RAPPORT D ACTIVITÉ

Transcription

1 RAPPORT D ACTIVITÉ

2 CONTENTS Foreword Nuclear Physics Research Nuclear structure 2 Nuclear dynamics and thermodynamics 9 Theoretical physics and phenomenology 22 Interdisciplinary Research Nuclear waste management 29 Medical and industrial applications 36 Group «Interactions Fondamentales et nature du Neutrino» (GRIFON) Precise correlation measurements in nuclear beta decay 42 High resolution study of low energy charge exchange collisions with a MOT (magnetooptical trapped) target 44 Towards a new measurement of the neutron Electric Dipole Moment (EDM) 26 Search for neutrinoless double beta decay 48 Activités Techniques et Administratives Service administratif 55 Bureau d études et mécanique 56 Service électronique et microélectronique 59 Service informatique 66 Service instrumentation 69 Documentation 73 Qualité et soutien aux projets 74 Hygiène et sécurité 75 Diffusion du savoir Enseignement 77 Formation par la recherche 78 Formation permanente 79 Valorisation 83 Actions de communication 84 Conférences et rencontres scientifiques 86 Informations générales Personnels permanents 92 Organigramme 93 Chercheurs associés 94 Glossaire 95

3 FOREWORD Le présent rapport d'activité couvre la période Malgré une situation financière tendue, il témoigne, nous l'espérons, du dynamisme des équipes de recherche avec le concours remarquable de l'ensemble des services du Laboratoire. Malgré sa taille relativement modeste, le Laboratoire couvre un large ensemble de thématiques qui va de la recherche fondamentale à la recherche interdisciplinaire à vocation sociétale. Sans être exhaustif, voici un bref résumé de nos activités sur la période considérée: En physique nucléaire, l'équipe 'Dynamique et Thermodynamique' a poursuivi l'analyse des campagnes de mesures avec le détecteur INDRA portant sur l'étude des réactions nucléaires et prépare les tests sur faisceaux des premiers modules de FAZIA, le futur multidétecteur de particules chargées. Le groupe"structure" est engagé dans d'ambitieux programmes expérimentaux à RIKEN, auganiletà ISACportantsurlesnoyauxexotiquesrichesenneutrons.Ilprépareaussiune prochaine expérience à ISOLDE. En theorie et phénoménologie, le groupe a produit d'importants résultats dans les méthodes Monte Carlo quantiques et dans l'étude de l'équation d'état de la matière nucléaire dans les objets stellaires. En ce qui concerne le groupe"interactions Fondamentales et Nature du neutrino", l'expérience nedm au PSI portant sur la mesure du moment électrique dipolaire du neutron est maintenant dans une phasedeprisededonnées.enmêmetemps,legroupepréparelaphaseiidel'expérience.la présentepériodeavul'achèvementetlafinalisationdelaprisededonnéessurnemo3aulsm dédiée à la recherche de l'émission double-beta sans neutrinos. Le groupe est maintenant largement impliqué dans la construction du démonstrateur de SuperNEMO. Les expériences de recherche de courants exotiques dans la décroissance beta menèes au GANIL sont en phase d'analyse. De beaux résultats ont été obtenus en collaboration avec des physiciens atomistes dans le domaine de l'interaction ion-atome et dans l'étude du phénomène de shake-off. Le groupe "Aval du Cycle" a poursuivi l'expérience GUINEVERRE du programme FREYA sur le réacteur sous-critique VENUS-F au SCK. Il prépare en même temps les expériences sur la future ligne NFS à SPIRAL2. L'équipe"Applications médicales et industrielles" mène ses recherches dans le domaine de la hadronthérapie à travers les programmes France-Hadron et Rec-Hadron. Il participe activement au projet ARCHADE et a aussi initié de très fructueuses collaborations avec le monde industriel.

4 En dehors de l'appui aux projets des équipes de recherche, les services du Laboratoire sont engagés dans des développements propres. En particulier, le système d'acquisition FASTER a atteint maintenant un niveau de maturité qui permet son déploiement sur nos expériences et dans de nombreux projets en France et à l'étranger. Notre contribution au projet SPIRAL2 s'est poursuivie et accrue avec notamment une importante contribution des membres de l'atelier au montage de l'accélérateur. Le développement du RFQ pour SPIRAL2 poursuit son cours. Le savoir-faire acquis dans ce domaine va permettre notre participation au développement de la ligne basse énergie de S3 et dans l'équipement du hall expérimental DESIR. Le Laboratoire comporte une forte composante d'enseignants-chercheurs. Ces derniers ont la tâche difficile de mener de front leur activité d'enseignement et de recherche. Leurs nombreuses prises de responsabilité dans les formations et diplômes font du Laboratoire un acteur reconnu au sein de l'université de Caen Basse Normandie et de l'ensicaen. Depuis de nombreuses années, nous sommes engagés dans de multiples actions de vulgarisation auprès des jeunes et du grand public. Cette activité s'est poursuivie à travers diverses manifestations et rencontres. A noter l'accueil d'un nombredeplusenplusgranddestagiairesdetoutniveaudeformation. Certaines aspects primordiaux, non'quantifiables', n'apparaissent pas à la lecture d'un tel rapport. D'abord l'engagement sans failles des personnels dans les projets du Laboratoire et ce, malgré des perspectives de carrière souvent difficiles. Ensuite, l'excellente ambiance de travail dans une atmosphère conviviale et détendue. Enfin, à l'heure de la multiplication des sources de financement, le haut niveau de cohésion, de mutualisation et d'entraide entre les équipes et les services qui font du Laboratoire bien autre chose qu'un simple'hôtel à projets'. C'est un plaisir de remercier l'ensemble des personnes qui ont pris part à l'élaboration de ce rapport. Une mention particulière à Samuel Salvador pour la mise en œuvre de la partie scientifique et à Sandrine Guesnon pour l'important travail de mise en forme de l'ensemble du document. En vous souhaitant une bonne lecture, Dominique Durand Directeur du Laboratoire

5 RESEARCH NUCLEAR PHYSICS Nuclear structure Nuclear dynamics and thermodynamics Theoretical physics and phenomenology 1

6 Nuclear structure N.L. Achouri, F. Delaunay, S. Leblond*, J. Gibelin, F.M. Marqués, N.A. Orr, M. Pârlog, M. Sénoville* Collaboration : D. Durand (LPCC), G. Lehaut (LPCC), M. Colonna (INFN Catania), H. Hamrita (IPN Orsay/CEA Saclay) *PHD students The Nuclear Structure (or Exotiques ) group is active in the investigation of the structure of neutron-rich nuclei using the probes of direct reactions and decay spectroscopy. In the direct reaction studies, the structure of light (A<50) neutron-rich nuclei, including haloes, clustering and correlations and shell structure, is explored using energetic radioactive beams. Two different approaches are employed: (i) at high energies (>100 MeV/nucleon) and close to the dripline nucleon knockout, breakup, inelastic excitation and Coulomb dissociation; and (ii) at low energies (~5 10 MeV/nucleon) and closer to stability nucleon transfer. The high-energy reaction studies, which have been the main focus of the group s reaction studies activities over the course of the last two years, are undertaken at the Radioactive Isotope Beam Factory (RIBF) at RIKEN where beam intensities 3 or 4 orders of magnitude higher than elsewhere, are available for the light near dripline nuclei. At the RIBF experiments are carried out, with radioactive beams delivered by the BigRIPS fragment separator, using the ZDS zero-degree spectrometer coupled to the DALI2 NaI array and, since Spring 2012, the SAMURAI spectrometer plus NEBULA neutron array. One of the goals of the group in the next few years is to upgrade, through a doubling of the number of scintillator walls, the NEBULA array ( NEBULA-Plus ) to enable us to exploit to the maximum the unique beams available at the RIBF and to explore, in particular, multi-neutron decaying systems and the most exotic neutron-rich systems accessible. This project is, at the time of writing, the subject of a grant request EXPAND made to the ANR. Our complementary transfer reaction studies typically neutron addition to the beam via (d,p) in inverse kinematics at lower energies and closer to stability employ at GANIL-SPIRAL1, the TiaRA Si-strip array coupled to the EXOGAM Ge-array and the VAMOS spectrometer. In the near future experiments will be undertaken employing beams, such as 16 C, not available with SPIRAL1, prepared using the LISE3 separator. In the case of our TRIUMF based work, the beams are delivered by the ISAC2 facility (which offers a suite of beams unavailable at SPIRAL1) and the SHARC Si-strip array coupled with the TIGRESS Ge-array is employed for the measurements. Owing to the lack of a suitable spectrometer, zero-degree detection is provided by a thin scintillator plus stopper foil setup developed at LPC. The main priority in the near future at ISAC is the measurement of the d( 28 Mg,pγ) 29 Mg reaction which will complement our earlier work on d( 24,26 Ne,p) 25,27 Ne [1,2] and further help map the transition into the island of inversion around N=20. The second main theme of the group s research is centred on the investigation of structure through β-decay, and in the context of neutron-rich nuclei, the study of β-delayed neutron emission. Presently this activity is focussed on R&D for a new neutron time-of-flight array which has included extensive neutron beam measurements at CEA/DAM-Arpajon. In addition, a proof of principle experiment is under preparation for ISOLDE aiming at a measurement of the β-delayed two-neutron decay of 11 Li. Extensive source testing, in particular in terms of investigating neutron-gamma discrimination techniques, digital signal processing and new scintillators, has also been carried out at LPC. This work has benefited greatly from the untiring support of our colleagues at LPC who have developed the FASTER digital acquisition system. 2

7 Structure at and beyond the neutron dripline: 18,19 B and 21,22 C Collaboration: Tokyo Institute of Technology (Japan), RIKEN (Japan) and the SAMURAI Collaboration The investigation of the light neutron-rich dripline nuclei, including in particular those exhibiting haloes, is a central theme of nuclear structure physics. In the present work a series of measurements, aimed at elucidating the structure of the two heaviest candidate two-neutron halo systems, 19 B and 22 C [3-5], and the associated unbound sub-systems 18 B and 21 C, the level schemes of which are critical to the defining the 17 B-n and 20 C-n interactions for three-body models, have been undertaken. In addition to being of direct importance to halo physics, 18,19 B and 21,22 C are of considerable interest in terms of the evolution of shell-structure far from stability as they span the N=14 and 16 sub-shell closures below doubly-magic 22,24 O. The measurements were accomplished using the SAMURAI spectrometer [6] coupled to the large area neutron array NEBULA [7] and were performed as part of the first phase of SAMURAI experiments following the successful commissioning in Spring The analysis to date has concentrated on the fragment+neutron channels and, in particular, 17 B+n which is known to exhibit a strongly interacting virtual s-wave threshold state [8]. Beyond the intrinsic physics interest noted above, a well defined threshold state provides an ideal means to validate the calibration and analysis procedures. In addition to accessing 18 B via proton removal from 19 C, which should populate almost exclusively s-wave strength, the complementary probe of neutron removal from a 19 B beam has been investigated. Fig. 1 shows the reconstructed 17 B+n invariant mass (or relative energy) spectra for the two reactions which were undertaken at around 240 MeV/nucleon. As may be clearly seen the proton removal populates a very narrow threshold structure, the form of which is consistent with the strongly interacting s-wave virtual state deduced by Spyrou et al. [8]. The neutron removal, however, in addition to the threshold peak, shows clear evidence for the population of a state or states in the region of MeV. The further analysis of these preliminary results is currently underway, including two-proton removal from 20 N, which is expected to populate preferentially d-wave strength in 18 B. The analysis of the data sets for the analogue reactions populating 21 C C( 22 C, 20 C+n), C( 22 N, 20 C+n) and C( 23 O, 20 C+n) are also in progress. The work outlined here forms part of the PhD thesis of S. Leblond who acknowledges the support provided in terms of a 6 month RIKEN Nishina Center International Program Associate fellowship in Fig. 1: Preliminary results for the 17 B+n relative energy spectra obtained for proton and neutron removal reactions at 240 MeV/nucleon. ANIME: a simulation code for NEBULA Collaboration: Tokyo Institute of Technology (Japan) As noted in the overview to our group s activities, our experimental program at RIKEN, which aims to explore the neutron dripline and beyond, relies on the coincident detection of charged fragments using the SAMURAI spectrometer and beam velocity neutrons (E~250 MeV) with the NEBULA multi-element plastic scintillator array. The response of the neutron array is a key element in the analysis of these experiments. There are two existing approaches for the description of this response: 1) GEANT4, that uses intra-nuclear cascade models; 2) MENATE, that describes individually all possible reaction channels on H and C. The former, however, functions very much as a black box which is difficult to modify if discrepancies with the data appear, while the latter was developed for energies well below 100 MeV, and for the specific setup of cylinders as for the DEMON array. 3

8 In the present work, we have opted for a third approach: the development from scratch of a new code ANIME ( Algorithms for Neutron Identification in Modular Experiments ), that treats the individual reaction channels at higher energy in a relatively simple manner, with a more user-friendly geometrical interface. NEBULA consists of multiple planes of vertical plastic scintillator bars, in which neutrons are tracked until they interact with either H or C. The interaction probability for each reaction channel is calculated using the MENATE_R database, which has been extended beyond 100 MeV. The reaction kinematics are treated as quasi-free scattering on n/p/α inside C, taking into account the corresponding separation energies and intrinsic momentum distributions. The outgoing neutron angular distribution is calculated from the phenomenological parameterization of the DEMONS code for the scattering off nucleons [9], and as an energy-dependent exponential in cosθ for the other channels. The recoiling protons are tracked as they move through the array with the same geometrical interface as the neutrons, and the energy deposited by all charged particles is transformed into light and propagated to the two photomultipliers coupled to each bar. We have checked the performance of ANIME with data acquired during the commissioning of SAMURAI+NEBULA using the 7 Li(p,n) reaction. As an example, Fig. 2 shows the light output recorded in NEBULA from mono-energetic neutrons at 250 MeV. The agreement is as good at 200 MeV, as well as for the multiplicity of the number of individual bars hit and the relative angle between them. The latter is essential in order to understand cross-talk in the array that is, the interaction of one neutron at several points within NEBULA that may mimic the detection of several neutrons. ANIME will be employed in constructing a cross-talk filter for the analysis of reaction channels with more than one neutron in the final state, as well as in our planned extension of NEBULA to 4 walls. Fig. 2: Light output, Q (MeVee), for 250 MeV neutrons interacting with the NEBULA array. The data are compared with the results obtained using the ANIME code (dashed line). The contributions of individual reaction channels are shown. R&D for a new time-of-flight neutron array Collaboration: CIEMAT-Madrid (Spain), CEA-DIF Bruyères-le-Châtel Owing to the large Q β values and the low neutron binding energies of the daughter nuclei, the β-decay of very neutronrich nuclei is often followed by the emission of neutrons from unbound states. The detection of such relatively low-energy neutrons (<5 MeV) is therefore crucial to constructing complete decay-schemes. In order to improve the detection performance and, in particular, provide for a multi-neutron detection capability, a new modular time-of-flight array based on discriminating scintillators and coupled to a digital data acquisition system is being developed by our group. Existing neutron arrays based on large plastic scintillator bars, such as the TONNERRE array [10], present limitations. First, owing to the absence of pulse-shape discrimination, the time-of-flight spectra are contaminated by a background arising from the ambient γ-rays and cosmic muons, which renders the identification and measurement of weak neutron transition delicate. Furthermore, two-neutron detection is extremely difficult as multiplicity-two events are dominated by random coincidences involving γ and cosmic rays. In addition, TONNERRE suffers from a limited energy resolution, asymmetric lineshapes and a relatively high neutron energy threshold (~300 kev). The following strategies, as described in earlier reports, have been adopted to overcome these limitations. These include, in particular: Limited scintillator volumes viewed by large-diameter photomultiplier tubes and digital signal processing to lower the threshold and reduce the lineshape asymmetry. Relatively thin detectors at increased distances (>2 m) to improve the energy resolution. To allow for multi-neutron detection by reducing the background with pulse shape discrimination, using liquid scintillators or discriminating solid organic scintillators. A modular array with variable geometry to limit cross-talk and optimise cross-talk rejection schemes. 4

9 The module design that has been adopted is based on a BC501 liquid scintillator cell with a diameter of 20 cm and a depth of 5 cm, viewed by a 13 cm photomultiplier tube through a light guide. The design was characterised in a series of measurements undertaken using monoenergetic neutron beams at the CEA-DIF Bruyères-le-Châtel facility and the FASTER digital acquisition developed here at LPC. Intrinsic efficiencies and cross-talk probabilities were measured at several neutron energies in the range of interest (1 15 MeV) in order to validate the simulations and the kinematical cross-talk filters developed for higher energies [11]. These data are the first measurements of cross-talk probabilities below 14 MeV neutrons energy. Fig. 3: Intrinsic detection efficiency as a function of the neutron energy. The measurements (squares) are compared to simulations (solid and dashed lines) for two different thresholds. An example of the results obtained for the efficiency measurements are shown in Fig. 3. As may be seen, simulations employing the MENATE code [12] and a revised version of MENATE developed for use within GEANT4, predict efficiencies in good agreement with the data. In terms of the cross-talk measurements, the setup employed is illustrated schematically in Fig. 4 whereby neutrons scattered from the unshielded module [A] to the shielded one [B] were measured for different relative positions of the two modules. Fig. 5 displays, for 2 MeV incident neutrons, the time-of-flight measured between the two detectors for events identified as neutrons. Also shown are the results of a MENATE based simulation incorporating only the active volume of each detector. Reasonable agreement was obtained between the experimental and simulated cross-talk probabilities for the full range of incident neutron energies (1 15 MeV) and detector relative positions (θ AB ). The next step that will be undertaken will be to test the efficacy of cross-talk filters on these data, and perform more realistic simulations including the inactive materials such as the detector housing. Fig. 4: Schematic view of the detector configuration used for the cross-talk measurements. 5

10 A proof-of-principle experiment accepted at the ISOLDE facility at CERN is envisaged to be run in the near future. The principal goal of this experiment will be the detection of two β-delayed neutrons in coincidence (from the decay of 11 Li which has the highest presently known two-neutron emission probability) and, for the first time, the measurement of their energies and angular correlations. The coincident detection of two neutrons is currently being tested using a 252 Cf source with some 10 neutron detectors, an LaBr 3 scintillator for the time-of-flight start and elements of the FASTER digital acquisition system. In order to investigate the possible utility of a reduced scintillator volume on the neutron-γ discrimination at low energy and to explore scintillators other than the usual liquids, we have undertaken a series of measurements to characterise small cylindrical (5 cm diameter 5 cm thick) samples of organic scintillators: crystals (p-terphenyl, trans-stilbene), liquid scintillators (BC501A and NE213 for reference purposes, deuterated BC537) and discriminating plastics (EJ , CP197 from CEA/LCAE). Whereas the crystals show a light yield twice as large as that of the usual liquids, their discrimination performance is not significantly better. In particular, the neutron-γ separation obtained at low energy with liquids and the p-terphenyl crystal are similar. The currently available discriminating plastic scintillators cannot compete with the liquids and the crystals in terms of discrimination. The deuterated BC537 liquid scintillator exhibits a smaller light yield and a poorer quality discrimination as compared to BC501A/NE213, and therefore is not a viable alternative. Finally, it is worthwhile noting that reducing the diameter of the liquid scintillator cell from 20 cm (see above) to 5 cm provides for a lowering of a factor of two of the threshold at which neutrons can be unambiguously identified. The work presented here forms part of the PhD thesis of M. Senoville [13]. Fig. 5: Time-of-flight between detectors A and B (see Fig. 2) for events identified as a neutron in both detectors, measured with 2 MeV neutrons (blue spectra) for three different angles (q AB ) and an analysis threshold of 100 kevee. The results of a simulation including only the active volumes are shown by the red spectra. Investigation of the compressions modes in unstable Nickel isotopes Collaboration: GANIL, IPN Orsay, ATOMKI (Hungary), KU Leuven (Belgium), Konan University (Japan), KVI Groningen (Netherlands), MSU (USA), Notre Dame (USA), RNCP (Japan), RIKEN (Japan), USC (Spain) The study of collective excitation modes, such as the Isoscalar Giant Monopole (ISGMR) and Dipole (ISGDR) resonances, has been pursued in stable nuclei over much of the last three decades with the aim of determining the incompressibility (K) of nuclear matter [14]. This fundamental property is of significant importance as it dictates the excitation energies of the compression modes and, in terms of the equation of state, it plays a crucial role in describing nuclear collisions and supernovae resulting from the collapse of very heavy stars. Through extensive experimental and theoretical studies, the incompressibility, K, has been relatively well determined in stable nuclei. The asymmetry term in the expansion of K, however, has been poorly determined, since it requires the investigation of compression modes over a broad isotopic chain. In addition, in exotic nuclei new phenomena are expected to occur, such as pygmy resonances with multipole strengths reflecting the collectivity arising from the neutron or proton-skins relative to the core. With this goal in mind, and following the first successful measurement of the ISGMR and ISGQR in 56 Ni [15], two experiments have been performed at GANIL using secondary beams produced with the LISE3 separator: a search for the ISGDR in 56 Ni and the measurement of the ISGMR and ISGQR in 68 Ni. The three measurements employed the MAYA active target filled either with deuterium or helium (+quencher) gases following the tests described in Ref. [16]. 6

11 The 68 Ni experiment employed a 50 MeV/nucleon beam of intensity 10 4 pps on both deuterium and helium gas targets. Excitation energy spectra deduced for the α( 68 Ni,α) 68 Ni* reaction are shown in Fig. 6. The GMR was determined to lie at 21.7±1.9 MeV and evidence for a soft monopole mode, predicted but never observed, was found at 13.2±0.5 MeV. The corresponding angular distributions, analysed using Distorted Wave Born Approximation with Random Phase Approximation transition densities, indicate that the GMR exhausts a large fraction of the energy-weighted sum rule and that neutrons mainly contribute to the soft monopole mode. Both experiments using deuterium and helium gas provided coherent results providing added confidence in or conclusions and demonstrating the relevance of alpha inelastic scattering in inverse kinematics in order to probe both the GMR and soft modes in neutron-rich nuclei. This work formed part of the PhD of M. Vandebrouck [17] and a manuscript has been submitted for publication in Physical Review Letters. The analysis of the inelastic scattering of 56 Ni on helium is currently ongoing in order to locate the ISGDR of 56 Ni. The excitation energy spectrum for 56 Ni has been reconstructed and simulations have been performed to estimate the angular acceptances and the efficiency of the reconstruction. In order to refine our understanding of the setup, the angular distribution for the elastic scattering of 56 Ni on helium has been derived and a preliminary result is shown in Fig. 7. We note that the minimum in the cross-section is a slightly shifted compared to DWBA predictions and the origins of this discrepancy are now being investigated. This work forms part of the PhD of S. Bagchi (University of Groningen). Fig. 6: Excitation energy spectrum for the α( 68 Ni,α) 68 Ni* reaction for a) all measured angles and b) for θ CM =5.5. In both cases data are fitted with Lorentzian distributions centred at 13.2 (red), 15.7 (blue) and 21.7 MeV (red lines) corresponding to the soft GMR, GQR and GMR, respectively. The broad peak above 25 MeV (dotted line) corresponds to several additional multi-polarities such as L = 1,3. Finally, we note that the 56 Ni + He data is also being exploited in order to explore the cluster nature of 56 Ni [18]. Fig. 7: Differential angular distribution for the elastic scattering of 56 Ni on helium. The dashed line is the result of a DWBA calculation. 7

12 References [1] W.N. Catford et al., Phys. Rev. Lett. 104 (2010) [2] S. Brown et al., Phys. Rev. C85 (2012) (R) [3] K. Tanaka et al., Phys. Rev. Lett. 104 (2010) [4] N. Kobayashi et al., Phys. Rev. C83 (2012) [5] L. Gaudefroy et al., Phys. Rev. Lett. 109 (2012) [6] T. Kobayashi et al., Nucl. Instr. Meth. B317 (2013) 294. [7] Y. Kondo et al., RIKEN Accel. Prog. Rep. 45 (2012) 131; [8] A. Spyrou et al., Phys. Lett. B 683 (2010) 129. [9] W.C. Sailor et al., Nucl. Inst. Meth. 277 (1989) 599. [10] A. Buta et al., Nucl. Instr. and Meth. A455 (2000) 412. Publications The N = 16 spherical shell closure in 24 O Tshoo K., Satou Y., Bhang H., Choi S., Nakamura T. et al. Physical Review Letters 109 (2012) Search for Superscreening effect in Superconductor Ujic P., de Oliveira Santos F., Lewitowicz M., Achouri N.L., Assié M. et al. Physical Review Letters 110 (2013) Structure of unbound neutron-rich 9 He studied using singleneutron transfer Al Kalanee T., Gibelin J., Roussel-Chomaz P., Keeley N., Beaumel D. et al. Physical Review C 88 (2013) Limited Asymmetry Dependence of Correlations from Single Nucleon Transfer Flavigny F., Gillibert A., Nalpas L., Obertelli A., Keeley N. et al. Physical Review Letters 110 (2013) Core excitations and narrow states beyond the proton dripline: The exotic nucleus 21 Al Timofeyuk N.K., Fernández-Domínguez B., Descouvemont P., Catford W.N., Delaunay F. et al. Physical Review C 86 (2012) Well-developed deformation in 42 Si Takeuchi S., Matsushita M., Aoi N., Doornenbal P., Li K. et al. Physical Review Letters 109 (2012) Comment on "First Observation of Ground State Dineutron Decay: 16 Be" Marqués F.M., Orr N.A., Achouri N.L., Delaunay F., Gibelin J. Physical Review Letters 109 (2012) Spectroscopy of 18 Na: Bridging the two-proton radioactivity of 19 Mg Assié M., Santos F. D. O., Davinson T., De Grancey F., Achouri N.L. et al. Physics Letters B 712 (2012) [11] F. M. Marqués et al., Nucl. Inst. Meth. A450 (2000) 109. [12] P. Désesquelles et al., Nucl. Inst. Meth. A307 (1991) 366. [13] M. Senoville, Développement d un nouveau multi-détecteur de neutron, Thèse, Université de Caen Basse-Normandie (2013) [14] M. N. Harakeh and A. van der Woude, Giant Resonances: Fundamental High-Frequency Modes of Nuclear Excitation, Oxford University Press, Oxford, [15] C. Monrozeau et al., Phys. Rev. Lett. 100 (2008) [16] J. Pancin et al. JINST 7 (2012) [17] M. Vandebrouck, Première mesure des résonances géantes isoscalaires dans un noyau exotique riche en neutrons : le 68 Ni avec la cible active Maya, Thèse, Université Paris Sud Paris XI (2013) [18] H. Akimune et al. J. Phys.: Conf. Series 436 (2013) Low-lying neutron f p-shell intruder states in 27 Ne Brown S.M., Catford W.N., Thomas J.S., Fernandez-Dominguez B., Orr N.A. et al. Physical Review C 85 (2012) Resonances in 19 Ne with relevance to the astrophysically important 18 F(p,(\alpha)) 15 O reaction Mountford D.J., Murphy A. S. J., Achouri N.L., Angulo C., Brown J.R. et al. Physical Review C 85 (2012) Direct mass measurements of 19 B, 22 C, 29 F, 31 Ne, 34 Na and other light exotic nuclei Gaudefroy L., Mittig W., Orr N.A., Varet S., Chartier M. et al. Physical Review Letters 109 (2012) Electrostatic mask for active targets Pancin J., Gibelin J., Goth M., Gangnant P., Libin J.F. et al. Journal of Instrumentation 7 (2012) P01006 In-beam spectroscopic studies of 44 S nucleus Caceres L., Sohler D., Grévy S., Sorlin O., Dombradi Z. et al. Physical Review C 85 (2012) One-proton breakup of 24 Si and the 23 Al( p, γ ) 24 Si reaction in type I x-ray bursts Banu A., Carstoiu F., Achouri N.L., Catford W.N., Chartier M. et al. Physical Review C 86 (2012) Resonances in 11 C observed in the 4 He( 7 Be,α) 7 Be and 4 He( 7 Be,p) 10 B reactions Freer M., Achouri N.L., Angulo C., Ashwood N.I., Bardayan D.W. et al. Physical Review C 85 (2012) One and two neutron removal reactions from the most neutronrich carbon isotopes Kobayashi N., Nakamura T., Tostevin J.A., Kondo Y., Aoi N. et al. Physical Review C 86 (2012) β-delayed neutron emission studies Gómez-Hornillos M.B., Rissanen J., Taín J.L., Algora A., Kratz K.L. et al. Hyperfine Interactions 223 (2012)

13 Nuclear dynamics and thermodynamics (INDRA-FAZIA collaborations) L. Augey *1, R. Bougault, M. Kabtoul *2, E. Legouée *3, N. Le Neindre, O. Lopez, M. Parlog, E. Vient Collaboration : D. Durand (LPCC), G. Lehaut (LPCC), M. Colonna (INFN Catania), H. Hamrita (IPN Orsay/CEA Saclay) *PHD students 1 since october 2013, 2 until july 2013, 3 until october 2013 The determination of the nuclear equation of state (EOS) is one of the key issue concerning Nuclear Physics. The characterisation of its dependence in term of density, temperature and isospin are mandatory to describe accurately as well heavy-ion collisions and properties of neutron stars. The EOS can be seen as the macroscopic consequence of the properties concerning the underlying nucleon-nucleon (NN) interaction in nuclear matter. Studying the EOS is then directly related to the study of NN interaction, namely its density dependence via many-body correlations, and its isovector properties via the symmetry term of EOS. In order to probe these features, we currently use heavy ion induced reactions in the Fermi energy domain and perform exclusive measurements using the 4π array INDRA. This allows to access to the dynamical (transport properties) and the thermodynamical features of hot and compressed nuclear matter. INDRA is a international collaboration grouping 5 institutes : GANIL Caen, IPN Orsay, LPC Caen, Laval University (Québec) and INFN Napoli (Italy). INDRA is in operation since 1993 and 8 large data takings (campaigns) have been performed at GANIL (stable beams + SPIRAL1/CIME beams) in France and GSI in Germany. The collaboration is composed by 18 physicists + 3 PHD + 1 post-doc (2013) and still continue to maintain INDRA in order to be ready for SPIRAL2 and GANIL beams in a near future. The collaboration is also deeply involved since 10 years on the next-generation 4π array; it is the FAZIA project. Taking advantage from the experience concerning 4π arrays, we are currently developping a new prototype of 4π detector and are in the present time in phase 2 of the FAZIA program. This phase consists in building a fully operational demonstrator, composed of 12 blocks made of 16 identification telescopes Si-Si-CsI with their embedded digital electronics. Several research topics have been developped in the laboratory concerning the study of the dynamical and thermodynamical properties of nuclei with INDRA as well as instrumental developments for FAZIA. In section 1, we present an analysis concerning the study of transport properties in nuclear matter and the determination of some fundamental in-medium quantities such as the nucleon-nucleon mean free path and cross section. In section 2, we address temperature and excitation energy measurements from an experimental point of view; indeed, these observables are at the centre of any thermodynamical study and therefore also for the accurate determination of the nuclear EOS. In section 3, we present a recent experimental work concerning the evaluation of the symmetry energy term on the nuclear EOS. In section 4, we show an experimental program aiming at the evaluation of the best Pulse Shape Analysis which can be achieved with highly homogeneous silicon detectors for the FAZIA project. At last, Section 5 is devoted to the modelization of current signals produced in Silicon detectors, in order to optimize the Pulse Shape Analysis for FAZIA. 9

14 In-medium effects in nuclear matter in the Fermi energy range collaboration with D. Durand and G. Lehaut Transport properties are critical in the description of the supernova collapse and the formation of a neutron star [1]. They are also one of the fundamental ingredients for microscopic models [2] and contribute for the determination of the equation of state via the underlying in-medium properties of the nuclear interaction. Transport properties of nuclear matter are probed with heavy-ion induced collisions (HIC) by looking at dissipation phenomena in term of energy and isospin diffusion. In the Fermi energy domain, transport features should exhibit the interplay between Mean-Field (nucleus) and individual (nucleons) effects, especially when looking at the energy dissipation reached in central collisions where the overlap between the two incoming partners becomes maximal [3]. Fig. 1 displays the mean isotropy ratio R E [3] as a function of the incident energy for 6 different symmetric systems ; this compilation consists in 40 experimental determination and illustrates the large body of data available with INDRA. In this study, we have selected central collisions by using the total charge multiplicity as detailed in [3,4]. The data are compared to the expected R E values for full transparency (blue curve) and full stopping (red curve). In a simple picture for the central collisions consisting in 2 separate Fermi spheres with relative momentum given by: P rel =αp rel0, where P rel0 is the relative momentum according to the incident (relative) energy between the 2 incoming nuclei of the reaction. The above-mentioned situations correspond then to α=0 for full stopping and α=1 for full transparency. Fig. 1: Mean isotropy ratio R E for protons as a function of incident energy in central collisions. The symbols correspond to different symmetric systems. The blue and red curves are the theoretical predictions for full transparency (blue) and full stopping (red) respectively. From [4]. From the isotropy ratio, we evaluate the stopping ratio reached in such central collisions. This latter is computed as the reduced distance d=(r E -R E (α=0))/(r E (α=1)-r E (a=0)) between the 2 extreme scenarii. It is instructive to note that the quantity d 2/3 can be nicely scaled as a function of the characteristic size of the system A 1/3 as shown in [4]. This scaling suggests that the stopping ratio, measured by d 2/3, is related to the size of the system; in a Glauber scenario, the stopping is indeed related to the in-medium NN cross section σ NN and the average distance crossed by the scattered nucleons. Therefore, one can use d as an estimate of the NN mean free path λ NN or the associated cross section σ NN related by the simple formula obtained from kinetic theory: λ NN 1/ρσ NN. This is done in Fig. 2 where we plot the estimated λ NN from the simple formula: λ NN R/d 2/3, with R=r 0 A 1/3 and r 0 =1.25 fm. 10

15 Fig. 2: In-medium mean free path λ NN as a function of the incident energy. From [4]. We obtain the incident energy dependence concerning the in-medium mean free path λ NN. We see that λ NN is maximal around E inc /A=40 MeV with a typical value λ NN =8-9 fm. The decrease observed at lower incident energy is not commented here since we believe that the reduced distance d is not valid when the Mean-Field dissipation is present; in this case, one has to evaluate properly the dissipation reached in central collisions, due to the 1-body dissipation term (friction). At variance, for incident energies larger than 40 A MeV, we consider the sudden approximation used as a reference for full transparency to be valid. Within this energy range, we observe a clear decrease for λ NN, from 8-9 fm at 40 A MeV toward a saturation around 4-5 fm at 100 A MeV. This latter result is in full agreement with both theoretical and experimental values around and above 100 A MeV [5,6,7]. To evaluate the magnitude of the in-medium effects, we then compare the NN cross section σ NN obtained from the λ NN values displayed by Fig. 2. We take into account the important effect due to the quantum nature of the nucleons (Pauli exclusion principle) as recommended in [8]. We obtain the reduction factors shown by Fig. 3. Fig. 3: In-medium reduction factor for the in-medium NN cross section. The curves correspond to different parametrizations used in recent theoretical descriptions. From [4]. The reduction factors are found to be quite large, ranging from 20% to 40%, indicating that in this incident energy range ( A MeV), the in-medium effects are far from being negligible and thus should be taken properly into account in any microscopic descriptions such as transport models. The best agreement is found with the phenomenological prescription proposed in [9] by Danielewicz [10]. It is also worthwhile to note that almost all prescriptions seem to converge toward the same value above 100 A MeV. 11

16 Thermometry and calorimetry studies Isospin transport during collisions around Fermi energy A previous study [11] allowed us to develop an experimental method giving the probability for a particle to be emitted by a hot Quasi-Projectile (QP), during a nuclear reaction around the Fermi energy. We decided to use this information to determine the probability for a particle of not being evaporated. Knowing this last probability, we can then characterize an eventual pre-equilibrium component or a neck emission, when this contribution exists. For two main reasons, the only way to make this work, is to observe symmetric or quasi-symmetric collisions. Firstly, there are principally binary collisions for these systems. Secondly, for obvious reasons of symmetry, the Quasi-Target (QT) should have the same physical behavior than the Quasi-Projectile, consequently the same probability of evaporating a given particle. This study has been done for different nuclear systems studied by the INDRA collaboration in the framework of a PhD [12]. We have wanted also to confirm experimentally the validity of hypotheses done to determine these different contributions and to show the effective quality of this method of isolation. The tool to attain this goal is to study the isospin diffusion and the isospin layout in the velocity space, during a nuclear reaction, for the first time at two dimensions. The INDRA Collaboration has studied during its fifth campaign the quasi-symmetric system Xe+Sn at 32 A MeV using several isotopes of these both nuclei. It has thus used four isotopic combinations: Neutron Rich System [NN] Proton Rich System [PP] Mixed System with Proton rich projectile and Neutron rich target [PN] Mixed System with Neutron rich projectile and Proton rich target [NP] To study the way in which the densities of neutrons and protons during a nuclear reaction are distributed in the velocity space (defined in the c.m.), respectively along the parallel and the perpendicular axis to the beam, we have defined Rami's ratios [13] for different elementary squares in this space. These ratios are determined from ratios of ratios of different isotopes as proton/deuteron obtained for the different considered isotopic configurations of collisions [13]. The ratio is normalized to 1 for a neutron rich zone and to -1 for proton rich zone. The ratio is equal to 0 if there is an equilibration of isospin. Fig. 4: Rami's ratios (for p/t) as a function of the perpendicular and parallel velocities in the c.m. frame for the system 124 Xe Sn at 32 A MeV with a specific selection of QP. For a specific selection of Quasi-Projectiles (angles defined in the laboratory frame between 4 and 6 and velocity between 0.1c and 0.12c, in the c.m. frame), we present, in Fig. 4, for the system 124 Xe+ 124 Sn at 32 A MeV, the Rami's ratios (obtained with protons/tritons) as a function of the perpendicular and parallel velocities in the c.m. frame. We have kept events with a QP detected on one side of the beam. The perpendicular component of particles is negative if the particle is at the opposite side of the QP. 12

17 Fig. 5: Bidimensional maps of experimental probabilities for a proton as a function of the perpendicular and parallel velocities in the c.m. frame for the system 124 Xe Sn at 32 A MeV with a specific selection of QP. The mean velocity of the QP is indicated in Fig. 4 as well as the limit between the front and the back in the QP frame. For comparison, we present, for the protons, in Fig. 5 different maps of probabilities in the velocity space for the same different selections of events than for Fig. 4. We observe a remarkable qualitative agreement between the two methods of isolation of the different contributions produced during a deep inelastic reaction around Fermi energy. For example, the blue parts of Fig. 4, corresponding to the QP contribution (important memory of the initial isospin), are completely compatible with the map of probability of being evaporated by the QP. We have the same trend for the other respective contributions. For the studied very peripheral collisions, there is not isospin balance between the QP and the QC. Only the pre-equilibrium component around the c.m velocity, is equilibrated of this point of view as well as the pre-equilibrium, that we observe on the Coulombian circles. It seems also that there is a fast process of fragmentation, keeping an important memory of initial isospin of the nuclei in collision. Indeed, we find a contribution coming from the target at the front of the emission sphere of the QP (in red in Fig. 4). We have therefore seen that the use of nuclear systems with an important gradient of isospins between the two partners, can be a very interesting tool to use with a 4π setup, to really understand the mechanism and the sequence of a nuclear reaction in the velocity space. By using specifically the Rami's ratio in the velocity space, we have shown the great interest of this representation to study the isospin transport during nuclear reaction and we have moreover confirmed experimentally the validity of our method of probability determination. 13

18 New experimental approaches of the classical caloric curve to study the disintegration of hot nuclei The richness of the set of data, collected by the INDRA collaboration during the last twenty years, enabled us to build a set of caloric curves for nuclei of various sizes, by using, for the first time, a single experimental set-up and a single experimental protocol. The experimental difficulties met usually to measure the temperature and the excitation energy of hot nuclei created by nuclear reactions, have brought us to approach the calorimetry by a new method and to perform in a different way the usual thermometry of such nuclei. We will therefore present the different caloric curves thus obtained in Fig. 6, for Quasi-Projectiles produced by symmetric or quasi symmetric reactions at different incident energies (systems Xe+Sn, Ni+Ni, Ar+KCl ) [12]. For all systems, at all incident energies, a change of behavior is observed, a clear break of slope corresponding to a change of the mode of de-excitation of the hot nuclei. QP QP Fig. 6: Experimental caloric curves obtained for different systems and incident energies.. A certain number of theoretical calculations showed that hot nuclei support an increase of temperature until a maximal temperature, called limiting temperature T lim, beyond which the nucleus may fragment [14-16]. This disintegration of the hot nuclei is due to Coulomb instabilities. This phenomenon is observed in the framework of our study as we can seen it in Fig. 7 for the system Xe+Sn. Indeed, there is a total agreement between the apparition of a break of slope in the caloric curve and the reach of this limiting temperature. Fig. 7: On the left, experimental caloric curves for the system Xe+Sn at different incident energies. On the right, evolution of measured temperatures as a function of QP mass for the same systems and energies (the theoretical curves came from the references [15,16]). We thus observe clearly a transition from a nuclear Fermi gas to another state which might be a gas of particles and fragments. 14

19 Quantifying EOS symmetry energy with Xe+Sn reactions collaboration with M. Colonna This contribution is part of M. Kabtoul thesis and is related to an accepted article in the European Physics Journal [17]. The density (ρ) dependence of the symmetry term of the Nuclear Equation of State can be parameterized as where ρ 0 is the nuclear saturation density. The first term is related to Pauli correlations; the second term is the potential part. The value of the γ exponent is linked to the asy-stiffness (γ 1) or asy-softness (γ <1) of the potential part. The value of γ is presently unknown [18]. 32 A MeV 124,136 Xe+ 112,124 Sn reactions were studied with the INDRA multidetector. Only products detected in the forward centre of mass hemisphere are considered. Observables were measured as a function of an impact parameter, considered as a dissipation scale. The scale is given by the total transverse energy of the light charged particles (Z=1 and 2) detected in the forward c.m hemisphere ( ). Low (large) transverse energies correspond to peripheral 12 (central) collisions. Isospin equilibration Isospin, (N-Z)/A, transport tends to equilibrate the isospin content between the projectile and the target. This has been studied as a function of the impact parameter using the isospin transport ratio [19] : the index H refers to the n-rich system ( 136 Xe+ 124 Sn) and L to the n-poor system ( 124 Xe+ 112 Sn), M to the mixed reactions 136 Xe+ 112 Sn and 124 Xe+ 124 Sn whose total N/Z are the same. The triton multiplicity has been used as isospin observable (x). The evolution of R t with impact parameter is displayed in Fig. 8. We recover the previous section result, i.e. for very peripheral collision no isospin equilibration is observed where as isospin equilibration is reached above a transverse LCP energy of about 100 MeV which corresponds to impact parameters below 6 fm. Fig. 8: Isospin transport ratio calculated for the measured multiplicity of tritons, for the four Xe+Sn systems at 32A MeV, as a function of dissipation. Symmetry energy from isospin diffusion The chosen isospin sensitive variable is the fragment (Z>2) multiplicity difference between the 136 Xe+ 112 Sn and 124 Xe+ 124 Sn systems. It is presented as a function of the impact parameter scale in figure 8. Only products detected in the forward c.m. hemisphere are considered and quasi-fusion events are removed [20]. The measured fragment multiplicity is then the sum of fragments coming from quasi-projectile, QP, de-excitation and from mid-rapidity. 15

20 For >100 MeV isospin equilibrium is reached as demonstrated above. Thus the QP de-excitation properties are the 12 same for the two systems, in particular the multiplicity of emitted fragments. Therefore the fragment multiplicity difference reduces to the difference between mid-rapidity multiplicities. Assuming that these multiplicities are not modified by the deexcitation stage, the measured fragment multiplicity difference can be directly compared to transport model predictions for primary fragments without any after burner hypothesis. This avoids resorting to a de-excitation code. Stochastic Mean Field (SMF) [21] calculations were performed using two different parameterizations of the symmetry energy (γ=1 and γ 0.5). If we now compare data and simulated values (Figure 9), it appears that the asy-soft (γ 0.5) case does not follow the experimental trend, whereas the asy-stiff calculation well matches the data for b<6 fm ( >100 MeV). For 12 more peripheral collisions, the comparison does not hold because isospin equilibrium is not reached, thus simulations and data diverge. Fig. 9: Difference of the fragment multiplicities for the systems, 136 Xe+ 112 Sn and 124 Xe+ 124 Sn, versus the transverse energy of light charged particles. Close points show the experimental data. Squares are related to SMF calculations using two parametrizations of the symmetry energy. (asysoft with γ=0.5 and asystiff with γ=1) Pulse Shape Analysis for the FAZIA project Concerning the FAZIA project, the group was mainly involved in the capability and improvement of the so-called Pulse Shape Analysis (PSA) for identification of stopped particles in one single silicon detector. During last period, two main achievements were obtained. Comparison of rear and front side injection for PSA identification For the same FAZIA telescope (Si 300 µm-si 500 µm-csi) and electronic chain we recorded data, taken in the same beam and target configuration, consisting of particles produced in heavy ion collisions at intermediate energy. The experiment was performed in two steps. The first one with particles entering by the low electric field side (rear side injection) in both silicons while in the second step they encountered first the high electric field (front side injection). This was simply obtained by turning both silicon detectors by 180. The silicon detectors fulfilled the last FAZIA specifications obtained during a few years of R&D in terms of resistivity homogeneity over the whole surface (20 x 20 mm²), adequate crystal cutting along the main axis to avoid channelling and bespoke electronic and digitization chains. In these conditions we were able to fairly compare both configurations to determine the best solution in term of particle identification both in the usual E-E (particles punching through the first silicon Si1 and stopped in the second Si2) and PSA (for nuclei stopped either in Si1 or Si2) methods. E-E identification technique It has been established that for the standard E-E technique no significant variations of the identification capability between both configurations have been observed. A very good charge separation for all incident particles as well as an equal impressive isotopic discrimination up to Z=23 have been obtained with the same good quality criteria [22]. 16

21 Pulse shape identification technique Fig. 10: PSA technique: Energy vs rise-time of the charge signal for particles stopped in the first Silicon (Si1). Particles punching through the detector have been removed. From [22]. For the front side injection configuration, the correlation between the energy and the maximum of the current signal (I max ) does not give any visible identification. All elements merge together in a very compact cloud, corresponding to a strong correlation between the energy and the maximum current. Thus the maximum amplitude of the current signal is not a good PSA variable when the fragments enter through the high electric field side. Regarding the Energy vs Charge rise-time" correlations shown in Fig. 10, we obtain in both cases identification maps, although, the shape of the correlation is very different in the two cases. For the front side injection, the charge rise-time continuously decreases with decreasing energy for ions of any Z value. On the contrary, for rear side injection we observe, for a given Z and starting from high kinetic energies, a rise-and-fall trend of the rise-time. For slow ions, this rise-and-fall produces a ridge where all Z values merge together, whatever the particle is. In both cases a no-identification zone is visible for each line at low energy, defining a Z- dependent identification threshold. These thresholds will be determined more precisely in the following. Particle identification thresholds for rear and front configuration At first sight, the rear side injection method may seem more efficient, since it enlarges the ridge range. We need a quantitative way to estimate the PSA identification thresholds. Therefore we apply the Figure of Merit" (FoM) protocol for adjacent peaks in the particle identification spectra. The FoM is defined as: whereµ 1 andµ 2 are the centroids,σ 1 andσ 2 the standard deviations of two Gaussians fitted to adjacent peaks. A value of FoM=0.7 was conventionally chosen in order to extract a low energy threshold above which we realize a good identification. In the case of two Gaussians, isolated, of equal intensity peaks, corresponds to a ratio peak/valley=2 and a correct identification of 95% of the events (as an example, FoM=1 corresponds to 99%). The quantitative FoM method was applied to both matrices of Fig. 10 in order to judge the identification quality for both configurations. The FoM=0.7 identification limit criterion was again adopted. Fig. 11: Thresholds expressed in term of range in Silicon material for Z identification with E(300µm)-E technique (black thick line) and with PSA technique (energy vs charge rise-time: red points are for rear side injection and blue points for front side injection. 17

22 The identification thresholds are summarized in Fig. 11 in terms of the range in Silicon, where a spectacular improvement on the identification energy threshold for the rear side injection technique is observed (red line and full symbols). For the front side injection case the range for identification varies from a minimum of 170µm to about 250 µm, whereas in the rear side injection case the minimum range presents a continuous increase, from 30 to 150µm. Under-depleted silicon detectors Fig. 12: Energy versus Current maximum correlations at different bias voltages.. Fig. 13: (Colour on line) Charge identification thresholds estimated from visual inspection of Energy versus Current maximum correlations (empty squares). Thresholds for the E- E techniques are also shown as filled triangles for 300 µm silicon thickness. In a recent test, we have explored the identification capabilities of reverse mounted partially depleted detectors. In such a configuration, the fragments enter the detector through an undepleted region, where the electric field is nominally zero. In the following we will focus on PSA via the Energy vs Current Maximum" method, since for partially depleted detectors it shows the most promising results. In Fig. 12, Energy vs Current Maximum" correlations are shown for different bias voltages applied to a 500 µm thick Si2 stage. The full depletion voltage is 290 V. An improvement of isotopic separation, above the identification energy thresholds, with decreasing bias voltage can be clearly spotted in the figure. However the better mass resolution capability comes at the price of higher identification energy thresholds. Visual inspection of Fig. 12 permits to evaluate the identification energy thresholds for different elements, reported as a function of Z in Fig. 13. From Fig. 13, it is apparent that at 105 V and 130 V bias voltages the energy thresholds for charge identification are slightly lower than those for mass identification. We would like to stress that the detector under test did not allow isotopic identification via PSA when biased at full depletion voltage. In fact its doping uniformity is only about 6%, while previous tests performed by the Collaboration showed that a doping uniformity of about 1% FWHM or less is needed for isotopic identification at depletion bias voltage. On the other hand, when not fully depleted, PSA of detector signals allowed for both charge and mass separation of fragments, though charge identification energy thresholds were higher than at full depletion. Under-biasing the first stage of a E-E telescope, one could still lower the energy thresholds for PSA isotopic identification, palliating thus an eventually poor doping uniformity. 18

23 Description of current signals in Silicon detectors collaboration with H. Hamrita The high frequency digitization of the current signals induced by heavy ions in highly resistivity-homogeneous neutron transmutation doped (n-td) silicon detectors put in evidence the dependence of their shape on the type and energy of the incident particle (Fig. 14) and related it to the characteristic local energy loss de/dx. Fig 14: Mean current signals induced by ions of about 100 MeV impinging on the rear side of a n-td silicon detector. This observation was recently interpreted in terms of charge carrier collection in a dielectric image. Our simple formalism developed in collaboration with IPN Orsay supposes that electrons (e) and holes (h), created along the track of the ionizing particle, are living for a while as exciton-like couples oriented by the electric field reigning in the detector. Multiplied by their volume concentration, the electric moments of these dipoles lead to a supplementary dielectric bulk polarization described by an enhanced relative permittivity ε r >ε r (ε r =11.7 for silicon) implying a local distortion of the electric field [23]. In a cylindrical geometry, ε r is connected to the instantaneous linear density of carriers N(x,t), initially given by N 0 (x)=(1/w) de/dx, (w=3.62 ev being the energy per e h pair creation): 1 kn,, 1 The dissociation of the charge carrier couples is supposed to take place with a constant probability λ per time unit: dn,, dt 2 as long as the e-h pair linear density overpass a threshold, of low value N th, the remainder being allowed to break down without any delay. All the separated carriers drift towards the appropriate electrode by inducing, in accord with the Shockley- Ramo s theorem, the current signal characteristic to each particle and eventually allowing its identification. A fit procedure, based on the above equations and three fit parameters:λ, k and N th, was used to get the best description of the individual shape of the mean signal induced by several ions of known energy see e.g. Fig

24 Fig. 15: Comparison between the simulated signals (dashed curves) and the mean experimental ones (solid curves) for two ions [24]. Fig. 16 (left) shows the dependence of the fit parameter k on the linear initial carrier density <N 0 > averaged over the particle range l, while in Fig. 16 (right) one may see the relative variation of the related dielectric susceptibilityχ=ε r -1: 0 (3) which connects the dielectric polarization vector to the electric field strength. Fig. 16: Parameter k values (symbols) vs <N 0 >. The curves correspond to the fit with: the derivative of a Heaviside function (as suggested by the integral ʃ(k-k 1 )d<n 0 > in the inset) plus a constant term k nm (solid curve) or the ratio of a monomial and a quadratic polynomial raised to a real power (dashed curve) (left). The relative increase of the dielectric susceptibility at t=0 (symbols) averaged over the range of the ion vs <N 0 >. The curves correspond to the mentioned functions. The straight line simply assumes k=k 1 (right). The dielectric susceptibility is nearly doubling around 250 pairs/nm, speaking about a huge dielectric polarization eventually locally induced for a few nanoseconds [23]. The simulation shown in Fig. 15 is very sensitive to the parameter λ and the associated time constantτ=1/λ is presented in a synthetic manner in Fig. 17. By means of such a simple model and the evidenced connection of its main parameters to the stopping powers and the electric field, we count to develop a global procedure of heavy ion identification in silicon detectors. Fig. 17: The dissociation time constant versus the ratio of <N 0 > and the averaged initial electric field, disturbed <Fin> (open symbols, dashed line) or not <F> (solid symbols, solid line). 20

25 References [11] E. Vient, Mémoire HDR, Université de Caen -Basse Normandie (2006), [12] E.Legouée, Ph.D Thesis, Université de Caen-Basse Normandie [1] J.M. Lattimer and M. Prakash, The Physics of Neutron Stars, Science (2013), 304, 536 (2004) [13] F. Rami, Phys. Rev. Letters 84, 6 (2000) [2] C. Fuchs and H.H. Wolter, Eur. Phys. J. A 30, 5-21 (2006) and [14] S. Levit, Nucl. Phys. A 437, (1985) refs. therein [15] Y. Zhang, Phys. Rev. C 54, 1137 (1996) [3] G. Lehaut et al. (INDRA collaboration), Phys. Rev. Lett. 104, [16] L. Zhang, Phys. Rev. C 59, 3292 (1999) (2010) [17] R. Bougault et al., Eur. Phys. J. A, Special topical issue on [4] O. Lopez, INPC proceedings, EPJ web of conferences, (2013) Symmetry energy (2014) [5] A. Rios and V. Somà, Phys. Rev. Lett. 108, (2012) [18] M.B. Tsang et al., Phys. Rev. C 86, (2012) [6] P.U. Renberg, D.F. Measday, M. Pepin, P. Schwaller, B. Favier, [19] J. Rizzo et al. Nucl. Phys. A 806, 79 (2008) and C. Richard-Serre, Nucl. [20] M. Kabtoul, PHD Thesis, University of Caen (2013) Phys. A 183, (1972) [21] M. Colonna et al., Nucl. Phys. A 742, 337 (2004) [7] A. Nadasen et al., Phys. Rev. C 23, (1981) [22] N. Le Neindre et al., Nucl. Instr. Meth. A 701, 145 (2013) [8] K. Kikuchi and M. Kawai, Nuclear matter and Nuclear Collisions, [23] M. Parlog et al. (FAZIA collaboration), Nucl. Instr. and Meth. in Ed. North Holland, New York (1968) Phys. Res. A 613 (2010) 290 [9] D. Coupland et al., Phys. Rev. C 84, (2011) [24] H. Hamrita et al. (FAZIA collaboration), Nucl. Instr. and Meth. [10] P. Danielewicz, Acta. Phys. Pol. B 33, 45 (2002) in Phys. Res. A 642 (2011) 59 Publications Constrained caloric curves and phase transition for hot nuclei Borderie B., Piantelli S., Rivet M.F., Raduta A. R., Ademard G. et al. Physics Letters B 723, 1-3 (2013) Nuclear multifragmentation time-scale and fluctuations of largest fragment size Gruyer D., Frankland J.D., Botet R., Ploszajczak M., Bonnet E. et al. Physical Review Letters 110, 17 (2013) Comparison of charged particle identification using pulse shape discrimination and ΔE E methods between front and rear side injection in silicon detectors Le Neindre N., Bougault R., Barlini S., Bonnet E., Borderie B. et al. NIM A 701 (2013) Isospin transport in 84 Kr + 112,124 Sn collisions at Fermi energies Barlini S., Piantelli S., Casini G., Maurenzig P.R., Olmi A. et al. Physical Review C 87 (2013) Effects of irradiation of energetic heavy ions on digital pulse shape analysis with silicon detectors Barlini S., Carboni S., Bardelli L., Le Neindre N., Bini M. et al. NIM A 707 (2013) N and Z odd-even staggering in Kr + Sn collisions at Fermi energies Piantelli S., Casini G., Maurenzig P.R., Olmi A., Barlini S. et al. Physical Review C 88 (2013) New isospin effects in central heavy-ion collisions at Fermi energies Gagnon-Moisan F., Galichet E., Rivet M.-F., Borderie B., Colonna M. et al. Physical Review C 86 (2012) A single-chip telescope for heavy-ion identification Pasquali G., Barlini S., Bardelli L., Carboni S., Le Neindre N. et al. European Physical Journal A 48 (2012) 158 Correlations between emission timescale of fragments and isospin dynamics in 124 Sn+ 64 Ni and 112 Sn+ 58 Ni reactions at 35A MeV De Filippo E., Pagano A., Russotto P., Amorini F., Anzalone A. et al. Physical Review C 86 (2012) Particle identification using the (DELTA)E-E technique and pulse shape discrimination with the silicon detectors of the FAZIA project Carboni S., Barlini S., Bardelli L., Le Neindre N., Bini M. et al. NIM A 664 (2012) X-ray fluorescence from the element with atomic number Z=120 Frégeau M.O., Jacquet D., Morjean M., Bonnet E., Chbihi A. et al. Physical Review Letters 108 (2012)

26 Theoretical physics and phenomenology F. Aymard*, D. Durand, F. Gulminelli, O. Juillet, A. Leprevost* *PHD students New Quantum Monte Carlo methods for the nuclear shell model The nuclear shell model with configuration interaction is a powerful theoretical framework for studying the nuclear structure [1]. Unfortunately, the exponential scaling of the many-body space with the number of valence nucleons or the size of the single-particle basis strongly restricts its applicability. Quantum Monte-Carlo (QMC) methods are attractive techniques to overcome such limitations by offering an alternative to the diagonalization of the Hamiltonian. Indeed, these approaches reduces the many-body problem to a set of numerically tractable one-body problems describing independent particles that randomly walk in fluctuating external fields. To date, the shell model Monte-Carlo method (SMMC) is the principal application of QMC approaches to the shell model [2]. With schematic effective interactions, the SMMC method exactly reproduces the properties of even-even and N=Z odd-odd nuclei at zero and finite temperature. However, with realistic effective interactions or for other kinds of nuclei, the method is plagued by a dramatically vanishing signal-to-noise ratio that reveals the so-called fermion sign/phase problem. Moreover, the SMMC approach cannot achieve a detailed spectroscopy of nuclei. In such a context, we have proposed a new QMC approach for the shell model with the aim of reconstructing spectroscopy of nuclei with a well-managed sign/phase problem [3]. The originality of this phaseless QMC formalism, firstly suggested by S. Zhang and H. Krakauer in quantum chemistry [4], relies on an approximate wave function assuming two crucial roles. First, the trial state initiates and guides the Brownian motion in order to improve the efficiency of the method according to the importancesampling technique. Second, it is also used to control the sign/phase problem through a constraint on stochastic realizations in the spirit of fixed-node ab-initio calculations. Reconstructing shell model eigenstates implies that the trial wavefunction must have the same quantum numbers. Furthermore, the quality of the constrained-path approximation depends on the quality of the approximate state. A good trial wavefunction can thus be obtained via a Hartree-Fock-like approach with variation after projection onto angular momentum (VAP). This strategy for restoration of broken symmetries is rather usual in nuclear theory but remains at the cutting edge of variational methods. Indeed, as a superposition of symmetry-related independent-particle states, the VAP solutions can absorb correlations beyond the mean-field level. All the performed calculations prove that the VAP method yields good approximations for all the considered observables and thus offers a relevant trial state. Finally, the phaseless QMC scheme we proposed is based on a VAP wavefunction to initiate, guide, and constrain the stochastic paths. The yrast energies obtained for sd- and pf-shell nuclei with realistic effective interactions agree remarkably well with the values from exact diagonalization: For any spin, phaseless errors do not exceed 0.3% with statistical error bars about kev [3]. A convincing example is given by the spectrum of 27 Na (these yrast energies are partially reproduced on the right panel of Fig. 2). Indeed, the odd-mass nuclei are the most pathological cases in QMC simulations because they induce a particularly serious sign/phase problem even with schematic interactions. In addition, the phaseless QMC approach accurately reproduces the binding energies of 56 Ni (see Fig. 1). The formalism also allows for a complete reconstruction of lowlying spectroscopy through the determination of VAP wavefunctions orthogonal to the previously computed one. For instance, if ψ refers to the VAP ground state for any given angular momentum, the approximate first excited state of same spin is obtained by minimizing the energy with the wavefunctionq ψ, where ψ is an independent-particle state projected onto the desired quantum numbers andq the projector onto the subspace orthogonal to ψ. 22

27 By using the VAP solution as the trial state in the phaseless QMC approach, one then achieves a stochastic sampling of the true first excited state. The preliminary results obtained are promising, as shown in Fig. 2 where samples of results from this extension of the method are reported, for the first excited states of 28 Mg and 27 Na respectively. Again, a very good agreement between QMC and the exact results is obtained. In conclusion, the phaseless QMC approach, with a variational symmetry-restored wave-functions to guide and constrain walkers, may be considered as a powerful tool to address the structure of nuclei out of reach of conventional shell model treatments that usually require strong truncations of the configuration space. Fig. 1: Binding energies of 56 Ni as obtained with the VAP and phaseless QMC methods compared to the exact values extracted from [5]. The realistic GXPF1A effective interaction is considered [6]. The lighter areas indicate the QMC statistical errors. Fig. 2: Observables for the two first 0,2 and 5/2,3/2 states of 28 Mg and 27 Na respectively, as obtained from VAP and QMC calculations and compared to exact values with the realistic USD [7] effective interaction. 23

28 Microscopic modelling of star matter We are involved since several years in the theoretical modelling of the matter equation of state in the density and temperature conditions where it can be described by nucleonic degrees of freedom [8], and in its applications for the understanding of the neutron star structure and core-collapse supernova evolution. These works are in collaboration with IFIN (Bucarest), IPNO (Orsay), IPNL (Lyon) and supported by the ANR SN2NS ( ). At zero temperature, all the information is contained in the nuclear energy functional in its isoscalar and, more important, isovector channels. Though generally agreed that the symmetry energy plays a dramatic role in determining the structure of neutron stars and the evolution of core-collapsing supernovae, little is known in what concerns its value away from normal nuclear matter density and, even more important, the correct definition of this quantity in the case of inhomogeneous matter. Indeed, nuclear matter traditionally addressed by mean field models is uniform while clusters are known to exist in the dilute baryonic matter which constitutes the main component of compact objects outer shells. We have investigated the meaning of symmetry energy in the case of clusterized systems and the sensitivity of the proto-neutron star composition and equation of state to the effective interaction. To this aim we have developed an improved Nuclear Statistical Equilibrium (NSE) model [9], where the same effective interaction is consistently used to determine the clusters and unbound particles energy functionals in the self-consistent mean-field approximation. In particular, it is well known that cluster self-energies should be deeply modified in the nuclear medium. We have explored the ground-state properties of nuclear clusters embedded in a gas of nucleons with the help of Skyrme-Hartree-Fock microscopic calculations [10]. We parameterize their density profiles in spherical symmetry in terms of basic properties of the energy density functionals used and propose an analytical, Woods-Saxon density profile whose parameters depend, not only on the composition of the cluster, but also of the nucleon gas. We have studied the clusters energies with the help of the local-density approximation, validated through our microscopic results. We found that the excluded volume effect does not exhaust the in-medium effects and an extra isospin and density-dependent energy shift has to be considered to consistently determine the composition of subsaturation stellar matter. The symmetry energy of diluted matter is seen to depend on the isovector properties of the effective interaction, but its behavior with density and its quantitative value are strongly modified by clusterization. Our studies provide a simple, but microscopically founded modeling of the properties of clusterized matter at both zero and finite temperature, for direct use in consequential applications of astrophysical interest. At super-saturation densities, the description of stellar matter is complicated by the emergence of the strangeness degree of freedom. In collaboration with LUTH (Meudon), we have evaluated the phase diagram of a system constituted of neutrons and L-hyperons in thermal equilibrium in the mean-field approximation [11]. We have shown that this simple system exhibits a complex phase diagram with first and second order phase transitions. In a successive paper [12], we have analyzed the complete three-dimensional space given by the baryon, lepton and strange charge. We show that the phase diagram at subsaturation densities is strongly affected by the electromagnetic interaction, while it is almost independent of the electric charge at supra-saturation density. As a consequence, stellar matter under the condition of strangeness equilibrium is expected to experience a first as well as a second-order strangeness-driven phase transition at high density, while the liquid-gas phase transition is expected to be quenched (Fig. 3). An RPA calculation indicates that the presence of this critical point might have sizable implications for the neutrino propagation in core-collapse supernovae. Fig. 3: Borders of the phase-coexistence domains at zero temperature and strangeness chemical potential. Upper: (n, p, L)-mixture in the baryon versus charge density plane. Lower: (n, p,l, e)-mixture. Red: liquid-gas phase transition of non-strange dilute nuclear matter; blue: non-strange to strange phase transition. The arrows mark the directions of phase separation. 24

29 Pairing and alpha-clustering at high excitation The modification of nuclear structure at high excitation energy is poorly known. We particularly focus on pairing effects in the nuclear level densities, and on the possible persistence of alpha-clustering in light even-even nuclei beyond the threshold for multi-alpha emission. These studies are done by means of a series of dedicated experiments in collaboration with the GARFIELD experimental collaboration. Our first study has focused on the reactions 32 S+ 58 Ni and 32 S+ 64 Ni at 14.5 A MeV [13]. Evidence was found for important oddeven effects in isotopic observables of selected peripheral collisions corresponding to the decay of a projectile-like source. The influence of secondary decays on the staggering was studied with a correlation function technique. It was shown that this method is a powerful tool to get experimental information on the evaporation chain, in order to constrain model calculations. Specifically, we show that odd-even effects are due to interplay between pairing effects in the nuclear masses and in the level densities. In a successive experiment [14], dissipative 12 C+ 12 C reactions at 95 MeV were fully detected in charge with the GARFIELD and RCo apparatuses at LNL. A comparison to a dedicated Hauser-Feshbach calculation allows to select events which correspond, to a large extent, to the statistical evaporation of highly excited 24 Mg, as well as to extract information on the isotopic distribution of the evaporation residues in coincidence with their complete evaporation chain. Residual deviations from a statistical behavior were observed in alpha yields and attributed to the persistence of cluster correlations well above the 24 Mg threshold for 6 alpha s decay (Fig. 4). Fig. 4: Experimental (black dots) and calculated (red lines) relative energy distributions of the two alpha s in coincidence with an oxygen in dissipative (left) and non-dissipative (right) events. In the left panel, a zoom on the low relative energy region with a reduced energy binning is shown in the figure inset, to better see the structures of the energy correlation. Generic phenomena in nuclear physics at finite temperature The development of finite temperature mean-field and cluster models for nuclear physics and astrophysics applications has allowed us to evidence generic phenomena in statistical mechanics, which can potentially lead to interdisciplinary applications. A first application concerns ensemble inequivalence in finite systems [15] as well as at the thermodynamic limit [16]. We explore the conditions under which the particle number conservation constraint deforms the predictions of fragmentation observables as calculated in the grand-canonical ensemble. We derive an analytical formula allowing extracting canonical results from a grand-canonical calculation and vice-versa. This formula shows that exact canonical results can be recovered for observables varying linearly or quadratically with the number of particles, independent of the grand-canonical particle number fluctuations. We explore the validity of such grandcanonical extrapolation for different fragmentation observables in the framework of the analytical Grand Canonical or Canonical Thermodynamical Model [(G)CTM] of nuclear multifragmentation (Fig. 5). It is found that corrections to the grandcanonical expectations can be evaluated with high precision, provided the system does not experience a first-order phase transition. 25

30 In particular, because of the Coulomb quenching of the liquid-gas phase transition of nuclear matter, we find that mass conservation corrections to the grandcanonical ensemble can be safely computed for typical observables of interest in experimental measurements of nuclear fragmentation, even if deviations exist for highly exclusive observables. A second application [17] consists in the exploration of the connections between the description of interacting particles systems in terms of energy density functionals, as it is done for self-consistent mean-field nuclear physics models, and fractional exclusion statistics (FES) introduced in different condensed matter applications. We have considered a generic interacting particle system in the quasi-classical limit and in the mean-field approximation. We have defined the FES quasiparticle energies, we calculate the FES parameters of the system and we deduce the equations for the equilibrium particle populations. The FES gas is ideal", in the sense that the quasiparticle energies do not depend on the other quasiparticle levels populations and the sum of the quasiparticle energies is equal to the total energy of the system. We have proved that this FES formalism is equivalent to the semi-classical or Thomas Fermi (TF) limit of the self-consistent mean-field theory and the FES quasiparticle populations may be calculated from the TF populations by making the correspondence between the FES and the TF quasiparticle energies. Fig. 5: In the upper-left panel and lower left panel canonical (solid lines) and grand canonical (dotted lines) mass distribution and largest cluster probability distribution are shown for A=50 (black) and 400 (red) at T=4 MeV. In the upper-right panel and lower right panel the same observables are plotted for a system A=200 at T=3 MeV (black) and 7 MeV (red) ). 26

31 References [1] E. Caurier, G. Martínez-Pinedo, F. Nowacki, A. Poves, & A. P. Zuker, Rev. Mod. Phys. 77, 427 (2005). [2] S. E. Koonin, D. J. Dean, & K. Langanke, Phys. Rep. 278, 1 (1999), and references therein. [3] J. Bonnard & O. Juillet, Phys. Rev. Lett. 111, (2013). [4] S. Zhang & H. Krakauer, Phys. Rev. Lett. 90, (2003). [5] S. Pittel & B. Thakur, Acta Phys. Pol. B 42, 427 (2011). [6] M. Honma, T. Otsuka, B. A. Brown, & T. Mizusaki, Eur. Phys. J. A 25, 499 (2005). [7] B. H. Wildenthal, Prog. Part. Nucl. Phys. 11, 5 (1984). [8] F.Gulminelli, Neutron rich nuclei and the equation of state of stellar matter, Phys. Scripta T152, (2013) [9] Ad. R. Raduta, F. Aymard, F. Gulminelli, "Clusterized nuclear matter in the (proto-)neutron star crust and the symmetry energy", EPJA 50 (2014) 24 [10] P. Papakonstantinou, J. Margueron, F. Gulminelli, Ad.R. Raduta, "Densities and energies of nuclei in dilute matter", Phys. Rev. C 88, (2013) [11] F. Gulminelli, Ad. Raduta,M. Oertel, "Phase transition toward strange matter", Phys. Rev. C 86, (2012) [12] F. Gulminelli, Ad. Raduta,M. Oertel, Coulomb effects in strangeness-driven phase transition of stellar matter», Phys. Rev. C 87, (2013) [13] M. D'Agostino, M. Bruno, F. Gulminelli, L. Morelli, G. Baiocco, & the GARFIELD collaboration, Towards an understanding of staggering effects in S+Ni collisions, Nucl. Phys. A 875, 139 (2012). [14] G. Baiocco, L. Morelli, F. Gulminelli & the GARFIELD collaboration, alpha-clustering effects in dissipative C+C reactions at 95 MeV, Phys. Rev. C 87, (2013) [15] G. Chaudhuri, F. Gulminelli & S.Mallik, On the effect of particle number conservation in nuclear fragmentation, Phys. Lett. B 724, 115 (2013). [16] F. Gulminelli & Ad. R. Raduta, «Ensemble inequivalence in supernova matter within a simple model», Phys. Rev. C 85, (2012). [17] D.V. Anghel, G.A. Nemnes & F. Gulminelli, "Equivalence between fractional exclusion statistics and self-consistent meanfield theory in interacting particle systems in any number of dimensions", Phys. Rev. E 88, (2013). Publications α-clustering effects in dissipative 12 C+ 12 C reactions at 95 MeV Baiocco G., Morelli L., Gulminelli F., D'Agostino M., Bruno M. et al. Physical Review C 87 (2013) Transformation between statistical ensembles in the modelling of nuclear fragmentation Chaudhuri G., Gulminelli F., Mallik S. Physics Letters B 724 (2013) Densities and energies of nuclei in dilute matter at zero temperature Papakonstantinou P., Margueron J., Gulminelli F., Raduta A. Physical Review C 88 (2013) Strangeness-driven phase transition in star matter Gulminelli F., Raduta A., Oertel M., Margueron J. Physical Review C 87 (2013) Equivalence between fractional exclusion statistics and Fermi liquid theory in interacting particle systems Anghel D.V., Nemnes G.A., Gulminelli F. Physical Review E 88 (2013) A Constrained-Path Quantum Monte-Carlo Approach for the Nuclear Shell Model Bonnard J., Juillet O. Physical Review Letters 111 (2013) Exotic spin, charge and pairing correlations of the twodimensional doped Hubbard model: a symmetry entangled meanfield approach Juillet O., Frésard R. Physical Review B (Condensed Matter) 87 (2013) Towards an understanding of staggering effects in dissipative binary collisions D'Agostino M., Bruno M., Gulminelli F., Morelli L., Baiocco G. et al. Nuclear Physics A 875 (2012) Phase transition towards strange matter Gulminelli F., Raduta A., Oertel M. Physical Review C 86 (2012) Ensemble inequivalence in supernova matter within a simple model Gulminelli F., Raduta A. Physical Review C 85 (2012)

32 RESEARCH INTERDISCIPLINARY RESEARCH Nuclear waste management Medical and industrial applications 28

33 Nuclear waste management G. Ban, T. Chevret*, F-R. Lecolley, J-L. Lecouey, G. Lehaut, N. Marie-Nourry Collaboration : LPSC Grenoble, IPHC Strasbourg, SCK.GEN Mol (Belgique), CEA Cadarache, CEA/IRFU Saclay, GANIL Caen *PHD student Over the past two years, in the framework of the GUINEVERE experiment and FREYA program we have pursued the experiments carried out at the subcritical, lead moderated VENUS-F fast core. Pulsed Neutron Source measurements (PNS) and short continuous beam interruptions were performed and analysed in order to estimate the reactivity of various configurations of the reactor. The second part of our activities was devoted to the development of an original experimental device in the framework of the FALSTAFF project whose objective is the study of fission using the neutron source facility (NFS) at SPIRAL2. The GUINEVERE Experiment Brief description The GUINEVERE (Generator of Uninterrupted Intense NEutrons at the lead Venus REactor) experiment [GUI] is dedicated to feasibility studies for Accelerator Driven Systems (ADS) which are envisaged in partitioning and transmutation strategies. It aims at providing a zero power experimental facility to investigate sub-criticality on-line monitoring procedures and to validate simulation tools. These issues are of major importance in view of the achievement of a future powerful ADS such as the MYRRHA project [MYR]. The GUINEVERE facility is hosted at the SCK CEN site in Mol (Belgium) and consists in the coupling of the fast VENUS-F reactor to a neutron source provided by the GENEPI-3C accelerator. The fast VENUS-F reactor consists of square fuel assemblies (FA) composed of a 5 5pattern mixture of fuel and solid lead rodlets, the latter acting as a fast system coolant. Radially and axially the fissile zone is surrounded by lead reflectors. The outer side length of a FA is 80 mm. The fuel is 30% 235 U enriched metallic uranium provided by CEA. The FA are arranged in a cylindrical geometry (~800 mm in diameter, 600 mm in height). The VENUS-F core is equipped with six safety rods, two control rods (CR) and an absorbent rod (PEAR rod). The GENEPI-3C accelerator [GUI] provides neutrons via T(d,n) 4 He fusion reactions. It accelerates deuteron ions up to the energy of 220 kev and guides them onto a tritiated target located at the VENUS-F core centre. This provides a quasi-isotropic field of about 14 MeV neutrons. It can be operated in continuous mode or in pulsed mode with adjustable frequency. 29

34 In a first step, a critical configuration called CR0 was loaded and experimentally characterized [CR0]. In a second step, a subcritical configuration called SC1 (k eff ~0.96) was obtained by replacing the four central FA by the device devoted to the accelerator pipe hosting. Estimate of the reactivity of SC1 configuration using the MSM method The MSM method has been used to estimate the reactivity of the so-called SC1 configuration of the VENUS-F core for different heights of the two control rods (see table 1). This reference measurement of SC1 subcritical level configurations by MSM method was used to estimate the reliability of the other methods of reactivity determination which could be applied in industrial ADS facilities. These methods (PNS, Source Jerk, Beam Interruption and others) have been investigated in the GENEPI-3C-driven subcritical VENUS-F core in the framework of the FREYA Project [FREYA] during the past two years. Configuration CR height (mm) ρ$ MSM ρ$ AREA ρ$ BIM SC1/CR@600 mm ± ± ±0.02 SC ± ± ±0.02 SC1/CR@200 mm ± ±0.02 SC1/CR@0 mm ± ± ±0.03 Table 1: Average reactivity value given by the Area method (AREA) and the Beam Interruption method (BIM) compared with the MSM reference value, for the different reactor configurations. Pulsed Neutron Source measurements Area method The current-to-flux technique was proposed to be combined to absolute reactivity measurements in order to establish a complete on-line reactivity measurement procedure for ADS [MUSE]. The absolute reactivity values are foreseen to be deduced from dynamical measurements requiring source variations. The study of the techniques used to analyse such measurements is one of the purposes of the GUINEVERE program. To evaluate their accuracy, they have to be applied in Pulsed Neutron Source (PNS) conditions for a given reactivity and their results have to be compared to the reference value given by the MSM method. One of the first methods investigated was the Area method [AREA] also referred as the Sjostrand method. It is based on the analysis of the time dependent response of detectors placed in the reactor to a pulsed neutron excitation and it allows determining in a straightforward way the reactivity of a subcritical nuclear reactor with no input from theoretical calculations as long as the assumptions of the neutron point kinetics holds in the reactor. Indeed within the one-delayed group approximation the equation of the time decrease of the neutron population (Eq. 1) after a pulse exhibits two components: a fast one due to prompt neutrons and a slow one due to delayed neutrons leading by simple integration over time respectively to the prompt surface A p and the delayed surface A d. Then, the ratio of these two surfaces gives directly the value of the antireactivity in dollars (Eq 1). ρ $ = A p A d = ρ β eff (1) Experimentally, for a set of pulses repeated with a fixed frequency, a single PNS histogram (an example is shown in Fig. 1) is constructed by summing all the detector time responses as a function of the time elapsed after the neutron pulse. After integrating the time spectrum to get the surfaces A p and A d, the antireactivity can be calculated using Eq. 1. Fig. 1: Time-dependent PNS histograms obtained with four different fission chambers. 30

35 The area method was applied to reaction rates measured by ten fission chambers during the PNS experiments for the three different subcritical configurations obtained by moving the control rods. The beam frequency was tuned at 220 Hz. Fig. 2 shows the results for the SC1 configuration. Similar results were obtained for the other configurations derived from SC1. Reactivity values extracted according to Eq. 1 are represented by solid dots. The error bars were calculated by taking into account the statistical as well as the systematic errors. The horizontal dashed line represents the reactivity of the subcritical configuration as measured using the MSM method, while the solid horizontal lines show the uncertainty range on the MSM value. One notices a dispersion of the results, which seem to depend on the detector location in the reactor. Three groups can be identified. The first one contains only the CFUF34 detector, which is the only one located in the reactor core. It is also the only one from which the reactivity value obtained with the Area method is in very good agreement with that of the MSM method. The second group gathers six detectors, which are located either at the core-reflector interface or in the corners of the grid, in the inner part of the reflector. The last detectors (RS10075, CFUL653 and CFUL659) form the third group and are located rather far away from the core, in the outer part of the reflector. Clearly the Area Method fails at providing the correct value of the reactivity when the detectors are not in the core. The effect seems to be stronger when the detector is farther from the core. Fig. 2: Uncorrected (solid dots) and corrected (open squares) reactivity values extracted from detector counts for the reactor configuration SC1. The MSM reference value is the dashed line and its uncertainty range is given by the solid lines. If the dispersion of the reactivity values given by the Area method is due to spatial effects, it should be possible to use Monte Carlo simulations of neutron pulses to correct for these effects since Monte Carlo simulations transport neutrons without approximations. MCNP [MCNP] correction factors were then calculated for each configuration and each detector location with a simplified version of the VENUS-F reactor. Corrected values are symbolized by open squares on Fig. 2. Except for the fission chambers installed in the outer lead reflector, the corrected values are all compatible with the value given by the MSM method. Finally, discarding the results obtained for the fission chambers located in the outer part of the reflector, the average corrected value of reactivity was calculated for the three configurations studied. To calculate the uncertainty, it was assumed conservatively that the correlations are at maximum between the values given by the detectors. As can be seen in Table 1, the agreement between the MSM reactivity and the Area Method is remarkable. Continuous Beam measurement - Beam Interruption Method Thanks to the presence of an external source in an ADS, one can extract the reactivity of the sub-critical reactor using interruptions of the source in a continuous mode within the neutron point kinetic model (Equation 4). G G t 1 Λ eff dn βieff λ it βieff λ it λ it' ρ$ ( t) = 1+ +n0 ( ) ( ) e + e n t' e dt' n t βeff dt i= 1 βeff i= 1 βeff t'= 0 (2) With ρ $ (t) the reactivity in dollar, n(t) the neutron population, n(t) the constant neutron population level before the beam interruption, Λ eff the mean generation time, β eff the effective delayed-neutron fraction, β ieff the effective delayed-neutron fraction of the delayed-neutron group i, λ i the decay constant of the delayed-neutron group i, and G the number of delayedneutron groups. If one assumes that point kinetic holds in the reactor, all the detector count rates evolve the same way as the neutron population and n(t) can be replaced in Eq. 2 by count rates from any detector and the reactivity is then readily extracted, once the kinetic parameters have been calculated using the deterministic code ERANOS [ERA]. During the experiment, beam interruptions were performed with a period of 25 ms (40 Hz) and the source was switched off for 2 ms. Fission event coming from fission chambers (FCs) were time stamped over a time range including each beam trip plus and minus 300 ms. For each FC, the histogram is obtained by summing all the time responses as a function of the time elapsed after each source jerk. Figure 3 shows histograms normalized to the same maximum for different detector location: in the core (CFUF34), in the inner part of the reflector (RS-10071) and in the outer part of the reflector (RS-10075). 31

36 Fig. 3: MCNP simulations (red with concrete walls, black without) compared to experimental data (blue) for three detectors at three different locations: CFUF34 in the core (left), RS in the inner reflector (centre) and RS in the outer reflector (right). At first, the neutron population decreases right after the source jerk, which corresponds to the prompt neutron population decrease. Then, more or less rapidly depending on the position of the detector, the neutron population tends to reach its delayed neutron level. As can be observed, the shape of the neutron population histogram over time strongly depends on the position of the detector in the reactor, as in the case of the PNS experiments. The CFUF34 detector seems to be the only one in agreement with Point Kinetics. Moreover, the farther the detector from the centre of the reactor, the more different the experimental shapes are from that predicted by Point Kinetics, and the later the neutron population reaches its delayed neutron level. This clearly indicates the presence of spatial effects that are not considered in Point Kinetics. In Fig. 4, the reactivity values (denoted raw reactivity ) given by the analysis of ten time dependent FC count rate histograms using equation 2 and taking into account several corrections (dead time, duty cycle, etc ) are shown as solid black circles and compared to the reference values given by the MSM method. Fig. 4: Raw reactivity values (solid dots) and corrected reactivity values (open squares) for each detector and for the four configurations studied: SC1/CR@0 mm at the upper left corner of the figure, SC1/CR@240 mm at the upper right corner, SC1 at the bottom left corner and SC1/CR@600 mm at the bottom right corner. The MSM reference value is symbolized by the dashed line (red), and the solid lines (red) are its uncertainty range. 32

37 As expected, the reactivity extracted depends strongly on the detector position. The CFUF34 detector, which is located inside the core and which exhibits a neutron population shape closer to that given by Point Kinetics, is the only one giving an anti-reactivity in agreement with the MSM reference values. The anti-reactivity values obtained using the six detectors located in the inner reflector are significantly underestimated and gathered around the same value. As for the detectors located in the outer reflector, they give reactivity values even farther off. This is not surprising since they exhibit the time responses which are the least similar to those given by Point Kinetics. In order to investigate the origin of these strong spatial effects that are not considered in Point Kinetics and lead to such scatter of the results, Monte Carlo simulations were carried out using MCNP and a simplified reactor geometry. At first, simulations were done without the concrete walls surrounding the reactor vessel and fail to reproduce the diversity of the experimental FC time histograms, all the simulated count rate evolution shapes being similar. However, when simulations include the concrete walls around the reactor, experimental shapes are well reproduced (see Fig. 3). Neutrons leaving the reactor can collide with the concrete wall elements, and thus may have their energy greatly reduced by collisions on light elements in the walls. Since the FC deposits are made of 235 U whose cross-section is the largest for low-energy neutrons, the influence of a small amount of slowed-down neutrons on the detector count rates can become quite significant. It appears that the concrete walls must be considered as a part of the reactor reflector. The fact that taking into account the concrete walls in the geometry allows reproducing the experimental data with a very good agreement opens up a way to correct the raw reactivity values obtained experimentally. The corrected results are shown as open squares in Fig. 4. An impressive consistency between the MSM reference values and the corrected ones derived from beam interruption analyses is observed. As expected given the comparison between experimental data and Point Kinetics, the reactivity obtained from the CFUF34 detector located in the core is almost left unchanged by the correction. Except for CFUL01-658, we observe that all detectors provide final reactivity values in agreement with the MSM method for all the configurations studied. It is important to note that, in industrial ADS, it might be difficult to install detectors in the reactor core due to the high flux that would prevail in it. Being able to correct the spatial effects that occur in the inner and the outer reflector is therefore an important result. To conclude with the extraction of the reactivity by studying the evolution of the neutron population during a beam interruption, Table 1 gathers the reactivity value obtained for each configuration by averaging the results from all detectors, and the MSM reference values. The uncertainties were computed assuming maximum correlations between the values given by the detectors. The agreement between the MSM technique and the beam interruption analysis method is very good. Conclusion Taking advantage of the different operating mode of the GENEPI3-C accelerator, two different methods to estimate the reactivity of a sub-critical reactor based on the analysis of the evolution of the neutron population have been tested on data collected at the GUINEVERE facility within the FREYA project: the Area method with a Pulsed Neutron Source and the Beam Interruption method with a Continuous Beam. The data analysis, using point kinetics theory, has been applied to count rates obtained with ten fission chambers located in the VENUS-F reactor. First, various shapes for the time dependent FC count rates were observed depending on the FC position. That indicated the presence of spatial effects, that appear to get stronger as the location of the detector is farther away from the core centre. MCNP simulations have then been used to compute correction factors in order to correct the raw reactivities obtained from the point kinetics analysis. Finally, corrected reactivity values are compatible with the MSM reference ones. The FALSTAFF Project Brief description of the project The FALSTAFF project [FALS] aims at providing highly constraining data to significantly improve the description of the fission process. More specifically, the goal is to measure the neutron multiplicity as a function of the fragment characteristics (mass, nuclear charge and kinetic energy) in neutron-induced fission of specific actinides in the MeV range. New developments on microscopic calculations and the future generation of nuclear reactors are two of the main motivations for new experimental programs devoted to the study of fission. 33

38 Ionization chamber with scintillating gas In order to minimize energy straggling of fission fragments, one possibility is to use ionization chamber (IC) not only to identify and measure the energy of the fission fragments but also to measure their velocity via the well-known time-of-flight technique. However the time response of an IC does not reach the resolutions required for a good determination of the fission fragment velocity. That is the reason why we have developed an IC filled with scintillating gas and coupled to a pair of photomultipliers (PMT) through transparent windows (Fig. 5). The light emitted by the gas provides the stop signal for the timeof-flight measurement. Fig. 5: Sketch of the ionization chamber with scintillating gas During the past two years, this scintillating IC has been qualify using several gas (N 2, CF 4 ) at different pressure, ranging from atmospheric pressure down to 200 mbar, with alpha particle and fission fragment emitted by a 252 Cf source. The energy resolution (table 2) has been determined using a tri-alpha source at atmospheric pressure. E alpha (MeV) σ (kev) N 2 σ (kev) CF Table 2: σ (kev) of each alpha peaks for the different gas Fig. 6, left part, shows the correlation distribution between the number of photo-electron detected in each PMT for CF 4 and N 2 gas respectively, at atmospheric pressure and with an alpha source. The coincidence time resolutions extracted from the gaussian fits (Fig. 6, right part) are σ=0.34 ns in CF 4 and 1.4 ns in N 2. Fig. 7 shows the correlation distribution between the number of photo-electron detected in each PMT for the CF 4 gas at different pressure and with a californium 252 source. By selecting events upper the red line, we have then measured the coincidence time resolution associated to fission fragment. At the lowest pressure (250 mbar) a coincidence time resolution of σ=210 ps was found, leading to a time resolution of σ=150 ps for each PMT. Fig. 6: Left: correlation plot of the PMT, right: time distribution of correlated events in both PMT. Top: CF4, bottom: N2. 34

39 Fig. 7: Correlation plot of the PMT at different pressure in CF 4. Conclusion Despite a good time resolution when using the CF 4 gas, the performance of our scintillating ionization chamber does not meet the FALSTAFF requirement if one wants to measure fission fragment masses after neutron evaporation with an accuracy of 1 mass unit. However, this kind of detector can be envisaged in other applications, e.g. cross-section measurement of alpha particle production in neutron induced reaction on oxygen below 20 MeV with the use of a gas mixture (CF 4 + CO 2 ). This latter reaction is of interest for the community and is one of the request of the High Priority Request List of the NEA/OCDE [HPRL] References [AREA] N.G. Sjostrand, Arkiv för Fysik Band 11 nr 13, 233 (1956) [CR0] W. Uyttenhove et al., Experimental Results from the VENUS-F Critical Reference State for the GUINEVERE Accelerator Driven System Project, Proceeding of the Int. Conf. on Advancements in Nuclear Instrumentation, Measurement Methods and their Application, ANIMMA, Ghent, Belgium (June ). [ERA] M. Carta, private communication [FALS] F.R. Lecolley et al., AccApp 2013, Bruges (Belgium) [FREYA] FREYA collaboration, FP Publication [GUI] A. Billebaud et al., The GUINEVERE Project for Accelerator Driven System Physics, Proceedings of Global 2009, Paris, France (September 6-11, 2009). [HPRL] [MCNP] MCNP - A General Monte Carlo N-Particle Code, Version 5, LA-ORNL, RSICC LA-UR , Los Alamos National Laboratory (2003) [MUSE] MUSE collaboration, 5th EURATOM FP-Contract#FIKW- CT Deliverable #8: Final Report (2005) [MYR] H.A. Abderrahim et al., MYRRHA Technical Description, Technical Report for the OECD MYRRHA Review Team, SCK CEN, Belgium (2008). Experimental Results From the VENUS-F Critical Reference State for the GUINEVERE Accelerator Driven System Project Uyttenhove W., Baeten P., Ban G., Billebaud A., Chabod S. et al. IEEE Transactions on Nuclear Science 59 (2012) A. Billebaud, A. Kochetkov, S. Chabod, X. Doligez, G. Lehaut, F.-R. Lecolley, J.-L. Lecouey, N. Marie, F. Mellier, V. Bécares, D. Villamarin, G. Vittiglio, H.-E. Thyébault, W. Uyttenhove, J.Wagemans FREYA project, 7th EURATOM FP-Contract # Deliverable 1.1: Current subcritical core results, S.Di Maria, A. Kochetkov, G.Mila, S.Argiro, M.Carta, F. Gabrielli, G. Vittiglio, S. Chabod, P. Gajda, N. Marie, W. Uyttenhove, G. Lehaut, A. Billebaud, X. Doligez, F.-R. Lecolley, J.-L. Lecouey, V. Bécares, D. Villamarin, Y.Romanets FREYA project, 7th EURATOM FP-Contract # Deliverable 1.2: Deep subcritical experiments,

40 Medical and industrial applications G. Boissonnat*, J. Colin, D. Cussol, J. Dudouet*, J.M. Fontbonne, M. Labalme, S. Salvador *PHD students The "Medical and Industrial Applications" team is involved in dosimetry measurements for medical and industrial purposes since its creation. For eight years the group is strongly involved into the development of beam monitors and carbon fragmentation studies for hadrontherapy. Hadrontherapy consists in irradiating cancerous tumours with light nuclei such as proton or carbon ions. Proton therapy is now widely spread worldwide. Carbon therapy is growing in importance. To be as efficient as possible in irradiating the tumour, all physics and biological processes which may occur during the treatments must be kept under control. A specific software, the Treatment Planning System or TPS, is used to define the machine parameters for a given patient, pathology and accelerator. Once these parameters are determined, different setups are necessary to control the irradiation process. Nuclear physicists can contribute to hadrontherapy in two ways: by optimizing the dose calculation module of TPS by studying the physical processes involved in the irradiation process ; by designing and building devices which can help to monitor the beam and which may allow controlling the dose deposition in the patient. The "Medical and Industrial Applications" team is also strongly involved in the ARCHADE project. This centre will be dedicated to the medical, biological and physical research in carbon-therapy and will be located at Caen. The group contributes to FRANCE-HADRON which gathers all the scientific terms in medicine, biology and physics which contribute to the development of hadrontherapy in France. Beam monitors The "Medical and Industrial Applications" team is developing beam monitors for the radio-biology experiments and for treatment centres. The use of swept pencil beams is more and more common in proton-therapy. It consists in delivering the dose by scanning the tumour with several beam spots. Each spot corresponds to a given beam location, energy and fluency. The main advantage compared to a passive beam dose delivery which uses beam range shifters and boluses to conform the dose to the tumour geometry is that less matter is set in the beam and hence less secondary particles (mainly neutrons) are produced. The price to pay is that the beam delivery is more complex. The correlations between the beam fluency, its energy and location have to be accurately controlled all along the irradiation. The beam monitor has to be as transparent as possible in order to minimize its disturbance on the beam (angular spreading, intensity attenuation, energy diminution). 36

41 In order to minimize the irradiation time during proton-therapy treatments, the trend is to increase proton beam intensities. The former IC2/3 beam monitor is not well suited anymore. In collaboration with the Ion Beam Applications (IBA) Company, new studies have been initiated to design and build a proton beam monitor for high intensities up to 10 9 ions per second. This is the subject of the PhD thesis of G. Boissonnat. A beam monitor for radio-biology experiments at GANIL has also been developed and tested at GANIL in September 2013 in the framework of the FRANCE-HADRON collaboration. This beam monitor called DOSION III is an adaption of the IC2/3 beam monitor for GANIL beams. Fragmentation studies The "Medical and Industrial Applications" team is also studying the fragmentation processes of carbon ions which contribute to spread the dose deposition beyond the Bragg peak. Although hadrontherapy has an obvious ballistic advantage compared to conventional radiation therapies, fragmentation processes may reduce this advantage. They occur when a projectile hits a nucleus present in the tissues. Secondary fragments produced by this interaction are much lighter and have a velocity close to the velocity of the projectile. As a result, the secondary particles have a longer range and deposit some dose in and beyond the Bragg peak of the initial projectile. Fig. 1: Schematic view of the experimental set-up of the May 2011 experiment at GANIL. The effect of the nuclear fragmentation process is twofold. The number of projectiles which do not experience a nucleusnucleus collision decreases strongly with respect to the penetration depth. Only one half of the initial carbon projectiles at 290 MeV/u reaches the maximum range. As a consequence, the dose deposition at the Bragg peak is strongly influenced by the nuclear reaction cross section. The other effect of the fragmentation process is the appearance of a tail beyond the Bragg peak. This so called "fragmentation tail" is mainly due to protons and alpha particles having a longer range. In addition, the secondary fragments may have different biological effects (cell death, mutation rates and metabolic changes) according to their nature. In order to compute accurately the dose deposition and the resulting biological effects, it is necessary to have an accurate knowledge of the fragmentation process of the projectile in human tissues. The ideal situation would be to have a valuable model which could predict the production rates of secondary particles and their angular and energy distributions. Uncertainties on the dose calculations are dominated by the uncertainties on fragmentation cross sections and on nuclear reaction models. Two experiments have been performed in May 2008 and in May 2011 with the ECLAN reaction chamber at the GANIL G22 beam line. These experiments have been performed in the framework of the GDR MI2B and in collaboration with the IPHC Strasbourg, IPN Lyon and SPhN Saclay. The carbon energy was 95 MeV/u for both experiments. A schematic view of the May 2011 experimental set-up is shown on Fig. 1. It included five three-layer E/ E/E telescopes for charged particles detection. The telescopes were fixed on rotating stages of the ECLAN chamber. This allowed covering angles ranging from 4 to 70. For the may 2008 experiment, six PMMA targets (C 5 H 8 O 2 ; d=1.19 g/cm 3 ) of different thicknesses: 5, 10, 15, 20, 25 and 40 mm were used. For the May 2011 experiment, the experimental set-up was very similar and thin C, CH 2, Al, Al 2 O 3 and Ti target were used. The nuclear reaction cross sections and the fragments production rates for H, C, O and Ca (close to Ti) nuclei have been extracted from this experiment. These nuclei are the most abundant nuclei in human tissues (more than 90%). This experiment used the FASTER acquisition system. The measurements for a thin PMMA target have also been performed for cross-checking. In September 2013 at GANIL, a complementary experiment has been performed in the framework of the FRANCE-HADRON collaboration to measure the secondary fragments production cross sections at 0 for carbon ions at 95 MeV/u colliding thin C, CH 2, Al and Ti targets. The particle identification was done by using the standard E/E technique. 37

42 The results of the May 2008 experiment showed that none of the nucleus-nucleus collision models implemented in GEANT4 (Binary Intranuclear Cascade, Quantum Molecular Dynamics, Intra Nuclear Cascade) are able to reproduce the experimental data. The thin target experiments are best suited to constrain models since the fragmentation process occurs in a very small energy range. This is the PhD thesis subject of J. Dudouet. Angular distributions of some fragments are presented on Fig. 2. The double differential cross sections have been obtained with a good accuracy (~5% to 10%) for almost all isotopes lighter than carbon nucleus. It has also been shown that the cross sections for composite targets can be obtained within 5% accuracy by combining the cross sections obtained from individual nuclei. GEANT4 nucleus-nucleus collision models have been compared to the experimental data. As for the thick target experiments, none of these models are able to reproduce all the experimental data. The experimental data have been given a free-access on a web site ( Fig. 2: Angular distributions of differents fragments for a carbone projectile at 95 MeV/u hitting a thin carbon target. The group is also participating to the FIRST European collaboration, which has measured C-C reactions at 400 MeV/u. Particle charge identification using the time-of-flight wall of the 2011 experiment at GSI Darmstadt has been done in our group. Moreover, the preliminary design of an experimental set-up called FRACAS (FRAgmentation du CArbone et Sections efficaces) for the ARCHADE centre has been done. It will consist of a time-of-flight wall for particle charge identification. The mass measurement will be done by means of a magnet and tracking detectors located before and after the magnet. This set-up should be efficient for the whole energy range of medical interest, i.e. from 80 MeV/u to 400 MeV/u. The selection of the best suited materials and detection techniques for all these elements is under progress in collaboration with the IPHC Strasbourg. The group is also strongly involved in the FRANCE-HADRON collaboration. We are leading the Working Package 2 "Improving treatment planning in hadron-therapy, and are members of the management committees. The management of the GANIL beam time for the experiments performed in September 2013 has been done by the group. Collaboration with industry The dose measurements on workers manipulating nuclear wastes are of great importance. The contribution of low energy gamma and X rays (E<60 kev) to the global dose is not yet well measured. In the framework of the collaboration between the PIERCAN company and the LPC Caen, the "Medical and Industrial Applications" team has performed simulations and measurements to characterize the protection of gloves. In the framework of the collaboration between AREVA MELOX company and the LPC Caen, the "Medical and Industrial Applications" team has performed a prototype of a new dosimeter able to measure the dose on a large range in energy for nuclear workers. Two patents have been submitted. 38

43 Patents J-M. Fontbonne, C. Fontbonne, J. Colin, J. Jehanno Procédé d asservissement du gain et du zéro d un dispositif de comptage de photons à pixels multiples et système de mesure de lumière mettant en œuvre ce procédé Dépôt Fév under examination J-M. Fontbonne, J. Colin, C. Fontbonne, J. Jehanno Procédé de mesure de dose au moyen d un détecteur de rayonnements, notamment d un détecteur de rayonnements X ou gamma, utilisé en mode spectroscopie et système de mesure de dose utilisant ce procédé Dépôt Fév. 2013, under examination Publications Proton computed tomography from multiple physics processes Bopp C., Colin J., Cussol D., Finck C., Labalme M. et al. Physics in Medicine and Biology 58 (2013) 7261 Double differential fragmentation cross-section measurements of 95 MeV/u 12 C on thin targets for hadrontherapy Dudouet J., Juliani D., Angelique J.C., Braunn B., Colin J. et al. Physical Review C 88 (2013) Comparison of two analysis methods for nuclear reaction measurements of 12 C + 12 C interactions at 95 MeV/u for hadrontherapy Dudouet J., Juliani D., Labalme M., Angélique J.C., Braunn B. et al. NIM A 715 (2013) Benchmarking GEANT4 nuclear models for carbon-therapy at 95 MeV/A Dudouet J., Cussol D., Durand D., Labalme M. To be published in Physical Review C Characterization and performances of a monitoring ionization chamber dedicated to IBA-universal irradiation head for Pencil Beam Scanning Courtois C., Boissonnat G., Brusasco C., Colin J., Cussol D. et al. NIM A 736 (2014)

44 RESEARCH FUNDAMENTAL INTERACTIONS Precise correlation measurements in nuclear beta decay High resolution study of low energy charge exchange collisions with a MOT (magneto-optical trapped) target Towards a new measurement of the neutron Electric Dipole Moment (EDM) Search for neutrinoless double beta decay 40

45 GRoup Interactions FOndamentales et nature du Neutrino (GRIFON) G. Ban, C. Couratin *1, D. Durand, X. Fabian *1, X. Fléchard, B. Guillon, V. Hélaine *2, T. Lefort,Y. Lemière, A. Leredde *3, E. Liénard, F. Mauger, O. Naviliat 4, G.Quéméner Experimental facilities : Institut Laue-Langevin (ILL), Paul Scherrer Institute (PSI), GANIL/SPIRAL, Laboratoire Souterrain de Modane (LSM), CENPA (Seattle), CERN/ISOLDE. *PHD students 1 LPCTrap, 2 nedm, 3 MOT 4 NSCL/MSU East Lansing The GRIFON group at LPC Caen is involved in four experimental research activities : LPCTrap: search for exotic couplings in nuclear weak decay processes, MOT: High resolution studies of atomic collisions in a MOT, nedm: measurement of the neutron electric dipole moment, NEMO-3/SuperNEMO: search for neutrinoless double beta decay. Each member of the group participates to one or two of these experimental programs. The instrumental and technical know-how and specific skills acquired in these research activities are often shared within the GRIFON group: detection of electrons in the MeV energy range with PMT-based optical modules (LPCTrap/SuperNEMO), electric and magnetic fields computing (nedm/supernemo/lpctrap), Monte-Carlo simulations (SuperNEMO/LPCTrap), parallel computing (nedm/lpctrap). Most of these research activities are maintained within the framework of internationally recognized collaborations. The experi-mental setups are hosted in first-class facilities : Institut Laue-Langevin (ILL), Paul Scherrer Institute (PSI), GANIL/SPIRAL, Laboratoire Souterrain de Modane (LSM), Center for Nuclear Physics and Astrophysics (CENPA, Seattle), CERN/ISOLDE. For all these projects, the LPC Caen occupies a prominent position with well identified responsabilities and recognized skills. 41

46 Precise correlation measurements in nuclear beta decay In collaboration with : LPC Caen, CIMAP Caen, GANIL Caen, IKS/KUL Leuven, NSCL/MSU East Lansing, Univ. Granada, CENPA Seattle, CELIA Bordeaux, Argonne National Laboratory, NCNR Warsaw The LPCTrap setup installed at GANIL/LIRAT has been built to measure with high precision beta-neutrino angular correlation parameters, a, in nuclear beta decays [Ban13]. Such measurements provide the most stringent limits on exotic Scalar (S) or Tensor (T) type contributions to the nuclear weak decay process. Precisions at a level of at least 0.5% are needed to improve the current sensitivity to these exotic weak interaction components. In addition, in the case of a mirror decay, the measurement of a enables to precisely determine the mixing ratio, ρ, between the Gamow-Teller (GT) and the Fermi (F) components in the transition. Combined with the Ft value, this mixing ratio can be used to determine accurately V ud, the first element of the Cabibbo-Kobayashi-Maskawa (CKM) matrix. Again, measurements of a at a precision level of at least 0.5% in some selected mirror decays for which the parameters involved in the Ft values are accurately known would enable to reach a precision on V ud equivalent to the current precision obtained from the usual pure F transitions. In the LPCTrap setup, the radioactive ions are confined in a transparent Paul trap and the correlation parameter, a, is precisely inferred from the time of flight of the low energy recoiling ions detected in coincidence with the beta particles. Moreover, the recoil ion spectrometer gives access to the charge state distributions of daughter ions stemming from 1+ ions decay. The setup allows a very strong control of the systematic effects. The 6 He pure GT decay is the first transition studied with LPCTrap [Cour13]. The last experiment was performed in The recoil spectrometer enabled to measure for the first time the charge state distributions of the recoiling 6 Li ions produced after the β decay of 6 He 1+ ions. An electron shake-off probability, P exp = (36), was deduced from the data [Cour12]. The value is in perfect agreement with simple quantum mechanical calculations, P th = , based on the sudden approximation, which has then been proved to be well suited for such a pure electron shake-off process. A preliminary value has been estimated for the angular correlation parameter, a= (26) stat, but the realistic simulations needed to extract a have still to be refined to carefully manage the relevant parameters of the experiment. The number of recorded events, , should enable to determine a with an unprecedented statistical accuracy of (0.45% in relative precision). As far as the 6 He decay is concerned, the team now contributes in a new experiment involving magneto-optical traps (MOT) and installed at CENPA, Seattle [Knec13]. Our main contribution is the development of the recoil ion detector and the installation of the acquisition system FASTER developed at LPC Caen (see section "Administration and technical departments, Instrumentation" of the present report). The goal of the experiment is to reach the precision level of 0.1% in the measurement of a. The apparent success of 6 He experiment has favoured the upgrade of LPCTrap to make it operational with masses heavier than 6 and, in June 2011, the setup has been commissioned with the 35 Ar 1+ beam delivered by SPIRAL. The 35 Ar nuclei decay through a mirror transition dominated by the F component (93%). A total of good events was recorded in 32 hours of data taking. This enabled to measure for the first time the charge state distribution of the recoiling 35 Cl ions [Cour13b], which is in excellent agreement with theoretical values (see table 1). Charge Expt. results Calculation with Auger > (1.07) (0.44) 5.71 (0.27) 1.58 (0.21) 0.71 (0.18) Calculation without Auger <0.002 Table 1: Experimental ion charge-state relative branching ratios (%) compared to calculations with and without Auger ionizations. This analysis has highlighted the important role of Auger processes in electronic rearrangement of such ions. The number of 35 Cl atoms produced during the experiment was deduced from the number of beta particles detected in singles and the overall absolute detection efficiency for ions. This estimate leads to 72(10)% of neutral 35 Cl recoils, which is also consistent with the 73.9% ratio obtained from the theoretical calculations. 42

47 Fig.1: Experimental time-of-flight distributions of 35 Cl n+ ions produced by the β decay of 35 Ar 1+ ions confined in the Paul trap. The real experiment with 35 Ar was performed in June The total efficiency of LPCTrap (transmission & trapping: 0.38%), has enabled to record real coincidences in one week (Fig.1) [Ban13]. This statistics should allow a determination of a with an absolute uncertainty of (~0.2% in relative precision). Assuming a systematic uncertainty of the same order than the statistical one, the final result would constitute the most precise value ever obtained in a betaneutrino angular correlation measurement. The analysis is ongoing [Fabi]. As far as the 35 Ar decay is concerned, the team was also involved in WITCH runs at CERN/ISOLDE [Brei12]. Again our main contribution was dedicated to the recoil ion detector and the installation of the acquisition system FASTER developed at LPC Caen. In September and October 2013, the commissioning run and a first experiment were performed with 19 Ne which mainly decays through a mirror transition to the ground state of 19 F (BR= (20)%). To get rid of the huge contamination of stable molecular ions with mass 19 in 1+ charge state coming from the ECS source of SPIRAL, the beam lines were tuned in 2+ charge state. This enabled to run the experiment, but with a loss of a factor of 3 in the RFQ efficiency. Nevertheless, even with this charge state, the contamination of the beam with 19 F 2+ remained important and saturated the RFQ, limiting the effective number of trapped radioactive ions. Finally a total number of coincidences were recorded during 4 days of data taking (Fig. 2), which remains reasonable considering that the half-life of 19 Ne is ten times larger than in the case of 35 Ar. This statistics will enable to determine the precise charge state distribution of the recoiling 19 F ions. A comparison between fig. 1 and 2 shows that the higher charge states are less favored for 19 F ions, which is consistent with a lower probability of Auger effects in the decay of 19 Ne 1+ ions, as predicted by the theoretical calculations [Lien12]. This first analysis will be completed in the coming months. The statistics collected during the experiment will also enable to analyze systematic effects in the determination of a in some extreme conditions linked to 19 Ne as the recoil maximum kinetic energy is the lowest in this case ~1.1x10 5 good events F + Counts F F 3+ 0 Counts/channel ToF (µs) Fig. 2: Raw experimental time-of-flight distributions of 19 F n+ ions produced by the β decay of 19 Ne 1+ ions confined in the Paul trap. As far as the 19 Ne decay is concerned, the team was also involved in a precise T 1/2 measurement performed at GANIL [Ujic12]. Here our main contribution was the installation of the acquisition system FASTER. In conclusion, we have collected a large amount of data concerning three different transitions. The next year will be dedicated to the complete analysis of these data, to extract precise values of the beta-neutrino angular correlation parameters in the three cases. This will also enable to design an upgraded LPCTrap setup, to continue weak interaction studies with the new beams expected in the upgrade of SPIRAL, first at LIRAT and later in the DESIR hall. 43

48 High resolution study of low energy charge exchange collisions with a MOT (magnetooptical trapped) target In collaboration with CIMAP (Caen) and CELIA (Bordeaux) Laser cooling and trapping of atomic samples in a MOT (Magneto-Optical Trap) is now a first step for many exciting and innovating experiments in atomic physics such as Bose-Einstein condensate formation and superfluidity studies, electromagnetically induced transparency, photoassociation, quantum information, etc Another possible application is the confinement of radioactive atoms produced by nuclear reactions. The cold cloud of trapped exotic atoms constitutes a very clean source to perform precision measurements in nuclear β decay: using 6 laser beams and a quadrupolar magnetic-field, the atoms are held in a small volume, almost at rest and in a high vacuum, which allows the detection in coincidence of the decay products using surrounding detectors [Knec13]. Moreover, they can be easily polarized using lasers for correlation measurements in the decay of polarized nuclei [Pitc09]. Having for final objective the installation of such a device on the future DESIR facility at SPIRAL2, the LPC has designed and built a MOT for stable rubidium atoms. With stable atoms, the nice properties of a MOT mentioned above can be used to provide a cold atomic target for the study of ion-atom collisions at low energy. In such experiments, the most effective technique is the recoil ion momentum spectroscopy (RIMS) [Dorn00]. It gives access to the main observables of the collision (the Q of reaction and projectile scattering angle) through the measurement of the ionized target recoil momentum. The momentum change due to the collision being very small (a fraction of a.u.), the resolution is usually limited by the temperature of the target. The Rb target provided by the MOT is a cloud of about 10 7 atoms confined in a 1 mm 3 volume at a temperature below 200 mk. Such a low temperature does not limit the resolution on the momentum measurement and the coupling of a Rb MOT with RIMS (called MOTRIMS) results in very high precision experiments. The MOTRIMS setup designed and built at the LPC is shown in the fig. 3. It includes transverse extraction of the recoil ions with a 3D focusing electrostatic spectrometer, and a fast switch-off of the trapping B-field during data counting [Blie08]. MOTRIMS can in principle be employed, as in conventional RIMS, to probe a multitude of scattering dynamics [Dorn00]. In a first step, we focused on single charge transfer in low energy Na Rb(5s,5p) collisions. The performances of the device allowed the detection of weakly populated charge transfer channel (contributing to less than 1% of the total cross sections), and provided accurate relative cross sections for the active channels, along with their associated distributions in projectile scattering angle. The high resolution on the scattering angle measurement (~50 mrad) has even enabled to resolve the predicted diffraction-like oscillations due to the limited range of impact parameters leading to charge exchange process. The results have been then used to test and refine molecular closecoupling (MOCC) calculations performed at the CELIA in Bordeaux with unprecedented precision. This joint theoretical/experimental study was published in 2012 [Lerr12]. Fig. 3: Schematic view of the setup (see text and ref. [4] for details) 44

49 Fig. 4: Schematic representation of the experiment. A homogeneous magnetic field B h defines the quantization direction (z axis), and optical pumping leads to magnetic sublevels of the target state with well defined hyperfine quantum numbers m F, depending on the handedness of the laser pulse. The initial and final states are therefore quantized in the (x, y, z) reference frame. The Na + ions impinge on the oriented target with velocity v and impact parameter b, and the scattered projectile distribution is characterized by the spherical angles (q,j) in the (x col, y col, z col ) scattering frame. In a second step, we have used the opportunity to prepare the target with lasers to investigate the case of charge exchange scattering between Na + ions and oriented Rb(5p ±1 ) atoms. It is known from previous studies on similar systems that the differential cross sections (DCS) in projectile scattering angle for charge exchange exhibit asymmetries related to the coherence of the capture process (Fig. 4). However, most of these previous studies called for more precise measurements in order to reveal the exact angular dependence of the DCS and associated coherence parameters. In this respect, Na + + Rb collisions are particularly challenging since the projectiles are scattered in very forward directions [Lerr12]. To apply the experimental technique to the case of oriented 5p states, several improvements of the setup have been achieved. An additional laser, with circular polarized light (right or left) could be shined on the trapped atoms to prepare the target in a (5 2 p 3/2, F=3, m F =+3) state or a (5 2 p 3/2, F=3, m F =-3) state when the trapping magnetic-field is switched-off. Helmholtz coils were also added to the setup to provide a weak (4 Gauss), constant and homogeneous magnetic-field that defines a vertical polarization axis. Finally, two different diagnostics were developed to measure the polarization efficiency and the polarization rate. We found that more than 95% of the atoms were oriented within a time interval shorter than 5 µs. Experiments were performed with oriented (5 2 p 3/2, F=3, m F =±3) Rb(5p ±1 ) targets at E=5, 2, and 1 kev. The capture channels of interest Na + + Rb(5p ±1 ) à Na(nl) + Rb + were easily identified and selected using the recoil-ion-momentum component parallel to the projectile beam axis. Precise DCS in projectile scattering angles θ and ϕ (Fig. 3) were then derived from the transverse momentum components. We present in Figs. 5 (a - f) the weighted DCS sin(θ)s p+1à3p (θ,j) associated with the principal Na + + Rb(5p ±1 ) to Na(3p) + Rb charge exchange reaction, in terms of its four main contributions to which we refer to as left (ϕ=0), up (ϕ=π/2), right (ϕ=π), and down (ϕ=3π/2) (Fig. 4). These contributions are displayed as functions of Eθ, which is related to the impact parameter b. We observe in Fig. 5 that the up and down contributions to the DCS are symmetric, whereas the left and right ones exhibit strong asymmetry. As may be seen from Fig. 3, the rotation of the electron flow inherent to the initial oriented state breaks the symmetry of left (y>0) and right (y<0) scatterings while it preserves the up-down symmetry because of reflection symmetry with respect to the (x, y) plane. To proceed more quantitatively, the quantum-mechanical origin of the asymmetry has been investigated using adequate MOCC calculations (Fig. 5). The asymmetries in the DCSs, observed here with unprecedented resolution, were then used to access to the related coherence properties of the capture process in the different charge exchange channels. This work has enabled not only the theoretical description to be improved but also marked out the limits of the single-active-electron and frozen-core approximations. More details about this work can be found in [Lerr12b]. Fig. 5: Weighted DCSs for the charge exchange reaction Na + + Rb(5p +1 ) à Na(3p) + Rb + at E=1 [(a),(b)], 2 [(c),(d)], and 5 [(e),(f)] kev, as functions of Eθ. The histograms are the measurements, while the continuous (red) lines correspond to MOCC calculations. 45

50 Towards a new measurement of the neutron Electric Dipole Moment (EDM) In collaboration with: PTB (Berlin, Germany), LPC (Caen, France), JUC (Crakow, Poland), HNI (Crakow, Poland), JINR (Dubna, Russia), LPSC (Grenoble, France), University of Kentucky (Lexington, UK), KUL (Leuven, Belgium), CSNSM (Orsay, France), Sussex University (Brighton, SU), PSI (Villigen, Switzerland) and ETH (Zürich, Switzerland). Motivations Searches for permanent electric dipole moment (EDM) of particles are considered to be among the most important particle physics experiments at low energy since a non-zero value may reveal non-standard sources of CP violation and physics beyond the standard model (SM). More than 30 experiments are currently running or planned worldwide aiming at measuring the EDM of fundamental particles or systems such as electrons, neutrons, muons, atoms, molecules, etc... Beside the possible implications on the SM, the discovery of new CP violation sources is also required to explain the baryon asymmetry of the universe (BAU) [Sak67]. In this context, searches for the neutron EDM (nedm) have been started in the 50 s and have been pursued over more than 60 years lowering the nedm upper precision limit down to < e cm (90% CL) [Bak06]. The δ induced nedm predicted by the SM is 10-32±1 e cm while the SM extensions predict a neutron EDM in the range of e cm. Therefore, the next generation of neutron EDM experiments should be able to validate or exclude such models. The current limitation on the measurement precision is statistical. As a result, all the new nedm projects (7 over the world) are coupled to the building of high intensity ultra-cold neutron (UCN) sources. At the Paul Sherrer Institute (PSI) in Switzerland, the nedm experiment is taking place close to a spallation-induced UCN source which has recently been launched. Very first UCN were delivered in December 2010 and the UCN source commissioning was started in nedm experiment status During the last two years, UCN beam time has been shared between UCN source studies, nedm spectrometer tuning and nedm data taking. The experiment status is the following. Despite an increase of 40% since 2011, the UCN flux produced by the PSI source remains 15 times lower than initially anticipated. Reasons are not fully understood and studies are still ongoing in order to recover the lack of UCN production. On the other hand, the spectrometer is basically working. The current statistical sensitivity is e cm per day i.e. the best sensitivity achieved so far with the apparatus. However, the apparatus reliability is not optimal and will be further improved in order to increase the number of data taking days per year. From 2012 and 2013, the integrated statistical sensitivity is at best equal to e cm. The analysis is still ongoing and is part of V. Hélaine s thesis. Systematic errors are quoted at the level of e cm i.e. well below the statistical uncertainty. In these conditions, continuing the data taking for 3 more years will result in a new nedm measurement at a level comparable to the present limit but with a better control of systematic effects. If the full factor 15 in the UCN source intensity is recovered, we expect to improve the limit by a factor 4. In order to significantly improve the statistical sensitivity, a new spectrometer (n2edm) is under study. A statistical precision 5 times better than the former one is foreseen. Combined with the current UCN source performances, a limit of e cm could be achieved after 4-5 years of data taking, i.e. in about 10 years. Assuming the UCN source will reach its nominal performances, the e cm range will start to be explored. Beside the contribution to the running of the experiment itself, the LPC Caen is in charge of the development of a new spin analyzing system, the UCN detection and the 3D magnetic field measurements. Set-up of a new U-shape Simultaneous Spin Analyser (USSA) Based on GEANT4 simulations, a U-shape simultaneous spin analyser system has been built and tested at PSI (see the left panel of Fig. 6). The aim of such a new device is to perform UCN counting and to be able to simultaneously measure both UCN spin components. It will replace the current sequential spin analyser which induces UCN losses and depolarizations. A second NANOSC detector and a dedicated FASTER acquisition have been developed for the USSA. This project is part of the V. Hélaine s thesis. Fig. 6: left panel: picture of the simultaneous spin analyser (USSA); right panel: neutron frequency measurement performed with the USSA. 46

51 The USSA has been tested and compared to the current system for standard nedm runs. A preliminary measurement of the nedm statistical sensitivity is very promising showing an increase of 18.2(6.1)% with respect to the sequential device. As a result, the USSA has been installed below the nedm spectrometer for the future nedm data taking. Such results have to be further confirmed in Development of a new UCN gas detector Investigations about a novel UCN gas detector have been started. Two approaches are under study: a Micro-Pattern Gaseous Detectors (MPGD) based on the GEM (Gas Electron Multiplier) technology and a scintillating gas detector using one or two PhotoMultipliers Tubes (PMT). The aims are three fold: decrease the background sensitivity, increase the detection efficiency and handle large counting rates up to10 6 counts/s. A generic detector chamber has been built (see the left panel of Fig. 7). The vacuum tightness has been tested and a pressure limit down to mbar has been reached. The GEM version of the detector has been characterized using an alpha source for two gases: ArCO 2 and CF 4. A Maximum pulse duration of 150 ns has been measured, which fulfills the counting rate requirement. No discharge was observed for gains up to The scintillating version of the detector has been tested with a CF 4 / 4 He gas mixture with a partial 4 He pressure varying from 0% up to 10%. Using an alpha source and one photomultiplier, about 50 photons have been collected for a deposited energy of 6 MeV. The next step is the design of a new detector with two PMTs for the light readout. The goals are two fold: increase the photon collection efficiency and suppress the background by coincidence counting technique. Magnetic field mapping A precise knowledge of the 3D magnetic field inside the nedm apparatus is of crucial importance for correcting some systematics effects (see the previous progress report for more details). In order to measure the field components, we use either a 3D fluxgate or a vector caesium magnetometer. They are positioned on a new mapping device designed and built at LPC as shown in Fig. 8. To avoid field induced by eddy currents, this mapper is made almost fully out of non-conducting materials (PEEK, POM and ceramics). This device allows mapping a cylindrical volume by moving the probe inside the nedm vacuum tank (radial, vertical and azimuthal motions). Special care has been taken to ensure mechanical reproducibility. A mapping campaign has been performed during the winter 2013 with this device. One of the measured maps is shown in Fig. 8 where a magnetic anomaly is seen at the front left side of the nedm spectrometer. Further maps will be performed in 2014 looking for such magnetic pollution in order to remove them. Fig. 7: left panel: detector chamber picture; right panel: identification chart for the scintillating version of the detector. Fig. 8: left panel: picture of the new mapper into the vacuum chamber; right panel: field map of the spectrometer inside. 47

52 Search for neutrinoless double beta decay Collaboration NEMO3/SuperNEMO LPC (Caen, France), IPHC (Strasbourg, France), LAL (Orsay, France), Idaho National Laboratory (Idaho Falls, U.S.A.), ITEP (Moscow, Russia), UCL (London, UK), University of Manchester (Manchester, UK), JINR (Dubna, Russia), CPPM (Marseille, France), CENBG (Gradignan, France), LAPP (Annecy-le-Vieux, France), IEAP (Prague, Czech Republic), University of Texas (Austin, U.S.A.), LSM (Modane, France), University of Warwick Coventry, (UK), Osaka University (Osaka, Japan), Saga University (Saga, Japan), FMFI (Bratislava, Slovakia), LSCE (Gif-sur-Yvette, France), Imperial College (London,UK), Institut Universitaire de France (Paris, France), Jyväskylä University (Jyväskylä, Finlande), MHC (South Hadley, Massachusetts, U.S.A.), Institute for Nuclear Research (Kyiv, Ukraine), Charles University (Prague, Czech Republic). Motivations Many extensions of the Standard Model provide a natural framework for neutrino masses and lepton number violation. In particular the see-saw mechanism, which requires the existence of a Majorana neutrino, naturally explains the smallness of neutrino masses. The existence of Majorana neutrinos would also provide a natural framework for the leptogenesis process which could explain the observed baryon-antibaryon asymmetry in the Universe. The observation of neutrinoless double-β decay (0νββ) would prove that neutrinos are Majorana particles and that lepton number is not conserved. The isotopes for which a single-β is energetically forbidden or strongly suppressed are most suitable for the search of this rare radioactive process. The experimental signature of 0νββ decays is the emission of two electrons with total energy (E TOT ) equal to the Q- value of the decay (Q ββ ). The NEMO-3/SuperNEMO international collaboration has maintained an experimental program of research of the (0νββ) process for about 20 years. Currently, the LPC NEMO group is involved in two projects: the NEMO-3 experiment and the SuperNEMO project. NEMO3 final analysis After 8 years of data collection, the NEMO-3 detector has been stopped in february 2011 and dismantled at the Modane Underground Laboratory (LSM, Fig. 9). The NEMO collaboration now does the final analysis with the full statistics: ~35 kg y of 100 Mo and ~4.5 kg y of 82 Se. From 2007, the NEMO group at LPC Caen has been in charge of the quality survey of the NEMO-3 calorimeter energy calibration using the laser system. This task has been completed in 2012 and the final set of quality parameters has been delivered to the collaboration, corresponding to the individual survey of 2034 photomultiplier tubes (PMT) from 2003 to This deliverable is now used by the collaborators responsible of the analysis. This work is critical because the search for the (0νββ) process is very sensitive to the stability of the energy measurement. During 8 years of running, the majority of the 1940 PMTs have shown a very good stability (<1%) of their gain. However, a few percents of the PMTs had been observed with gain fluctuations larger than what was acceptable (>2%). The laser survey system has been used to identify the PMTs with such a problematic behaviour. This approach allows to reject these PMTs from the analysis and leads to elaborate a safe dataset, particularly for the search of the (0νββ) process at high energy, where one wants to achieve the best signal/background ratio. The limit obtained on the half-life of the (0νββ) process for 100 Mo is T 1/2 (0nbb) > y [Bong13]. This is the best result ever obtained for this ββ isotope. In 2013, the group has been actively involved in the writing of the publication dedicated to the final NEMO3 results of (0νββ) with 100 Mo. Fig. 9: Left: the NEMO-3 detector within the radon-free tent at LSM (2004). Right: a neutrinoless double beta decay candidate event in the NEMO-3 detector. 48

53 Fig. 10: The E TOT distribution for 100 Mo in the NEMO-3 detector after 34.7 kg y exposure. In the [ ] MeV range (Q bb =3.034 MeV), 15 events have been observed. Low radioactivity measurements, dedicated analysis and Monte-Carlo simulations have been used to predict 18 background events. The NEMO-3 experiment shows no evidence of neutrinoless double beta decay. Fig. 11: Left: Exploded view of the SuperNEMO demonstrator module: the central planar source frame is surrounded by two tracking chambers (2034 open drift cells working in Geiger regime) and two calorimeter walls (520 optical modules with low-radioactivity 8 PMTs). Right: the SuperNEMO demonstrator will be installed in the LSM cavity in place of the NEMO-3 detector. SuperNEMO experiment construction status The SuperNEMO detector is the next generation experiment designed to search the neutrinoless double beta decay process at the y sensitivity level (sensitivity to the effective Majorana neutrino mass: ~50 mev). The design of the SuperNEMO experiment reuses and improves the NEMO-3 technology: it consists in 20 planar modules, each hosting about 5 kg of ββ enriched isotopes. After a R&D program from 2005 to 2011 in which the LPC Caen has been strongly involved (BiPo1 and Bipo2 prototype detectors for the measurement of the radioactivity of the source foils, DAQ development, analysis and simulation software, electronics development), the IN2P3 Scientific Council has validated a first phase of the project in 2011: the SuperNEMO demonstrator module [CSIN2P3]. This module is now in construction (Fig. 11). The data collection will start in the second semester of 2015 for 2.5 years. It will accomodate ~7 kg of 82 Se and should reach a sensitivity of T 1/2 (0νββ) ~ y. This intermediate step is necessary to prove that the NEMO technology will be able to reach the target sensitivity with 20 modules and a 100 kg source of 82 Se isotope. Several points will be addressed with the demonstrator: the calorimeter energy resolution should be validated at the level of 8% FWHM at 1 MeV, the source radiopurity should be measured at the level of 10 mbq/kg for 214 Bi and 2 mbq/kg for 208 Tl, the radon ( 222 Rn) contamination of the tracking chamber should be measured at the level of 0.15 mbq/m 3. The LPC Caen is involved in several working packages of the SuperNEMO project. 49

54 Readout and trigger electronics The LPC Caen NEMO group leader is the scientific coordinator of the SuperNEMO electronics work package. He is assisted by a senior engineer from LAL. This implies the management of five engineering and development teams: LPC Caen (FEAStraduction enterinert ASIC), LAL Orsay (calorimeter front end board and integration), University of Manchester (tracker front end board), University of Osaka (DAQ) and LAPP Annecy. The group is therefore strongly involved in the elaboration and design of the specifications of various core components of the SuperNEMO demonstrator front end electronics: the specifications and design of the tracker front end board's ASIC (FEAST). the specifications of the calorimeter front end board (with LAL) the trigger system and strategy (with LAL) the readout system and data format (with LAL) In addition, the LPC Caen participates to the specification and architecture design of: the DAQ system (with Osaka) the Control and Monitoring System (CMS, with LAPP) various interfaces (cabling, mechanics...) Several tasks have been achieved in the period: the hardware architecture and specifications of the front end electronics integration scheme has been finalized. This implies: 52 calorimeter front end boards for >700 channels, 57 tracker front end boards (>6000 channels), 6 control boards (CB), 6 crates with their custom backplanes, 1 trigger board (TB), dedicated unified bus and protocols, interfaces (DAQ, CMS, etc...), the final batch of 150 FEAST chips has been delivered, a new test bench had been produced to perform exhaustive tests of the FEAST ASIC in real readout conditions, the specifications of the trigger system for both calorimeter and tracker front end board, as well as for the control board is completed, parts of the specifications and design of the readout have been done. Software The NEMO group has also a strong involvement in the development of the SuperNEMO simulation and data processing off-line software. This task is organized in two main parts: the design and implementation of the generic multi purpose Bayeux C++ library and the development of the software tools that are specific to the SuperNEMO project (SuperNEMO demonstrator, BiPo3 detector): the Falaise C++ library. The Bayeux library A set of C++ library modules and companion applications has been designed to perform various core tasks of interest in the making of nuclear and particle physics simulation applications: these contributions have been packaged in the Bayeux library. This software package contains the following generic components in charge of: data modelling, generic serialization and advanced I/O system, object factories, data selection and data processing mechanisms(pipeline), numerical tools, generic geometry modelling (compatible with GDML/Geant4), generic vertex generation for Monte-Carlo inputs, generic event generation for Monte-Carlo inputs (radioactivy, ββ processes, etc...), generic electro-magnetic field modelling, generic Monte-Carlo simulation engine (based on Geant4). Bayeux has been designed with genericity in mind and thus can be used in the context of many different experimental setups (Fig. 12). It is now a rather mature and stable library. Fig. 12: A simple detection setup modeled with the Bayeux library and used through the Geant4 engine to simulate the emission of 1 MeV electrons from a radioactive source. This simulation framework is fully parameterized by a set of human friendly configuration files (geometry, vertex generator, event generator, management of Geant4 session), without the need to write a single line of C++ code. 50

55 Its geometry modeling engine and Geant4 based simulation engine are used not only by the SuperNEMO collaboration to simulate the SuperNEMO and BiPo3 detectors (see the Falaise library below) but also by some other experimental projects: simulation of high purity germanium detectors (HPGe) for low radioactivity measurements at LSM, simulation of the LPCTrap experiment. Some new projects have started to implement dedicated simulation tools using the core functionalities of Bayeux (LPCCaen/IM2NP collaboration for the prediction of soft error rate in microelectronics components and circuits). The use of the Bayeux library is also foreseen for educational purpose at the Université de Caen (nuclear physics labs and teaching activities related to radiation protection). The Falaise library The Falaise C++ library is a collection of software dedicated to the SuperNEMO project. It implements the official simulation and reconstruction tools for the SuperNEMO demonstrator module and the BiPo3 detector. This library uses Bayeux as its foundation library. It is still under development and is currently integrating the SuperNEMO prototype code written during the R&D phase of the SuperNEMO project. Falaise modules are: SuperNEMO/BiPo3 simulation tools and application Reconstruction tools and application Analysis tools Event display (Fig. 13) The final design and software integration of Falaise has started in 2013 and will continue during Fig. 13: A double-beta decay of 82 Se simulated from the source foil of the SuperNEMO module displayed by the SuperNEMO 3D visualization program. The tracks of the two electrons emitted from the source foil are shown. The response of the drift cells and the calorimeter blocks are simulated too. Magnetic coil Our group is also responsible of the SuperNEMO magnetic coil: its study, its design, and finally its implementation at the LSM Laboratory. Indeed, a magnetic field of about 20 to 35 Gauss along the vertical axis is needed to provide the charge recognition and reject background events. The former NEMO-3 coil works perfectly during data collection; it was simple, stable and robust. Moreover we could re-use most of its components, which reduces significantly the cost. Thus, except minor changes (shape, size, etc...), it was decided to use a similar design for the SuperNEMO coil; NEMO-3 feedback, test with a prototype build at LPC Caen and magnetic field simulations serving as support to validate the final demonstrator design. Two of the ten panels of the NEMO3 coils were delivered at LPC Caen on February All elements have been sorted, weighted, neatly cleaned, then reformed to built the reduced size prototype. It was thus a very useful tool to get it into one s head with mechanics and to validate the magnetic field computations. For his purpose, we use the code "Maentouch" developped by Gilles Quemener for the nedm experiment. This code, written in C++ in the ROOT software framework, is based on boundary element methods. Magnetic field measurements show that we have a good understanding and control of the magnetic field inside the coil (Fig. 14). In parallel, many tests have been done on the conductive contacts between copper bars. In fact, it must present a large mechanical robustness and an electrical resistivity as low as possible to avoid warm up of the detector. This prototype will now be used to test the SuperNEMO photomultiplier magnetic shieldings. 51

56 Fig. 14: On the left: geometry and discretization of the prototype within Maentouch software framework. On the right : Vertical component of magnetic field inside the prototype (lines: simulation for different iron magnetic permeability, red points: measurements). At present, the final mechanical design of the SuperNEMO coil is fixed. It will surround the entire detector (source foil, tracker and calorimeter), with a developed circumference of about 17 m and a height of about 3.5 m. It consists on about 200 turns of square copper wires of mm square section, held together by isolating delrin pieces. About forty panels of iron (10 mm thick) are use for back magnetic field. For detector access reasons, the coil will be split into panels connected together with copper contact tubes. The complete coil has a mass of about 9 tons (~3.5 tons of high radio purity copper and ~5.4 tons of ARMCO Iron). Note that like for all components of the facility, each element of the mechanical structure will be monitored, identified and listed in a database accessible to the collaboration. A sample will be systematically collected, radiopurity measured and archived. Miscellaneous responsibilities In parallel to scientific and technical tasks, the group is in charge of several community tools for the NEMO-3/SuperNEMO collaboration: the management of 12 mailing lists for 120 collaborators, the hosting of the collaboration's private Wiki-based web site, the hosting of the Subversion server for official software development, the responsibility of the NEMO-3/SuperNEMO public web site, the responsibility of the computing resources at CCIN2P3 for the collaboration. These tasks are carried out with the technical support of the LPC Caen IT service and the CCIN2P3. The group also participates actively to the SuperNEMO Institutional Board (IB) and Technical Board (TB). 52

57 References [Bak06] C. A. Baker et al., Phys. Rev. Lett. 97, (2006). [Ban13] G. Ban et al., Ann. Phys. (Berlin) 525 (2013) 576. [Blie08] J. Blieck et al., Rev. Sci. Instrum. 79, (2008) [Bong13] M. Bongrand, Latest NEMO-3 results and status of SuperNEMO, TAUP 2013 [Brei12] M. Breitenfeldt et al., IS433 experiment, CERN-INTC [Cour12] C. Couratin et al., Phys. Rev. Lett. 108 (2012) [Cour13] C. Couratin's thesis (2013). [Cour13b] C. Couratin et al., Phys. Rev. A 88 (2013) [CSIN2P3] SuperNEMO France, Searching for lepton number violation with the SuperNEMO experiment, Proposal for the participation of France in the construction of a preproduction prototype (demonstrator module) of the SuperNEMO double beta decay detector, Conseil Scientifique de l'in2p3 (2011) [Dorn00] R. Dörner et al., Phys. Rep. 330, 95 (2000) [Fabi] X. Fabian's thesis. [Knec13] A. Knecht et al., AIP Conf. Proc (2013) 636. [Lerr12] A.Lerrede et al., PRA 85, (2012) [Lerr12b] A.Lerrede et al., PRL 111, (2013) [Lien12] E. Liénard et al., Proposal E646S, GANIL PAC October [Pitc09] J.R.A. Pitcairn et al., Phys. Rev. C 79, (2009) [Sak67] A. Sakharov, JETP Letters, 5, (1967). [Ujic12] P. Ujic et al., Proposal E658S, GANIL PAC October Publications High resolution probe of coherence in low-energy charge exchange collisions with oriented targets Leredde A., Fléchard X., Cassimi A., Hennecart D., Pons B. Physical Review Letters 111 (2013) An endoscopic detector for ultracold neutrons Göltl L., Chowdhuri Z., Fertl M., Gray F., Henneck R. et al. European Physical Journal A: Hadrons and Nuclei 49 (2013) 9 Energy-dependent relative charge transfer cross sections of Cs+ + Rb(5s, 5p) Nguyen H., Brédy R., Fléchard X., DePaola B.D. Journal of Physics B 46 (2013) Precision measurements in nuclear β-decay with LPCTrap Ban G., Durand D., Fléchard X., Liénard E., Naviliat-Cuncic O. Annalen der Physik 525 (2013) Measurement of the transverse polarization of electrons emitted in free neutron decay Kozela A., Ban G., Białek A., Bodek K., Gorel P. et al. Physical Review C 85 (2012) Atomic-matter-wave diffraction evidenced in low-energy Na++Rb charge-exchange collisions Leredde A., Cassimi A., Fléchard X., Hennecart D., et al. Physical Review A 85 (2012) Undergraduate research opportunities in neutron activation analysis for local, regional and international students Landsberger S., Tipping T., Ezekoye O., Tamalis D., Lott V. et al. J. Radioanalytical Nuclear Chemistry 291 (2012) First Measurement of Pure Electron Shakeoff in the β Decay of Trapped 6 He + Ions Couratin C., Velten P., Fléchard X., Liénard E., Ban G. et al. Physical Review Letters 108 (2012) Electron shakeoff following the Β + decay of trapped 35 Ar + ions Couratin C., Fabian X., Fabre B., Pons B., Fléchard X. et al. Physical Review A 88 (2013)

58 ACTIVITÉS TECHNIQUES ET ADMINISTRATIVES Service administratif Bureau d études et mécanique Service électronique et microélectronique Service informatique Service instrumentation Documentation Qualité et soutien aux projets Hygiène et sécurité 54

59 SERVICE ADMINISTRATIF M. de Claverie (resp.), V. Devaux, A. Gontier, O. Guesnon, L. Lancien Service Administratif Composé de trois agents CNRS et d un agent universitaire, le service administratif assure un rôle d interface avec les tutelles de l unité (CNRS IN2P3, ENSICAEN, UCBN) ainsi qu un rôle d assistance et de conseil auprès de l ensemble des personnels du laboratoire. Le service traite tous les actes administratifs des domaines suivants : La gestion des personnels Le service effectue le suivi de tous les agents de l unité, ce qui se concrétise par divers actes administratifs (gestion des congés, gestion des compte-épargne temps, dossiers de carrière, avancement,etc ) ainsi que par la diffusion d informations réglementaires. De plus, le pôle administratif se charge de l accueil, de l aide à l installation des personnels non-permanents (13 doctorants, 25 à 30 stagiaires tous niveaux et 7 visiteurs étrangers par an) ainsi que de l instruction et du suivi des gratifications de stage et des demandes de recrutement des contractuels rémunérés sur les ressources propres de l unité. La gestion financière et la gestion des missions Le service assure l exécution du budget du laboratoire ainsi que le suivi des crédits. Il traite dans le respect des règles en vigueur une multitude d actes administratifs se traduisant par l engagement des dépenses (commandes, missions, achats carte bleue), la validation des ordres de missions et des réservations de billetterie et d hôtel, le calcul du montant des remboursements dû aux agents, la transmission des états de frais au paiement, la liquidation des factures, la gestion des immobilisations. Les crédits provenant du CNRS, de l ANR et de l Europe sont gérés sous GESLAB ; les crédits provenant du MESR, de la Région Basse-Normandie (1 en cours) et des contrats industriels (1 en cours), sont gérés sous SIFAC. La gestion des relations internationales Le LPC a plusieurs collaborations avec des organismes de recherche à l étranger via les accords de coopération IN2P3 (5 en 2012 et 3 en 2013), les programmes européens (2 contrats en cours) et les contrats ANR (1). Le service aide au montage des projets et collaborations, assure le suivi et la justification des contrats en partenariat avec le Service Partenariat et Valorisation et la Cellule Contrat de la Délégation Normandie. Enfin, il organise les conférences, colloques, workshop en collaboration avec les scientifiques. Le secrétariat Le service apporte une aide à la saisie et à la mise en forme de rapports, courriers, notes, en français ou en anglais. Il assure également des tâches plus générales comme l accueil téléphonique, le traitement et l acheminement des courriers ainsi que l approvisionnement en fournitures de bureau et le transport des colis en collaboration avec la plateforme ULISSE. Il gère également l annuaire du personnel, les badges d accès au site ainsi que les cartes de restauration. Services généraux Un agent CNRS est l interface avec le Service Technique Immobilier de l ENSICAEN. Il assure l entretien, l amélioration et l aménagement des locaux, la gestion et l entretien du parc automobiles (2 véhicules), la réception et la distribution des colis, les achats de proximité avec la carte achat. Il est également chargé de la mise aux normes du réseau électrique en collaboration avec une entreprise spécialisée et de la gestion des contrats de maintenance pour les équipements de l unité. 55

60 BUREAU D ÉTUDES ET MÉCANIQUE B. Bougard, P. Desrues, H. Franck de Préaumont, D. Goupillière, J. Lory, Y. Merrer (resp.), C. Pain Missions, Compétences, Moyens Le service mécanique est en charge de l étude, de la réalisation et de l intégration sur site des parties mécaniques des instruments scientifiques développés par le laboratoire. Il répond également aux demandes faites dans le cadre des activités d enseignements dispensées à l ENSI Caen et à l Université de Caen Basse Normandie. Le service est composé d un bureau d études (4 personnes) et d un atelier de fabrication (3 personnes). Le bureau d études dispose de moyens IAO/CAO (Catia V5, SmarTeam, Ansys 14) lui permettant d assurer la conception, les calculs et les dossiers de réalisation. L atelier est doté de moyens de fabrication conventionnels et numérisés ainsi que d un logiciel de CFAO (Mastercam V6). Il assure la réalisation et la mise au point des dispositifs conçus ainsi que leur montage sur sites expérimentaux. Nos moyens nous permettent la réalisation d ensembles de mécanique générale et de chaudronnerie ainsi que le polissage et le nettoyage des pièces. L atelier réalise la quasi-totalité des pièces et ensembles conçus dans la limite des capacités de nos machines. Les compétences Étude et conception mécanique. Simulation mécanique et thermique. Ingénierie mécanique, dossier industriel, suivi de réalisation. Fabrication mécanique et soudure. Montage, alignement et intégration sur site. Les moyens La conception CAO - logiciel Catia V5 La gestion de données techniques - logiciel SmarTeam Le calcul par éléments finis - logiciel Ansys 14 La CFAO - logiciel Mastercam Un centre d usinage à commande numérique. Un tour par apprentissage numérisé. Des machines outils conventionnelles et des moyens de soudage Des moyens de contrôles et de nettoyage. Le laboratoire est impliqué dans des projets dans le cadre de collaborations nationales et internationales, le service mécanique a en conséquence une bonne pratique du travail en mode projet et de la qualité. Les nombreux achats effectués tout au long de l année nécessitent la maîtrise des règles de consultations et d achats des différentes tutelles. Enfin, l utilisation régulière de la soustraitance nous permet d avoir une bonne connaissance du tissu industriel de notre domaine d activité. Réalisation, Montage et Alignement & Conception Mécanique CAO Catia V5 56

61 Projets et activités SPIRAL2 Phase 1 MONTAGE DES LIGNES ACCÉLÉRATRICES Au côté des équipes techniques du Ganil, montage et alignement des lignes accélératrices de la phase 1 du projet Spiral2. SPIRAL2 - Démonstrateur SHIRac2 (Refroidisseur quadrupôle à radiofréquence) Développement d un démonstrateur de refroidisseur de faisceau d ions à gaz. Étude et réalisation de l ensemble de la ligne sous vide et de ses équipements associés. SPIRAL2 RFQ-Cooler (Refroidisseur quadrupôle à radiofréquence pour Spiral 2) Étude du refroidisseur quadrupôle à radiofréquence pour la phase 2 du projet Spiral2. Cet ensemble répond aux exigences en termes de nucléarisation du projet (maintenance, maintien du confinement, ). SPIRAL2 IBE (Station d identification de faisceaux d ions radioactifs basse énergie) Étude d une station d identification de faisceaux radioactifs. Dispositif de détection composé d un dérouleur de bande sur laquelle les ions sont implantés et de détecteurs permettant la mesure en 2 points grâce à des détecteurs Germanium, des détecteurs Silicium et des scintillateurs solides. SPIRAL2 PTFI (Profileur Très Faible Intensité) Études et réalisation des parties mécaniques d un démonstrateur de profileur de mesure de faisceau de très faible intensité pour le projet Spiral2. LIRAT - EMILIE Dans le cadre d une collaboration avec le Ganil, étude et réalisation d un Debuncher destiné à être monté sur l installation Lirat au Ganil et sur le démonstrateur SHIRaC2 au LPC. Implantation et Alignement des châssis du LINAC Modèle CAO de l ensemble de la ligne Alignement et Montage de la ligne LME Ensemble du dispositif monté dans le hall du LPC RFQ-Cooler Modèle CAO du débuncher Etude et réalisation d ensembles dans le cadre de collaborations avec d autres laboratoires. Ensemble réalisé au LPC Réalisation d un piège à Ions pour l université King Khalid (Arabie Saoudite) Détecteur Ion de recul pour l Argonne National Laboratory (USA) 57

62 SUPER NEMO Pour la collaboration internationale, étude de la bobine du démonstrateur de SuperNemo qui sera installé au Laboratoire Souterrain de Modane. Etude et réalisation d un prototype de bobine permettant de caractériser et de valider les dimensions de la bobine du futur démonstrateur. Modèle CAO de la Bobine du démonstrateur de SuperNemo nedm (Le moment dipolaire électrique du neutron) Développement de dispositifs pour l installation nedm à l Institut Paul Sherrer (PSI) en Suisse. Étude et réalisation mécanique d ensembles de détection ainsi que de dispositifs de mesure associés. ARCHADE REC HADRON Pour le futur centre de recherche en hadronthérapie Archade, dans le cadre du projet Rec Hadron étude d une chambre à vide modulaire et de ses équipements associés (porte-cibles, plateau mobile, groupe de pompage ) MAPPER : Dispositif amagnétique de cartographie automatisé du champ magnétique à l intérieur de l enceinte FRAGMENTATION Etude et installation de dispositifs expérimentaux au Ganil. DOSION Réalisation de chambres à ionisation et intégration sur une ligne expérimentale au Ganil. FRACAS : Modèle CAO de l ensemble de la chambre à vide en Inox (Ø2m Lg 7m) FALSTAFF Etude, réalisation et montage d ensembles de détection destinés à mesurer les fragments de fissions induites par des neutrons dans le cadre du projet NFS. Travail réalisé en partenariat avec l IRFU et monté sur leurs installations à Saclay. FAZIA (Multi détecteur 4ϖ) Dans le cadre de la collaboration internationale, participation aux études du bloc de détection composé de détecteurs Silicium et de CsI. Etude de l assemblage de 12 blocs suivant différentes configurations et étude de leur intégration au sein d Indra. Réalisation d un prototype de support de bloc pour les expérimentations en cours. Détecteur Scintillateur réalisé au LPC Détecteur IC Dosion 3 Bilan Le service mécanique est engagé dans de nombreuses études et réalisations. La diversité des projets retenus par le laboratoire nécessite de la part des membres du service une adaptabilité de tous les jours. Les collaborations sont nombreuses et de plus en plus orientées à l internationale. L utilisation d outils de gestions de données techniques, de conduites de projets et d assurance qualité tend à se généraliser à tous les projets. Les fréquentes collaborations internationales nécessitent l usage de l anglais pour les membres du bureau d études et ceux de l atelier assurant les intégrations sur sites. Le service est très sollicité, son équipe et son organisation lui permettent de répondre aux multiples demandes des équipes de recherche du laboratoire. Modèle CAO de l ensemble de détection Dispositif monté à l IRFU à Saclay (juin 2013) Modèle CAO du dispositif de Mesure de résolution Intégration d un ensemble de 12 blocs à l intérieur de l enceinte Indra suivant une configuration «mur». 58

63 SERVICE ELECTRONIQUE ET MICROÉLECTRONIQUE F. Boumard, J.-F. Cam, S. Drouet, L. Fédor*, P. Laborie (resp.), J. Langlois, A. Leconte, L. Leterrier. * Contrat temporaire d avril 2012 à août Missions et Compétences Notre mission principale est de répondre aux besoins des expériences et projets de physique en concevant et en réalisant des systèmes dans les domaines de la microélectronique, l électronique analogique, l électronique de puissance et les radiofréquences. Cela implique de mener une veille technologique permanente, de construire des démonstrateurs et prototypes, d effectuer des tests. Nous avons une forte activité en R&D de manière à pouvoir rapidement nous adapter aux demandes à venir de la physique. Durant les deux dernières années, l activité du service a été principalement marquée par une implication forte dans deux projets majeurs : SPIRAL2 et SuperNEMO ; la croissance de la demande des physiciens en matière de RFQ coolers ; l avancée de notre R&D en préamplificateurs de charge, qui s est traduite par des applications et de nouvelles collaborations. Les paragraphes suivants donnent un aperçu de nos diverses activités. Activités liées à SPIRAL2 Le RFQ cooler en énergie des ions à la sortie du RFQ cooler a nécessité un long et délicat travail sur la conception et la validation de 2 détecteurs spécifiques. Les performances obtenues durant l année 2013 répondent aux exigences du cahier des charges en termes de transmission (>50 %) et d émittance ( 3 ϖ mm mrad). Concernant la dispersion en énergie ( 1 ev), les résultats se rapprochent très fortement de la valeur requise. Performances obtenues : Le démonstrateur SHIRaC au LPC CAEN Les travaux sur le RFQ cooler se partagent en deux projets : Le projet SHIRaC qui consiste à développer au laboratoire un démonstrateur permettant de valider les spécifications techniques du RFQ cooler conçu dans le projet RFQ Cooler de SPIRAL2. La construction du RFQ Cooler SHIRaC s est terminée fin De nombreuses campagnes de mesure ont été effectuées pour qualifier l instrument : mesures de transmissions, d émittances et de dispersions en énergie avec des faisceaux de différentes natures (en intensité et masse). Le démonstrateur a été travaillé pour améliorer sa fiabilité et ses performances en termes de manipulation de faisceau. La précision requise pour les mesures de dispersion 59

64 Intensité faisceau (na) Transmission (%) Emittance transverse (ϖ mm mrad) Dispersion en énergie lognitudinale (ev) 50 74,2 2,1 0, ,2 2,2 1, ,4 2,2 1,3 Résultats des mesures d émittance et de dispersion en énergie Concernant le projet RFQ cooler de SPIRAL2 Phase2, les travaux de conception du RFQ cooler adapté à l architecture du bâtiment de production de faisceaux radioactifs ont permis de rédiger la première version dossier de définition. L état d avancement de la conception de l instrument a été présenté lors d une revue de définition organisée par l équipe de direction du projet SPIRAL2. Les PTFI (Profileurs Très Faible Intensité) Un prototype d amplificateur d instrumentation à fort gain (46 db), large bande (50 MHz) et faible bruit (3 mv rms) a été réalisé en 2013 afin de traiter les signaux différentiels faibles niveaux (qq 100 µv) issus d un PTFI. Les tests de ce prototype avec un détecteur étant satisfaisants, la réalisation d une carte au format µtca incluant 5 voies amplificatrices a été lancée fin 2013 et devrait être testée début Le contrôle projet SPIRAL2 Comme présenté dans ce rapport d activité, le laboratoire est impliqué dans plusieurs tâches concernant SPIRAL2, future installation qui va étendre les capacités du GANIL en termes de faisceaux exotiques. Dans ce cadre, un ingénieur du service est contrôleur projet depuis Il est en charge du planning directeur du projet, de l organigramme des tâches et de la gestion des risques projet. Il est membre de la direction de ce projet. Prototype de l amplificateur Maquette 3D du RFQ cooler au LPC CAEN Carte µtca amplificatrice 5 voies Activités pour SuperNEMO Le circuit intégré FEAST (Front-End ASIC for SuperNEMO Tracker) a été conçu pour répondre aux exigences du système de tracking de SuperNEMO. Ce circuit intègre une électronique permettant de traiter jusqu à 54 voies. FEAST fournit une mesure de temps sur les signaux provenant des chambres à dérive avec un pas et une résolution inférieure à 15 ns. Cet ASIC a été conçu, routé et fabriqué en En 2012, 20 puces encapsulées ont été testées au LPC sur un banc de test générique qui nous a permis de valider les différentes caractéristiques comme par exemple : la résolution temporelle =3.62 ns rms, la non-linéarité différentielle=369 ps, la nonlinéarité intégrale =806 ps. A la suite de ces bons résultats, des tests sur 18 prototypes de cellules fonctionnant en régime Geiger ont été réalisés à Manchester (Royaume-Uni). Seuls les signaux anodiques ont pu être traités, les signaux cathodiques n étant pas câblés au niveau des détecteurs. Ces tests nous ont permis de : valider les fonctionnalités et les performances de FEAST monté sur des détecteurs Geiger avec des signaux anodiques, nous apercevoir que notre banc de test générique avait des problèmes de transmission de données. Courant 2013, un banc de test spécifique a été développé pour remplacer le précédent sur la base d une architecture plus simple. L interface utilisée est une interface USB et les premiers tests réalisés nous montrent que les transmissions s effectuent correctement quelque soit le flot de données. Aux vues des bonnes performances de FEAST, une série de 150 ASICs a été produite dans le but d équiper les cartes Front- End Board développées par nos collègues de Manchester et contribuant ainsi à la construction du prototype de SuperNEMO. Ces 150 ASICs seront testés et validés par un banc de test automatique dont la conception a débuté fin

65 Banc de test générique ASICs FEAST (boitier et puces nues) Production de 150 ASICs pour construction prototype SNEMO Banc de test spécifique USB Activités liées aux projets Emilie et Piperade Le projet PIPERADE (PIège de PEnning pour des ions RAdioactifs à DEsir) est piloté par le CENBG (Centre d Etudes Nucléaires de Bordeaux Gradignan). Celui-ci inclut un RFQ cooler buncher pour lequel il nous a été demandé d apporter notre expertise dans la conception du système de production des tensions Radio-Fréquences de l instrument. Le projet EMILIE (Enhanced Multi-Ionization of short-lived Isotopes at Eurisol) consiste entre autres à concevoir et expérimenter un RFQ debuncher. Notre mission est de concevoir et construire d une part le système de production des 2 tensions radiofréquences nécessaires à la production du champ de confinement du quadripôle et d autre part développer le générateur des tensions impulsionnelles appliquées aux 23 segments. La solution retenue pour générer les tensions de segments est composée de 2 générateurs arbitraires multivoies dont les signaux sont adaptés en amplitudes par 8 amplificateurs modulaires, les 8 tensions produites sont distribuées sur les 23 segments. Ce dispositif permet de produire les cycles de piégeages et d expulsion des ions stockés au centre du quadripôle. Le système de production des tensions RF est basé sur un circuit résonnant utilisant un condensateur variable haute tension et deux bobines à air, le couplage du circuit résonnant à l amplificateur de puissance est assuré par une boucle d induction. Banc de test RF du debuncher Vue de la carte CARAMEL Le choix des équipements nécessaires à la production des tensions de piégeages a été effectué, le circuit de distribution des tensions a été finalisé est pourra être implanté sur le RFQ debuncher. L instrument et ses équipements électroniques seront implantés sur le banc de test du démonstrateur du RFQ cooler de SPIRAL2. Les premiers tests devraient être effectués en début d année Activités pour DOSION La carte CARAMEL (CARte d Acquisition Multivoies Electromètres) est une carte d acquisition de faibles courants disposant de trois calibres : 3 pc, 6 pc et 12 pc. Elle peut intégrer des charges sur 12 bits à 10 µs ou 16 bits à 20 µs sur 16 voies simultanément. La gamme d entrée en acquisition de charge s étend de % de la valeur positive max de la plage de mesure (3, 6 ou 12 pc) à 3, 6 ou 12 pc selon le calibre. Elle dispose : de 32 entrées via un connecteur SAMTEC ERI8 d un connecteur Vita 57 LPC. La carte électromètre est une carte fille qui se connecte sur une carte mère FASTER. La carte mère fournira les alimentations, les horloges à la carte CARAMEL. La carte fille retournera à la carte mère les signaux numérisés. 61

66 Son implication dans DOSION avec un système constitué de deux cartes CARAMEL permet de pouvoir localiser le faisceau en X sur 32 voies, en Y sur 32 voies et aussi de quantifier l énergie collectée sur chaque voie X ou Y. R&D et contributions à la R&D du laboratoire R&D en préamplificateurs de charge intégrés Après une période d'étude et de conception, un ASIC a été soumis à la fonderie en janvier Cet ASIC est composé de 7 PAC (PréAmplificateurs de Charge) différents. Ces sept configurations nous ont permis de comparer les différentes structures en matière de performance. Les différences entre ces 7 PAC sont la résistance de contre-réaction (active ou passive), la polarité (unipolaire ou bipolaire), le gain (fixe ou configurable), et la connexion de substrat (commune avec la masse ou indépendante). Les tests faits en 2012 sur cet ASIC ont donné lieu à un rapport détaillé permettant la comparaison entre les différentes architectures en matière de bruit, linéarité, dynamique, charge équivalent de bruit, etc Carte de test de l ASIC C2SA Deux PAC ressortent de toutes ces comparaisons. Le premier est un PAC dont la polarité peut être choisie (soit unipolaire, soit bipolaire) et ayant une capacité de contreréaction de 1 pf. Ses principales caractéristiques : bruit large bande=265 µv (unipol.) et 545 µv (bipol.), linéarité de±1 % sur une dynamique de 1,8 pc, un CEB min=549 e - (unipol.) et 863 e - (bipol.). Le deuxième est un PAC dont la polarité peut être choisie (soit unipolaire, soit bipolaire), et ayant des capacités de contreréaction programmables de 0 à 15 pf par pas de 1 pf. Ses principales caractéristiques : bruit large bande=350 µv (unipol.) et 564 µv (bipol.), linéarité de ±0,7 % sur une dynamique de 1,7 pc (pour Cf=1 pf), un CEB min=841 e - (unipol.) et 1200 e - (bipol.). Layout de l ASIC C2SA Un test avec détecteur silicium a aussi été réalisé sur le premier PAC, ce qui nous a permis de mesurer sa résolution qui est de 24 kev LTMH avec une source tri-alpha et un shaping time de 0,5 µs. Dans la continuité de ce travail de R&D et en collaboration avec un collègue microélectronicien du GANIL, nous avons développé en 2013 un ASIC préamplificateur de charge configurable nommé C2SA pour Configurable Charge Sensitive Amplifier. Celui-ci a été conçu pour répondre aux besoins de la plupart des expériences de physique nucléaire. Les spécifications de ce PAC sont les suivantes : Gamme en énergie jusqu à 1,5 GeV, Taux de comptage < 250 khz, Polarité des signaux d entrée unipolaire ou bipolaire, Temps de montée des signaux d entrée >10 ns, Résolution de 10 kev Silicium à 5,5 MeV, Linéarité <1 %, Capacité d intégration sélectionnable de 1 pf à 42 pf, Résistance de contre-réaction sélectionnable de 280 kω à 5 MΩ. En plus de la fonction pré-amplification de charge, cet ASIC intègre : Un esclave I²C réalisé par l IPN Lyon pour le slow control, Une interface de secours en doublon de l esclave I²C, Un générateur d impulsions pour les tests/calibration (Amplitude réglable sur 10 bits), Un discriminateur à seuil sortant au standard LVDS (Seuil réglable sur 10 bits). La caractérisation de cette ASIC sera réalisée début

67 R&D en préamplificateurs de charge discrets Afin d étendre l offre sur l électronique frontale discrète du SEM, une R&D sur des PAC avec sortie temps a été menée en La particularité de ces PAC est d offrir, en plus du signal de charge, un signal impulsionnel très rapide homothétique en amplitude et minimisé en «walk» sur la gamme de fonctionnement. Le «walk» étant le déplacement temporel en fonction de la variation d amplitude du signal d entrée, il doit donc être minimisé afin de ne pas dégrader la mesure de temps de vol. Les principales caractéristiques de la version la plus performante sont : Sortie Charge : Bipolaire ; Plage de sortie ±2,6 V ; INL ±0,6 % ; CEBmin 570 e - pour Cf=2,6 pf et Rf=10 MΩ. Sortie temps : Signal impulsionnel négatif ; Temps de montée <2 ns ; Amplitude minimum -52 mv ; Walk 140 ps pour une dynamique d environ 30 ; Walk 60 ps pour une dynamique d environ 10. (Ces résultats sont donnés pour Cf=1 pf). Afin de finaliser la caractérisation de ces PAC, des mesures avec un détecteur seront effectuées en PAC avec sortie temps Carte de test de l ASIC Block_50 ps DNL voie 2 en ps DNL d une voie en ps DNL corrigée DNL DLL0 DNL DLL1 R&D sur la mesure de temps Bloc 50 ps : L équipe microélecronique du SEM, engagée dans la R&D sur la mesure du temps, a conçu fin 2009 un ASIC interpolateur de temps à 50 ps de pas quantification. Cet interpolateur met en œuvre une structure asservie innovante qui devrait permettre d atteindre une résolution temporelle d environ 30 ps. A cause de projets plus prioritaires, la caractérisation de cet ASIC a dû être mise en attente. En 2013, un banc de test a été réalisé et les premiers résultats sont très encourageants (DNL<±16 ps ; INL<±85 ps et Résolution <25 ps rms). Les tests seront finalisés en Résolution temporelle SCATS : SCATS (Sixteen Channel Absolute Time Stamper) est un ASIC marqueur de temps haute résolution, grande dynamique et fort débit réalisé en collaboration avec le LAL d Orsay pour l expérience SuperB. Cet ASIC comporte 16 voies de mesures de temps indépendantes permettant la mesure du temps de vol des particules ainsi que le marquage en temps des événements à des taux de comptage élevés (qq MHz/voie). Cet ASIC a été caractérisé en 2012 et les performances obtenues sont : DNL <±40 ps, INL <±110 ps et Résolution < 86 ps rms. En prenant en compte la courbe d INL, la résolution maximale a été diminuée à 65 ps rms. Dans un premier temps, SCATS a été encapsulé sous un boitier CQFP 120 au pas de 0,8 mm et afin d augmenter l intégration, le boitier est a été remplacé par un CQFP 128 au pas de 0,4 mm. Layout de SCATS Carte de test et boitiers de SCATS Contribution à la R&D sur les détecteurs diamants En vue d utiliser des détecteurs diamant segmentés double face pour la localisation spatiale de particules (diagnostic faisceau), le SEM a réalisé en 2013 une carte ETC (Eight Time Channels) au format VITAL57. Cette carte comporte 8 voies de mesure de temps/énergie réalisées à partir de deux ASICs : L ASIC NINO qui est un amplificateur/discriminateur multivoies conçu au CERN et l ASIC SCATS qui est un marqueur de temps haute résolution multivoies conçu par le LPC Caen et le LAL. Résolution temporelle entre 2 voies de mesure 63