the best properties characterizing inorganic scintillators (high efficiency, subnanosecond time resolution) with an energy resolution surpassed only by that of germanium detectors.

Preliminary results
In the analysis of the experiment the information from the Si telescopes is used not only for the selection of the reaction channel but also for the correlation of the gamma-ray energy with the excitation energy transferred to the target nuclei. This quantity can be measured with the Total Kinetic Energy Loss (TKEL) of the projectile, which is the difference between the Total Kinetic Energy (TKE) measured in an event and the energy corresponding to an elastic scattering event. The gamma spectrum obtained with the AGATA Demonstrator after selecting the inelastically scattered 17 O events is shown (in blue) in the upper panel of figure 3, together with the spectrum (in red) obtained with the additional requirement that the energy of the gamma rays equals the TKEL values within a window ±1.5 MeV wide. By examining these two spectra it is evident that this TKEL condition enhances the relative intensity of the 208 Pb gamma-ray transitions at 2615 keV (3 -) and at 4085 keV (2 + ), with respect to the background. In the bottom panel of figure 3 the same spectra described above, in this case for the LaBr 3 :Ce scintillator array, are displayed. The high resolution data obtained with AGATA show (see upper panel of figure 3) that the 4085 keV gamma line is splitted in two components. This splitting originates from Doppler shift due to the 208 Pb target nuclei recoil motion and the two components are associated to events in which the 17 O scattered ions are detected by the left/right silicon telescope, respectively. The speed of the recoils is of the order of 0.5% of the speed of light. While this value appears quite small it is enough to cause a shift of more than 10 keV for high-energy gamma rays. In order to perform a Doppler correction for the recoil, we calculated, with simple kinematics considerations, the velocity vector of the recoil associated to each pad of the silicon telescopes. The gamma emission direction was determined instead using the position information from AGATA detectors. The results of this procedure are shown in figure 4, where the AGATA spectra before (black line) and after (red line) Doppler correction are compared. As can be seen the splitting of 4085 keV gamma line is correctly removed once the Doppler correction is performed.
Although the gamma-ray spectrum in the 5-8 MeV energy range is dominated by E1 transitions, some E2 transitions are also present (this is also known from previous Nuclear Resonance Fluorescence experiments [46,47]). It is then important to have the possibility to separate the two contributions through the different angular distribution of the emitted gamma rays. In the case of the AGATA Demonstrator it is possible to measure the emission direction of each gamma ray with a 03001-p.3 remarkable precision (~ 1°), thanks to the Pulse Shape Analysis and tracking algorithms. We considered for each event the angle ( ,recoil ) between the gamma-ray emission direction and the 208 Pb recoil velocity vector (reconstructed using the information from the silicon telescope pad which detected the 17 O ion). Figure 5 shows the angular distribution obtained for the 4085 keV line. The measured variation in intensity (W( ,recoil )) as a function of the angle is well reproduced with the expected trend for the E2 transition in 208 Pb (2 + @ 4084 keV  g.s.). Although the use of tracking algorithms greatly improves the Peak/Total ratio for the AGATA Demonstrator [29,30], as compared to traditional HPGe arrays, still a significant number of counts in the spectra are associated to Compton events and to single/double escape events. This can be clearly seen, for example, in the AGATA spectra of figure 3 and figure 4. In order to remove from the spectrum line-shape the effect of such background, we applied the Compton unfolding techniques implemented in the RADWARE software package [48]. The response function of the AGATA Demonstrator was computed with GEANT4 [49,50], for photon energies in the range from 1 MeV up to 15 MeV. The result of this unfolding procedure is shown in figure 6. In this last figure the original gamma spectrum is displayed in the upper panel (black line), while the blue spectrum in the bottom panel is after the unfolding. The removal of the Compton background and of the single escape peaks in the blue spectrum is evident. One can see in particular the single escape peaks removed around the 5715 keV line (indicated with a black arrow). The spectra in figure 6 are displayed in the 5-8 MeV energy interval, dominated by E1 transitions associated to the PDR mode in 208 Pb. A comparison of the experimental results with theoretical calculations requires the extraction of the B(E1) for each state of the resonance from the experimental crosssections. In principle, the absolute cross-section obtained from the DWBA calculation should be compared to the experimental cross-section. We decided, however, to evaluate the relative strength of each state compared to a reference one, and scale all B(E1) values with the value for the reference state found in literature [32]. We used as a reference line the strongest E1 transition, at 5512 keV, using the value of B(E1) measured in [32]. In addition the energy dependence was deduced with DWBA calculations. Figure 7 displays the experimental values of the B(E1) of the PDR states in 208 Pb measured with our setup (in blue) and with the NRF technique (in green), convoluted with a Lorentzian curve with a width of 500 keV. There is a remarkable overlap between the two curves below ~6.5 MeV, while for higher energies there is an abrupt change in the response for the two different probes. This indicates that there is a splitting of the PDR 03001-p.4

NSRT12
in 208 Pb similar to what has been observed for lighter nuclei with the (α,α'γ) technique [14][15][16][17]. The lowenergy part of the resonance is excited equally well by heavy ions and photons, while the high-energy part is weakly excited by ions. In order to study the neutron decay from the GQR and GMR region different region of excitation energy above the neutron binding energy were selected. In particular the gamma-ray spectra measured with the AGATA Demonstrator corresponding to the decay from high lying states (>7 MeV) are displayed in figure 8. The spectra of

Conclusions
The present study, although in its preliminary form, has shown an interesting result and an interesting opportunity related to a future analysis. The interesting result concerns the E1 response in the pygmy resonance region. Similarly to what was found using the (α,α'γ) reaction, also in this case the results seem to indicate that there are two groups of states one with a more isoscalar character and the other with a more isovector character. Data aiming at studying the neutron decay of the Giant Quadrupole Resonance in the 208 Pb by the high resolution measurement of the following gamma decay are also presented in their preliminary form. The future analysis of these data is expected to provide information on the neutron decay of the GQR and GMR with unprecedented resolution. The neutron decay is important to shed light on the damping mechanisms of giant resonances and to extract the direct decay component. This allows to test theory in detail.