γ spectroscopy of 25 , 27 Ne and 26 , 27 Na

PHYSICAL REVIEW C 74, 064305 (2006) γspectroscopy of 25,27Ne and 26,27Na A. Obertelli, A. Gillibert, N. Alamanos, M. A. G. Alvarez,*F. Auger, R. Dayras, A. Drouart, N. Keeley, V. Lapoux, X. Mougeot, L. Nalpas, E. Pollacco, F. Skaza, and Ch. Theisen CEA-SACLAY DSM/DAPNIA/SPhN F-91191 Gif-sur-Yvette, France G. de France, B. Jurado,†W. Mittig, F. Rejmund, M. Rejmund, P. Roussel-Chomaz, and H. Savajols GANIL BP 5027, F-14076 Caen cedex 5, France A. Pakou and N. Patronis Department of Physics, The University of Ioannina, 45110 Ioannina, Greece (Received 24 March 2006; revised manuscript received 19 October 2006; published 7 December 2006) The γspectroscopy of 25,27Ne and 26,27Na was studied from the reaction of 26Ne with a deuterium target in inverse kinematics at 9.7 MeV/nucleon. The selectivity of the (d,p),(d,t), and (d,n) transfer reactions provides new spectroscopic information on low-lying states. The validity of the sd shell-model space for these nuclei is discussed. DOI: 10.1103/PhysRevC.74.064305 PACS number(s): 25.60.−t, 21.10.Pc, 23.20.Lv, 27.30.+t I. INTRODUCTION Two strong isospin-dependent phenomena involving the orbitals from the sd and fp shells have been experimentally suggested in the neutron-rich region for masses 20 ⩽A⩽40: the disappearance of the N=20 magic number and the appearance of N=16 as a new magic number. The first indication of the weakening of the N=20 shell closure has been given through mass measurements of sodium and magnesium isotopes [1,2]. The low excitation energy of the first 2+state [3] and the high reduced transition probability B(E2) [4]of32Mg can only be reproduced assuming an intruder configuration (2¯hω) with two neutrons in the fp shell for the ground state [5], showing that 32Mg is deformed. The same feature is suggested for 30Ne from a (p, p) measurement in inverse kinematics [6]. N=16 has been experimentally [7,8] suggested to be a possible magic number in the vicinity of 24O, as already predicted by several models [9–11]. This is interpreted as an enhancement of the spherical gap between the s1/2and the d3/2subshells of the neutron sd shell compared to its value for stable nuclei or as proton-neutron correlation effects [10]. In the latter description, the sd-fp shell gap is predicted to be considerably reduced compared to stability. Data are needed to clearly determine the shell structure of neutron-rich nuclei in the N=16 to N=20 region where both deformation and spherical shell gap evolution are expected to coexist. In this article, we report on the γspectroscopy of 25,27Ne and 26,27Na from the reaction of 26Ne with a deuterium target at 9.7 MeV/nucleon. New spectroscopic information about 25Ne, together with previous results for 27Ne [12], allow studying the intrusion of fp orbitals along the neon isotopic chain. The results obtained for 26,27Na *Present address: Departamento de FAMN, Universidad de Sevilla, E-41080 Sevilla, Spain. †Present address: Centre d’Etudes Nucl´ eaires de Bordeaux- Gradignan, F-33175 Gradignan Cedex, France. confirm the validity of the sd shell-model space for these neutron-rich sodium isotopes. II. EXPERIMENTAL SETUP The experiment was performed using the SPIRAL facility [13] of GANIL (Grand Acc´ el´ erateur National d’Ions Lourds). A26Ne beam was produced via an ISOL method: a 36S primary beam of 1 kW at 77.5 AMeV was fragmented and stopped in the thick carbon SPIRAL target. After the selection and acceleration by the CIME cyclotron, a pure 26Ne secondary beam was delivered at 9.7 MeV/nucleon with an intensity of ∼3000 pps. The charge state 26Ne5+was selected because there is no lower-mass contaminant with the same M/Q =26/5 ratio. A solid cryogenic 2H target (1 mm thick, 17 mg/cm2) developed at GANIL [14,15] was used. The choice of the target thickness results from a compromise between the low intensity and the low energy of the incoming 26Ne beam. Beamlike ejectiles were detected and identified with the VAMOS magnetic spectrometer [16,17]. VAMOS has a large momentum acceptance (±5%) and a large angular acceptance withamaximumof8 ◦(140 mrad) in the laboratory frame in the horizontal and vertical planes. In our experiment, the angular domain is restricted to [0◦,3◦]forthe(d,p) channel, due to the strongly inverse kinematics. Due to a geometrical cut by the beam pipe from the reaction chamber to the entrance of VAMOS, the polar angular acceptance was additionally limited to 2.7◦. The resulting cut was checked to be sharp. The detection of VAMOS was composed of two drift chambers to measure the kinematic properties of the ejectiles after the dipole, an ionization chamber for energy-loss measurements (E) and a scintillating plastic detector used to trigger the electronics. A time of flight (TOF) was measured between the plastic scintillator and a microchannel plate device (MCP) located upstream of the target (see Fig. 1). This detection has been designed for low-energy ejectiles by minimizing the windows along the ion path: the two drift chambers [18] 0556-2813/2006/74(6)/064305(7) 064305-1 ©2006 The American Physical Society A. OBERTELLI et al. PHYSICAL REVIEW C 74, 064305 (2006) Ne beam 26 Ne 27 D2 MCP Target EXOGAM Q1 Q2 Dipole Plastic Drift Ch. γ Ionization Ch. VAMOS 45 90 o o 135o FIG. 1. Experimental setup in the VAMOS vault. are juxtaposed so that they are constituted by one volume of isobutane delimited by only two 0.9-µm-thick Mylar windows. The ionization chamber has only one entrance Mylar window (1.5 µm thick), the plastic detector being glued to the exit of the chamber. The magnetic rigidity of VAMOS was centered on Bρ0=1.0 Tm. Ejectiles were identified in the focal plane with the horizontal position Xf, the TOF, and the energy loss E. AZidentification plot from E-TOF correlations is shown in Fig. 2: neon and sodium isotopes are produced and clearly separated. A direct A/Q assignment was done via TOF-Xf correlations [12]. Both singles and events in coincidence with deexcitation γrays were measured. γrays were measured with the EXOGAM γspectrometer [19] surrounding the target (see Fig. 1). In this experiment, 11 clovers (8 EXOGAM clovers and 3 EUROGAM size detectors) were positioned in the array at distances ranging from 11 to 17 cm from the center of the target. Four of them were located at forward angles in a 45◦ ring, 4 at 90◦and 3 at backward angles in a 135◦ring. Each clover consists of 4 ×4-fold segmented germanium crystals. The intrinsic energy resolution of the whole system was 2.6 keV full width at half maximum (FWHM) for a 1332-keV γtransition, and its photopeak efficiency for the same transition was determined to be 4.8%. A time measurement between the central-contact discriminators of EXOGAM and the plastic scintillator of VAMOS is used to select the true -10 -5 0 10 15 20 ∆E (arb. units) Time-of-flight (ns) Z=11 Z=10 Z=9 Coincidence VAMOS-EXOGAM 5 FIG. 2. (Color online) E-TOF identification plot for Zassignment. Only events in coincidence with a γray detected in EXOGAM are presented. A part of the total statistics is shown. 0 50 100 150 200 Without Doppler correction Ne 26 Counts / 8 keV 0 500 1000 1500 2000 2500 3000 3500 4000 0 20 40 60 80 100 β = 0.115Doppler corrected: E (keV) 2019(2) keV FIG. 3. γ-ray spectrum of 26Ne without (top) and with (bottom) Doppler correction. Sharp lines in the uncorrected Doppler spectrum are due to well-known transitions from room background. coincidences and substract the background due to random coincidences. III. RESULTS AND DISCUSSION A. Test case: 26Ne The first 2+excited state of 26Ne is well established at 2018 keV and has been used to validate the whole setup and analysis method. Excited states of 26Ne were populated via (d,d) and inelastic excitation in the Mylar windows of the target. The γ-ray spectrum of 26Ne is shown in Fig. 3, without (top) and with (bottom) Doppler correction. In the uncorrected spectrum, the intense low-energy exponential background and sharp lines are due to random coincidences, the noninteracting 26Ne beam being transmitted to the focal plane. It is not observed in the case of transfer-reaction products. The energy Eγwas measured from the central-contact electrode of the hit crystal. The emission angle θwas determined as the polar angle from the beam axis of the segment collecting the highest energy in the considered crystal. Addback corrections for Compton events were also performed when two adjacent crystals of a detector were hit in the same event. As the reaction vertex in the target is not reconstructed, only the mean value of the velocity βwas used. For example, the 26Ne beam is slowed down in the 2H target from β=0.142 to β=0.090. For 26Ne, a mean velocity β=0.115 was adopted, because for that value the Doppler corrected energies are the same at forward and backward angles for the well-known transition 2+→0+at 2.02 MeV, as illustrated in Fig. 4. Assuming a locally linear background, a Gaussian fit gives an energy of 2019(2) keV and a width of 49(4) keV FWHM. The observed transition corresponds to the γdecay of the first 2+state. Our measurement is in agreement with the 2018.2(1) keV [20] and 2024(5) keV [21] excitation energies previously measured. Estimated contributions to the width are gathered in Table I for both backward and forward angles. The main contribution comes from the large change of the velocity in the target due to the target thickness and the low-energy beam. The final estimated width (taken as the mean value between the 45◦and 135◦estimations) is 42 keV FWHM, in agreement with the 064305-2 γSPECTROSCOPY OF 25,27Ne AND 26,27Na PHYSICAL REVIEW C 74, 064305 (2006) 0 10 20 30 135 Ne 26 = 0.115β 5 10 15 90 Counts / 8 keV 1600 1800 2000 2200 2400 0 10 20 30 E (keV) 45 0 o o o FIG. 4. Doppler corrected deexcitation γ-ray spectrum of 26Ne for the three detection angles : 45◦(bottom), 90◦(middle), and 135◦ (top). The structure at ∼1600 keV in the 135◦spectrum comes from the Doppler correction of the 1460-keV background transition. 49(4) keV measured width. The 2H(26Ne,26Ne)2Hinelastic scattering allowed us to validate the Doppler reconstruction. B. 2H(26Ne,27Ne)1H We performed the spectroscopy of 27Ne below its neutron separation threshold Sn=1.43(11) MeV [12]. From the measurement of low-energy γtransitions in 27Ne, constraints on the multipolarity of these transitions and cross sections quantifying the selectivity of the (d,p) transfer reaction, we proposed a negative parity to a low-lying level at 765 keV with a spin Jπ=(1/2,3/2,5/2)−. Considering all existing data [12,22,23], the most likely assignment is Jπ=3/2− from the neutron p3/2orbital. This observation shows that the gap between the sd and fp shell is considerably reduced in 27Ne compared to stable N=17 isotones. More experimental information on 27Ne would help to understand the underlying shell structure: (i) the location of its first 7/2−excited state, distant of less than 1 MeV from the first 3/2−in all other N=17 isotones, and (ii) the quadrupole moment of its mass and charge distributions. C. 2H(26Ne,25Ne)3H A detailed spectroscopy of 25Ne is useful to test the validity of the sd shell-model space for its low-lying excitations. Shell-model calculations with the USD interaction [24] predict TABLE I. Contributions to the width of the 2019-keV peak of Fig. 3: the intrinsic energy resolution δEγ,the velocity uncertainty δβ, the angular uncertainties δθ due to the finite size of the segments and the scattering angle of 26Ne. Angle 45◦135◦ δEγ(keV) 3.6 4.2 δβ (keV) 34.4 40.6 δθ(segments) (keV) 18.6 15.8 δθ(26Ne) (keV) 6.2 5.3 FWHM (keV) 39.8 44.1 1 2 3 10 12 14 16 Z E* (MeV) 5/2 USD 5/2 expt. 3/2 USD 3/2 expt. S =2.74 MeV n 8 N=15 isotones + + + + FIG. 5. Comparison between published excitation energies [31] of the first 3/2+and 5/2+excited states above a 1/2+ground state for N=15 odd isotones and USD predictions. a(3/2 +,5/2 +) doublet in 25Ne at 1687 and 1778 keV, respectively. In this section, we provide new elements to determine the levels that correspond to this doublet. The 3/2+ state is described as a νd3/2neutron configuration, whereas the 5/2+state results mainly from the coupling of the neutron shell-model state νs1/2to a 2+proton excitation of the core. The structure of this 5/2+state and of the 3/2+first excited state are very different from each other. The (d,t) reaction from 26Ne is then expected to be very selective and strongly favors hole states in 25Ne like the 5/2+state of the doublet. The spectroscopic factor from 26Ne(gs) to the 5/2+state of 25Ne is predicted to be 2.3 from USD shell-model calculations, whereas the spectroscopic factor to the first 3/2+state is 0.4 [29]. Figure 5shows the systematics of the first 3/2+and 5/2+ excited states in N=15 odd isotones. USD calculations are compared to available data: for isotones with Z>10, the agreement is very good and suggests a good prediction for 25Ne. Spectroscopic information about 25Ne have been reported from the βdecay of 25F[20,25], from low-energy transfer reactions 26Mg(7Li,8B)25Ne [26], 26Mg(13C,14O)25Ne [27], and the one-neutron pickup 2H(24Ne,25Ne)1H[28], and also from higher incident-energy reactions: the one-neutron removal 9Be(26Ne,25Ne)X [23,29] and the breakup of 26Ne on 208Pb [30]. A comparison of most of the previous data and shell-model calculations within the sd shell-model space are shown in Ref. [25]. In the aforementioned experiments, no doublet was observed in the 1700-keV region within a range of 300 keV. The multinucleon transfers [26] and [27] were limited by an energy resolution not better than 100 keV. The suggestion of a doublet in Ref. [27] was not confirmed in more recent βdecay of 25F[20,25] with a much better resolution (a few keV). In Ref. [25], a 2096-keV transition has been confirmed, and the authors concluded that the USD doublet states correspond to excited states at 1702 and 2096 keV. These conclusions rely on log(ft) measurements and multipolarity constraints. The γ-ray spectrum of 25Ne obtained in this experiment is shown in Fig. 6: without (top) and with (bottom) Doppler cor- 064305-3 A. OBERTELLI et al. PHYSICAL REVIEW C 74, 064305 (2006) 0 20 40 Ne 25 without Doppler correction 0 20 40 Doppler corrected: β=0.128 Counts / 8 keV 0 20 40 Multiplicity 1 0400 800 1200 1600 2000 2400 0 4 8 12 Addback E (keV) (a) (b) (c) (d) FIG. 6. γ-ray spectra of 25Ne: without Doppler correction (a), Doppler corrected with a velocity β=0.128 (b), multiplicity 1 events only (c), and addback events (d). rection with a velocity of β=0.128. The corrected spectrum is divided into three distinct energy regions: around 1700 keV where a group of transitions is visible, a region with a very few counts at high energy and the low-energy part dominated by Compton events from higher-energy transitions. In Fig. 7, these 1550 1600 1650 1700 1750 1800 0 5 10 15 20 25 30 35 40 Counts / 8 keV Ne 25 Linear background Two Gaussians Total fit 1621(5) keV 1700(2) keV Multiplicity 1 Doppler corrected: β=0.128 2000 2500 3000 3500 4000 0 2 4 6 8 10 2075(25) keV 4025(100) keV E (keV) Counts / 25 keV S =4180(50) keV n 0200 400 600 800 1000 0 5 10 15 20 25 Counts / 8 keV 320(2) keV (a) (b) (c) FIG. 7. γspectrum of 25Ne. (a) The low-energy part of the spectrum shows a 320-keV transition, present in all angular rings. (b) Zoom of the 1700-keV region: the result of the fit for two transitions at 1621 and 1700 keV is shown. (c) High-energy region: the low background allows for sensitivity to transitions with low statistics. TABLE II. Observed transitions in 25Ne from 2H(26Ne,25Ne)3H (see Fig. 7). The experimental width of each transition is compared to a theoretical evaluation. The theoretical value does not take into account shifts from half-life effects. EγFWHMexp FWHMtheo (keV) (keV) (keV) 320(2) 5(4) 7 1621(5) 42(4) 36 1700(2) 51(7) 36 2075(25) 38(14) 45 4025(100) — 87 three regions are shown separately. It is worth noting that only a part of the momentum distribution of 25Ne was transmitted to the spectrometer. Therefore, we did not measure any absolute cross section for the (d,t) reaction channel. A zoom of the 1700-keV region [Fig. 7(b)]showstwo structures at ∼1620 and ∼1700 keV. We considered two possibilities: one (case 1) or two (case 2) transitions to reproduce the 1700-keV peak. We assumed Gaussian shapes for the transitions over a locally linear background. In case 1 [fit in Fig. 7(b)], the width of the peak is 51 keV FWHM compared to the expected 36-keV value (Table II), whereas a smaller value of 42 keV FWHM is obtained in case 2. However, it is difficult to conclude because a half-life of about 30 ps (the time necessary for 25Ne to go through the target) is enough to induce a broadening of the peak. Indeed, such a scenario is plausible for the first excited state of 25Ne: assuming a (3/2, 5/2)+state at 1700 keV, the transition to the 1/2+ground state corresponds to a E2orM1 transition (or a mixing M1/E2) with, according to Weisskopf estimates, a half-life expected to be of the order of 10 ps. That ambiguity combined with rather low statistics prevents any conclusion: no evidence could be found for a doublet at 1700 keV. The observed 1621(5)-keV transition, consistent with previous measured energies, corresponds to the decay of a 3321(6)-keV state to the 1700(2)-keV level. This 3321-keV excitation energy is in good agreement with the 3324-keV value recently published in Ref. [25] and with similar excitation energies measured via particle transfer in previous experiments [26–28]. The high-energy region of the Doppler-corrected spectrum is shown in the bottom panel of Fig. 7. The statistics are poor, and it is important to determine the random background component at these energies. In the γspectrum of 27Ne, 20 counts have been measured above 1800 keV and are considered as background for total statistics of 3654 counts, because the neutron separation energy of 27Ne is Sn=1.4MeV. For 25Ne, the high-energy spectrum (E>1800 keV) contains 137 counts for a total amount of 2662 counts. It leads to an estimation of 11% background relative to the total number of counts in the high-energy part of the 25Ne spectrum by analogy with 27Ne. It indicates that most of the high-energy counts observed in the spectrum come from the decay of 25Ne. One transition at 2075(25) keV is clearly visible in the high-energy spectrum of Fig. 7. Its width is consistent with the 064305-4 γSPECTROSCOPY OF 25,27Ne AND 26,27Na PHYSICAL REVIEW C 74, 064305 (2006) theoretical value as indicated in Table II. This transition has already been observed through γspectroscopy at 2030 [28], 2050(100) [30], and 2090(3) keV [25] and corresponds to the direct decay of a level observed via (7Li,8B) and measured by missing-mass measurement at 2030 keV [26]. This state has been proposed to correspond to the 3/2+state of the USD doublet [25]. From the selectivity of the (d,t) reaction, the ratio of the counts measured at 2075 keV to the counts in the 1700-keV transition is, after correction of the energy efficiencies of EXOGAM, 0.14, very close to the ratio of spectroscopic factors from USD calculations S(3/2+)/S(5/2+)=0.16. This comparison is a strong argument in favor of the conclusions of [25]: the 5/2+level of the USD doublet lies at 1700 keV, whereas the 3/2+lies at 2075(25) keV. The events whose energy is above 2500 keV are not directly assignable to transitions because the statistics are low. Nevertheless, we observe an amount of counts centered at 3000 keV that indicates the presence of high-energy states produced during the reaction. Five events are measured around 4025(100) keV, suggesting a possible direct branch to the ground state for the level(s) previously assigned at ∼4050 keV [20,26–28]. The low-energy part of the γspectrum of 25Ne is mainly composed of Compton events from high-energy transitions. Among the low-energy part of Fig. 7, we identified one transition present over a large background in every angular detection ring at 320(2) keV. This transition is reported in Table II with all the other measured transitions. This transition cannot be assigned to a specific level decay in this work and has not been observed in β-decay experiments despite a good energy resolution and low background. From this comparison, we can infer that the level scheme of 25Ne may contain more levels than observed in β-decay experiments as suggested by shell-model calculations. Conversely, we do not observe the 574- and 2186-keV transitions that are suggested by a β-decay experiment [25] to correspond to the decay of a 3889-keV level. This indicates that this state, not produced via (d,t), is not a neutron-hole excitation. The other structures in the low-energy spectrum of 25Ne are not present in all the angular rings and, therefore, are not considered as transitions. The level scheme obtained from this experiment is presented in Fig. 8 and compared to shell-model calculations performed with the USD interaction. Finally, we observed three already-known transitions in 25Ne produced from 2H(26Ne,25Ne)3H: 1700(2), 2075(25), and 1621(5) keV, which correspond to the decay of excited states at 1700, 2075, and 3321 keV, respectively. The 2075-keV transition is consistent with the 2090-keV transition observed in βdecay [25]. The selectivity of the (d,t) reaction, in comparison with spectroscopic factors from USD calculations, gives a strong argument to assess Jπ=5/2+to the 1700-keV level, and Jπ=3/2+to the 2075-keV state, as previously proposed [25]. High-energy counts at 4025(100) keV may indicate a branch for the direct γdecay of the already known 4050 keV excited state to the ground state. The obtained level scheme from one-neutron stripping is consistent with shell-model calculations performed within the sd shell-model space, showing no evidence for a strong component with low-lying fp orbitals in the 26Ne ground-state wave function. These fp orbitals do not seem to strongly influence the low-lying spectroscopy of 25Ne. 1621(5) Shell model USD This work Ne 25 1/2 5/2 3/2 5/2 3/2 7/2 2967 2971 0 1687 1779 3638 3593 4226 4249 4595 4648 9/2 7/2 5/2 3/2 5/2 + + + + + + + + + + + 0 1700 2075 3321 4025 4025(100) 1700(2) 2075(25) (5/2 ) (3/2 ) (1/2 ) + + + FIG. 8. Levels of 25Ne observed in 2H(26Ne,25Ne)3H. Prediction from shell-model calculations performed with the USD interaction are also presented. Going toward the neutron drip line along the neon isotopic chain, they appear to result in a negative-parity excited state at 765 keV in 27Ne [12] with a substantial amount in the ground state of 28Ne [23], and they are suggested to be intruders in the ground state of 30Ne [6]. D. Sodium isotopes In addition to neon isotopes, we have studied the spectroscopy of sodium isotopes with N=15 and N=16. Their spectroscopy, by comparison to USD shell-model calculations, tests also the validity of the sd shell-model space in that region of the nuclear chart. 1. 27Na The lowest excited states in 27Na still present some uncertainty on their spin and parity assignment. The spectroscopy of 27Na has already been studied via multinucleon transfer from 26Mg [32,34] and 14C[33]. 27Na ground state is known to be 5/2+from laser spectroscopy [35] and βdecay [36]. A low-lying 3/2+at 62 keV above the ground state has also been identified in Ref. [33]. Two transitions at 1663 and 1753 keV have been assigned to low-lying states at 1725 and 1815 keV that both decay to the 62-keV state, respectively. These states have been proposed to be 1/2 levels [33]from angular-momentum selection rules and γangular distribution for the 1663-keV transition. The parity assignment of these two states was made supposing that one of the states is the 1/2+state predicted by USD shell-model calculations and that the other state is a 1/2−intruder state from the proton pshell. The reaction 26Mg(18O,17F)27Na [32] populates significantly the 1725-keV level. M. W. Cooper et al. suggested that the 1725-keV level is a 1/2−state without excluding the opposite parity assignment [33]. In the following we resolve this uncertainty on the parity of the first 1/2 state. The γspectrum of 27Na is shown in Fig. 9, where the Doppler correction is performed with a velocity of β=0.105. 064305-5 A. OBERTELLI et al. PHYSICAL REVIEW C 74, 064305 (2006) C( C, Na)X 14 14 27 Na 27 2091 1/2+ 7/2+ 9/2+ 5/2+ 3/2+ 2197 1630 0 2224 2191 1815 1725 62 0 5/2+ 3/2+ (1/2+) (7/2+) (9/2+) (1/2- ) 1669(6) This work USD 14 1731 1663 1/2+ 0 Shell-model S =6.7 MeV n FIG. 9. Comparison of the present results among the 27Na level scheme, published data, and USD calculations. One transition is visible at 1669(6) keV, in agreement with the 1663-keV transition observed in Ref. [33]. Its width is 30(7) keV, consistent with the theoretical estimation of 30 keV. The observed transition is then identified as the decay of the 1725-keV level to the low-lying 3/2+excited state at 62 keV. The energy threshold of the EXOGAM array was too high to detect this low-energy transition. The statistics at higher energy is too low to identify any other excited state. It is important to determine the mechanism responsible for the production of 27Na to interpret properly the population of the mentioned excited state. We now show that the observed 27Na are mainly produced by a direct (d,n) transfer reaction. The production of 27Na from 26Ne+dmay have two different origins: the direct (d,n) reaction or a fusion-evaporation process 26Ne +d→28Na→27Na +n. (1) Nevertheless, two arguments are in favor of a direct reaction. (i) The very clean γspectrum suggests that the mechanism to produce 27Na from 26Ne+dwas selective and may be considered as a direct (d,n) reaction. Indeed, a fusion-evaporation process implies a statistical feeding of the excited states. As a statistical feeding depends mainly on the excitation energy and the spin of the concerned states, the two states at 1815 and 1725 keV should be fed at the same level from the statistical part of the feeding because they are almost at the same excitation energy and both assigned to have a spin 1/2. The lack of the 1753-keV transition, corresponding to the decay of the 1815-keV state to the 62-keV low-lying level, compared to the population of the 1669-keV level indicates that the population of the 1/2 states is not driven by a statistical law. (ii) The fusion Qfvalue is positive: Qf=14.2MeV. 0 1000 2000 3000 4000 5000 0 2 4 6 8 10 12 14 Na 27 E (keV) Counts / 16 keV Doppler corrected β=0.105 1669(6) keV FIG. 10. Doppler-corrected γspectrum of 27Na. The neutron separation energy of 27Na is Sn=6.726(7) MeV [37]. 0 200 400 600 800 1000 0 5 10 15 20 25 30 35 Counts / 8 keV Na 26 407(2) keV 368(2) keV 150(2) 232(3) keV 1500 2000 2500 3000 3500 4000 0 2 4 6 8 10 12 Counts / 20 keV E (keV) 1284(9) keV 1998(8) keV (a) (b) FIG. 11. Low-energy (a) and high-energy (b) part of the γspectrum of 26Na. The spectrum is Doppler corrected with a velocity β=0.095. The neutron separation energy of 26Na is Sn= 5.576(6) MeV [37]. In the case of the experiment, the kinematics imply an excitation energy of 33 MeV for 28Na. This excitation energy is very high compared to the one- and twoneutron separation energy of 28Na [Sn=3.52(8) MeV, S2n=10.37(8) MeV], showing that the one-neutron evaporation is not expected to be a favored decay of the compound 28Na. We then assume that 27Na was mainly produced via the direct (d,n) reaction. The populated levels produced from (d,n) should be proton particle states. This picture for the observed excited state is confirmed by a shell-model calculation within the sd shellmodel space with the USD interaction in which a 1/2+state is predicted at 1630 keV and described as a single-particle state with a π(d5/2)2(s1/2)1main configuration. Our data indicate then that the 1725-keV state is a 1/2+level (see Fig. 10). 2. 26Na 26Na has already been studied from transfer reactions [38– 40] and βdecay of 26Ne [41]. Recently, a detailed spectroscopy has been obtained from 14C(14C, d)26Na [42]. In our case, the origin of 26Na observed in the focal plane may be due to a charge exchange process or fusion followed by the evaporation of two neutrons, with a cross section expected to be two orders of magnitude higher in the latter case [43,44]. No spectroscopic selectivity in the population of the different states is therefore expected. The following transitions (quoted by arrows in the Dopplercorrected spectrum of Fig. 11) have been checked to be present in all the γ-detection angular rings to eliminate spurious peaks: 150(2), 232(3), 368(2), 407(2), 1284(9), and 1998(8) keV. Except for the 368-keV transition, these results are a confirmation of the transitions observed in Ref. [42]by another reaction channel. IV. CONCLUSION We performed the γspectroscopy of the neutron-rich nuclei 25,27Ne and 26,27Na from the reaction of 26Ne with deuterium in inverse kinematics at 9.7 MeV/nucleon. The use 064305-6 γSPECTROSCOPY OF 25,27Ne AND 26,27Na PHYSICAL REVIEW C 74, 064305 (2006) of a cryogenic D2target with the γspectrometer EXOGAM coupled to the magnetic spectrometer VAMOS gives access to deuteron-induced reaction and allows the high-resolution γspectroscopy of the reaction products even with a lowintensity beam (3000 pps of 26Ne in this experiment). For 25Ne produced via the one-neutron removal 2H(26Ne,25Ne)3H, no evidence for a doublet at 1700 keV is found, consistently with the conclusions of β-decay experiments [20,25]. A 2075(25)-keV level is confirmed and suggested to have Jπ=3/2+. A spin and parity Jπ=5/2+is assigned to the 1700-keV state. The observation of an unassigned transition at 320(2) keV and a consequent amount of high-energy γrays may sign the existence of bound high-energy excited states in 25Ne produced by neutron stripping. The relatively good agreement between shell-model calculations for 25Ne and the known levels of 25Ne may indicate that the sd-fp shell gap is rather large in 25Ne. In the present experiment, low-lying intruder states below 1 MeV have been observed in 27Ne [12]. In addition to recent conclusions for 28Ne [22,23], these results show that the intrusion of fp orbitals in the low-lying spectroscopy of neon isotopes occurs around N=17. For 27Na, the selectivity of the (d,n) transfer reaction allowed us to assign a 1/2+spin and parity to the 1731(6)-keV state, showing a good agreement with sd shell-model calculations. ACKNOWLEDGMENTS We acknowledge the technical staffs of GANIL for their excellent support: P. Dol´ egi´ eviez, P. Gallardo, and P. Robillard for the cryogenic target; J. Ropert and G. Voltolini for EXOGAM; and B. Lecornu and C. Mary for VAMOS. We also acknowledge the beam delivery teams of SPIRAL for the high quality of their work. [1] C. Thibault et al., Phys. Rev. C 12, 644 (1975). [2] N. Orr et al., Phys. Lett. B258, 29 (1991). [3] C. D´ etraz et al., Phys. Rev. C 19, 164 (1979). [4] T. Motobayashi et al., Phys. Lett. B346, 9 (1995). [5] A. Watt et al.,J.Phys.G7, L145 (1981). [6] Y. Yanagisawa et al., Phys. Lett. B566, 84 (2003). [7] A. Ozawa et al., Phys. Rev. Lett. 84, 5493 (2000). [8] M. Stanoiu et al., Phys. Rev. C 69, 034312 (2004). [9] M. Beiner, R. J. Lombard, and D. Mas, Nucl. Phys. A289,1 (1975). [10] T. Otsuka et al., Phys. Rev. Lett. 87, 082502 (2001). [11] A. Obertelli et al., Phys. Rev. C 71, 024304 (2005). [12] A. Obertelli et al.,Phys.Lett.B633, 33 (2006). [13] A. C. Villari et al., Nucl. Phys. A693, 465 (2001). [14] P. Dol´ egi´ eviez et al., Cryogenic system for a thin solid hydrogen target, Report Ganil A 00 01, November 29, 2000. [15] P. Dol´ egi´ eviez et al., Nucl. Instrum. Methods A 564, 32 (2006). [16] H. Savajols (VAMOS Collaboration), Nucl. Phys. A654, 1027c (1999). [17] R. Anne, Proceedings of EXON 2001, edited by E. Penionzhkevich and E. A. Cherepanov (World Scientific, Singapore, 2001). [18] E. Bougamont, internal report CEA/DSM/DAPNIA, ref. 6D6810E2100/303 (2002). [19] http://www.ganil.fr/exogam. [20] A. T. Reed et al., Phys. Rev. C 60, 024311 (1999). [21] M. Belleguic et al., Phys. Rev. C 72, 054316 (2005). [22] Zs. Dombradi et al., Phys. Rev. Lett. 96, 182501 (2006). [23] J. R. Terry et al., Phys. Lett. B640, 86 (2005). [24] B. A. Brown and B. H. Wildenthal, Annu. Rev. Nucl. Part. Sci. 38, 29 (1988). [25] S. W. Padgett et al., Phys. Rev. C 72, 064330 (2005). [26] K. H. Wilcox et al., Phys. Rev. Lett. 30, 866 (1973). [27] C. L. Woods et al., Nucl. Phys. A476, 392 (1988). [28] W. N. Catford et al.,J.Phys.G31, S1655 (2005). [29] J. R. Terry and J.-L. Lecouey, Nucl. Phys. A734, 469 (2004). [30] J. Gibelin et al., RIKEN Accel. Prog. Rep. 37, 53 (2004). [31] P. M. Endt, Nucl. Phys. A521, 1 (1990). [32] L. K. Fifield et al., Nucl. Phys. A437, 141 (1985). [33] M. W. Cooper et al., Phys. Rev. C 65, 051302(R) (2002). [34] I. Paschopoulos et al., Phys. Rev. C 18, 1277 (1978). [35] G. Huber et al., Phys. Rev. C 18, 2342 (1995). [36]D.E.Alburger,D.R.Goosman,andC.N.Davids,Phys.Rev.C 8, 1011 (1973). [37] LBNL Isotopes Project Nuclear Data Dissemination Home Page. Retrieved March 11, 2002, from http://ie.lbl.gov/toi.html. [38] E. R. Flynn and J. D. Garrett, Phys. Rev. C 9, 210 (1974). [39] K. I. Pearce et al., Phys. Rev. C 35, 1617 (1987). [40] H. Scheit et al., nucl-ex/0401023, 01/21/2004 (2004). [41] L. Weissman et al., Phys. Rev. C 70, 057306 (2004). [42] Sangjin Lee et al., Phys. Rev. C 73, 044321 (2006). [43] E. R. Flynn, J. Sherman, and N. Stein, Phys. Rev. Lett. 32, 846 (1974). [44] D. Ridikas et al., Phys. Rev. C 63, 014610 (2000). 064305-7