Home

Beyond the Neutron Drip-Line: Superheavy Oxygen Isotopes

1. anode strip width 25 mm strip length 40 cm active area 800 cm number of gaps 2 4 gap size 0 3 mm glass plate thickness 1 0 mm HV Lycron Spray anode iron steel plate thickness 4mm anode strip distance 0 3 mm gas in between outer electrode thickness 2mm comments all 8 strips the same longer than active area and with triangular strip end all strips are coupled with transformator to readout results time resolution 108 ps problems with readout of right side efficiency voltage 97 12 0kV cross talk at n n 92 amplitude of crosstalk x16 a ee ee ee nl y 100 200 300 400 500 rate in Hz cm 153 180 170 timeresolution in ps bof EEE ee En En A 100 200 300 400 500 rate in Hz cm APPENDIX C DATA SHEETS OF MRPC PROTOTYPES prototype name GSI 5 tested during beam time HZDR 30 08 09 02 09 09 number of strips structured anode 8 anode strip width 25 mm strip length 40 cm active area 800 cm number of gaps 2 3 gap size 0 3 mm glass plate thickness 0 5 mm HV Europlex anode iron steel plate thickness 4mm anode strip distance 0 6 mm gas in between outer electrode thickness 2mm comments all 8 strips the same longer than active area and with triangular strip end all strips are coupled
2. OSH 10 T Otsuka T Suzuki J D Holt A Schwenk and Y Akaishi Three Body Forces and the Limit of Oxygen Isotopes Physical Review Letters 105 3 032501 July 2010 OST01 A Ozawa T Suzuki and I Tanihata Nuclear size and related topics Nuclear Physics A 693 32 62 October 2001 OUF 02 T Otsuka Y Utsuno R Fujimoto BA Brown M Honma and T Mizusaki Frontiers and challenges of nuclear shell model EUROPEAN PHYSICAL JOURNAL A 15 1 2 151 155 SEP OCT 2002 3rd International Conference on Exotic Nuclei and Atomic Masses HAMEENLINNA FINLAND JUL 02 07 2001 Pan12 V Panin Fully Exclusive Measurements of Quasi Free Single Nucleon Knock out Reactions in Inverse Kinematics PhD thesis Technische Universitat Darmstadt Germany 2012 PBB 07 A V Prokofiev J Blomgren O Bystrom C Ekstrom S Pomp U Tip pawan V Ziemann and M Osterlund The TSL neutron beam facility RA DIATION PROTECTION DOSIMETRY 126 1 4 18 22 2007 10th Inter national Symposium on Neutron Dosimetry Uppsala SWEDEN JUN 12 16 2006 PBL 12 P Pawlowski J Brzychczyk Y Leifels W Trautmann P Adrich T Au mann C O Bacri T Barczyk R Bassini S Bianchin C Boiano K Boret zky A Boudard A Chbihi J Cibor B Czech M De Napoli J E Ducret H Emling J D Frankland T Gorbinet M Hellstr m D Henzlova S Hlavac J Imme I Iori H Johansson K Kezzar S Kupny A Lafriakh A Le F vre E Le
3. Kernreaktionen und Nukleare Astrophysik Institut f r Kernphysik GSI Helmholtzzentrum f r Technische Universit t Darmstadt Schwerionenforschung GmbH Beyond the Neutron Drip Line Superheavy Oxygen Isotopes Vom Fachbereich Physik der Technischen Universit t Darmstadt zur Erlangung des Grades eines Doktors der Naturwissenschaften Dr rer nat genehmigte Dissertation von Dipl Phys Christoph Caesar geboren in Mainz Darmstadt 2012 D17 1 Berichterstatter Prof Dr Thomas Aumann 2 Berichterstatter Prof Dr Joachim Enders Tag der Einreichung 11 07 2012 Tag der m ndlichen Pr fung 29 10 2012 We have to remember that what we observe is not nature herself but nature exposed to our method of questioning WERNER HEISENBERG PHYSICS AND PHILOSOPHY THE REVOLUTION IN MODERN SCIENCE 1958 LECTURES DELIVERED AT UNIVERSITY OF ST ANDREWS SCOTLAND WINTER 1955 56 Abstract The neutron unbound ground states of 7 O and O have been investigated using the LAND R B setup at GSI in Darmstadt Germany Relativistic secondary cocktail beams of A Z 3 and Z lt 10 at approximately 450 MeV u have been produced using fragmentation of a primary stable Ar beam on a 4 g cm Be target and subsequent sep aration using the FRagment Separator FRS After reaching the LAND R B setup in Cave C one proton removal reactions i e X F 2 O X p 7 O 40 42n were inves tigated using various tar
4. 2 Therefore 4O is truly remarkable because it is hard to excite implying that it is doubly magic and very tightly bound But it is located at the very limits of nuclear existence as the addition of even a single neutron is not possible Jan09 The region in the nuclear chart surrounding the neutron rich oxygen isotopes is of par ticular interest from an experimental as well as from a theoretical point of view In the last 30 years huge experimental efforts have been undertaken to gain in sight into the nuclear structure of neutron rich oxygen isotopes In 1985 Langevin et al LQB 85 found that O is unbound As experimental technique in flight identifi cation of fragmentation products of a Ar beam impinging on a tantalum target has been used Five years later Guillemaud Mueller et al GJK 90 showed that 7 O is unbound as well The experiment was conducted using a Ca beam and a tantalum fragmentation target Finally in 1997 it was shown that even O a nucleus which in a simple shell model picture is doubly magic Z 8 N 20 is particle unstable TAA 97 From then it took 10 years until the first spectroscopy of a neutron rich oxygen isotope beyond the neutron drip line has been performed Hoffman et al HBB 08 found that 250 is unbound by 770 keV In 2009 it was experimentally shown that 74O is doubly magic Jan09 HBBt09 KNP 09 This experimental finding implicates a shift of the magic numbers as dis
5. 72 6 1 ANALYSIS OF THE O CHANNEL 20 m A A S a A a E S E E E E O S O S O S D O E E S S O E S B S E E E E A E experimental data R 350 40 n J ren Breit Wigner with E 737 keV 0 9 and T 72 keV folded with simulated response matrix 0 8 Efficiency x Acceptance counts 200ke V THEORY NN 3N 0 6 Efficiency x Acceptance NN 3N residual 8 E in MeV rel Figure 6 1 Shown is the experimental data of the relative energy for O as black solid line On top as blue dotted line the best fit Breit Wigner line shape with R 4 fm E 737 keV T 72 keV and A 0 is shown In addition the LAND efficiency x ac ceptance curve is shown to highlight that the observed peak is not produced by an acceptance cut At the top of the panel two different theoretical predictions obtained using chiral EFT including three body forces are shown 73 CHAPTER 6 RESULTS T in MeV T in MeV 09 0 9 1 E in MeV 1 E in MeV T in MeV T in MeV E in MeV T in MeV 100 0 9 E in MeV Figure 6 2 2 dimensional x spectra are shown for methods A to E applied to the 7 O data using a Breit Wigner line shape as described by formula 5 1 As red solid line the X2 m 1 contour is indicated in each spectrum the yellow horizontal and vertical lines represent the errors determined using this contour The exact values are given in table 6 1 74 6 2 ANALYSIS OF TH
6. GAB 92 GJK 90 GMSZ11 Gon06 HBB 08 B Eberlein Aufbruchreaktionen des Halo Kerns He PhD thesis Johannes Gutenberg Universitat Mainz Germany 1998 H G Essel and N Kurz GSI Multi Branch System User Manual http web docs gsi de mbs v43 manual gm_mbs _i pdf 2003 Z Elekes MRPC simulations private communication 2009 B Friman and A Schwenk Three body interactions in Fermi systems ArXiv e prints January 2011 H Geissel P Armbruster K H Behr A Br nle K Burkard M Chen H Folger B Franczak H Keller O Klepper B Langenbeck F Nickel E Pfeng M Pf tzner E Roeckl K Rykaczewski I Schall D Schardt C Scheidenberger K H Schmidt A Schr ter T Schwab K S mmerer M Weber G M nzenberg T Brohm H G Clerc M Fauerbach J J Gaimard A Grewe E Hanelt B Kn dler M Steiner B Voss J Weck enmann C Ziegler A Magel H Wollnik J P Dufour Y Fujita D J Vieira and B Sherrill The GSI projectile fragment separator FRS a ver satile magnetic system for relativistic heavy ions Nuclear Instruments and Methods in Physics Research Section B Beam Interactions with Materials and Atoms 70 1 4 286 297 1992 D Guillemaud Mueller J C Jacmart E Kashy A Latimier A C Mueller F Pougheon A Richard Y E Penionzhkevich A G Artuhk A V Be lozyorov S M Lukyanov R Anne P Bricault C Detraz M Lewitow icz Y Zhang Y S
7. 149 APPENDIX C DATA SHEETS OF MRPC PROTOTYPES prototype name GSL 1 tested during beam time HZDR 03 02 09 04 02 09 number of strips structured anode 8 anode strip width 25 mm strip length 40 cm active area 800 cm number of gaps 2 4 gap size 0 3 mm glass plate thickness 0 5 mm HV C layer on glass Graphite Spray OKS anode iron steel plate thickness 4mm anode strip distance 0 3 mm glue in between outer electrode thickness 2mm comments formator directly coupled to FEE four upper strips long triangular strip end which is not inside the active area coupled with trans four lower strips short rectangular shape two with transformator one with resistances and one pictures results Note at that time the HZDR DAQ could only read out 4 channels at the same time time resolution 85 ps efficiency voltage 98 12kV determined using scalers cross talk at n n x96 amplitude of crosstalk x54 efficiency liper eer er r r 5 HV in kV 150 prototype name GSI 2 tested during beam time HZDR 15 06 09 19 06 09 number of strips structured anode 8 anode strip width 25 mm strip length 40 cm active area 800 cm number of gaps 2 4 gap size 0 3 mm glass plate thickness 0 5 mm HV
8. B 4 TACQUILA channel starts counting The clock counters of channel 1 16 are stopped by channel 17 Channel 17 starts his own clock counter like the other channels but this is stopped by the trigger coming on GTB So clocki7 gives an idea about the delay between channel 17 and the readout trigger coming on the GTB bus The 17 channel performs a normal TAC measurement like the other 16 channels Having these individual measurements the timing of one channel can be calculated using counter ti t tac t taci7 4 B 1 clock frequency where t tac is tac converted from channels to times reference clock channel 1 channel 2 2 H H channel 17 lt gt tac_17 trigger on GTB Figure B 26 Shown is how the TacQuila timing works for the explanation see text The calibration converting channels to ns of the TAC measurement is at the moment done following a method suggested by N Kurz One records a white spectrum and gets a histogram like shown in the upper pad of figure B 27 To convert the channels to times one uses the information that the width of the spectrum is given by the clock frequency This is described by 1 y a clock frequency Ar t channel B 2 where A whole integral B integral up to certain channel The disadvantage of this method is th
9. Resistive Plate Chamber Parallel Plate Chamber Kernfysisch Versneller Instituut Helmholtz Zentrum Dresden Rossendorf TElectron Linac for beams with high Brilliance and low Emittance The Svedberg Laboratory Saha Institute of Nuclear Physics 13 CHAPTER 2 THE NEULAND TIME OF FLIGHT NEUTRON SPECTROMETER investigating different designs and readout schemes Z Ez E ZB Z ZZ Figure 2 2 Schematic layout of the experimental setup at the TSL neutron beam fa cility PBB 07 The neutron production system the proton beam dump the neutron collimation the position of the MEDLEY DABt00 and MRPC setup are shown A schematic layout of the experimental setup at the TSL neutron beam facility is shown in figure 2 2 A proton beam from the Gustaf Werner cyclotron was used to produce neutrons via the Li p n Be reaction Q 1 64 MeV The used lithium target was 23 5 mm thick The proton beam current on target was typically in the range of 280 370 nA The proton energy loss in the target amounts to 2 5 MeV Thus the energy of the primary proton beam of 179 0 0 8 MeV results in a peak energy for neutrons of 175 MeV The produced neutron beam is shaped by a system of collimators with a cylindrical and a conical design The collimator limits the maximum size of the beamspot at the detector position 11 m downstream of the target to a radius of 5 cm The remaining proton beam is bent into a beam dump tunn
10. BIBLIOGRAPHY BM69 Bor11 BR69 BR05 BRO06 BST12 CAB 12 Cer W Walus Two phonon giant resonances in 136Xe Pb and 2 U Physical Review C 68 2 024317 August 2003 A Bohr and B R Mottelson Nuclear Structure volume 1 W A Benjamin New York 1969 K Boretzky Technical Report for the Design Construction and Commission ing of NeuL AND The High Resolution Neutron Time of Flight Spectrometer for R3B private communication 2011 P R Bevington and D K Robinsons Data Reduction and Error Analysis for the Physical Sciences McGraw Hill New York 2 edition 1969 B A Brown and W A Richter Magic numbers in the neutron rich oxygen isotopes Physical Review C 72 5 057301 November 2005 B A Brown and W A Richter New USD Hamiltonians for the sd shell Physical Review C 74 3 034315 September 2006 T Baumann A Spyrou and M Thoennessen Nuclear structure experiments along the neutron drip line Reports on Progress in Physics 75 3 036301 March 2012 C Caesar T Aumann D Bemmerer K Boretzky Z Elekes D Gonzalez Diaz J Hehner M Heil M Kempe V Maroussov O Nusair R Reifarth D Rossi H Simon D Stach A Wagner D Yakorev and A Zilges Neu LAND MRPC based detector prototypes tested with fast neutrons Nuclear Instruments and Methods in Physics Research Section A Accelerators Spec trometers Detectors and Associated Equipmen
11. O to be bound and determine thereby wrongly the location CHAPTER 1 INTRODUCTION of the neutron drip line The neutron drip line is experimentally still only verified for elements up to oxy gen BST12 as shown in figure 1 1 In theoretical calculations the exact course of the neutron drip line is difficult to predict In a Nature article from 2007 BAB 07 it is e g stated that the experimental proof for the existence of two neutron rich nuclei Mg and A1 suggests neutron drip line slant towards heavier isotopes To test theoretical calculations nuclei with an extreme proton to neutron ratio are especially important because they put the most strict constraints on those theories Therefore it is very important to expand the study of nuclei even to nuclei beyond the neutron drip line BST12 The next paragraph will discuss why neutron rich oxygen isotopes are special from a nuclear structure point of view The Nuclear Shell Model developed by Goeppert Mayer May48 and Jensen JSH49 gives an intuitive picture of the structure of atomic nuclei In this picture each nucleon moves independently in a mean field which is created by all other nucleons From this model one gains analogously to the atomic shell model for the electrons single particle orbits These are the so called SPE s This basic picture of nuclear shell structure leads to the concept of magic numbers Z and or N 2 8 20 meaning that a shell closure
12. 0 24 E 24 6 4 26 0 23 F 4 0 4 26 0 23 Table 6 3 Given are the values obtained describing the high energy part of the 6O 2n channel using a flat test distribution having a width of 50 keV NDF is the number of bins in the fit range minus the number of free parameters in the input function The x distributions have been fitted in the range of the minimum using a polynomial of 2 4 order The minimum and error have been determined using the fit For 7 O the same type of theoretical calculations as for the 25O have been performed those are indicated in figure 6 6 Here the ground state as well as the first excited state are predicted 77 CHAPTER 6 RESULTS 0 100 200 300 400 500 600 700 800 900 1000 0 100 200 300 400 500 600 700 800 900 1000 center of orig distribution in keV center of orig distribution in keV ZOE 100 am x 2 BAPL 90 80 70 60 0 100 200 300 400 500 600 700 800 900 1000 0 100 200 300 400 500 600 700 800 900 1000 center of orig distribution in keV center of orig distribution in keV 2 ne a ee ee ee ee ee er 0 100 200 300 400 500 600 700 800 900 1000 0 100 200 300 400 500 600 700 800 900 1000 center of orig distribution in keV center of orig distribution in keV Figure 6 4 Shown is x for the different methods as a function of the central energy of the test input distribution Flat distributions with a width of 50 keV are c
13. 17 1 12 20 completely Ol OO OO SI D DW rR m oO Table 5 1 Shown are LAND channels which could not be used for the data analysis Plane 10 was read out using the new TacQuila electronics and is excluded from this analysis These paddles are switched off in the simulation In a second step the individual threshold of each channel has to be given to be able to match the experimental and the simulated data The thresholds are determined from the experimental data using the procedure described in Ros09 As further input parameter also the experimentally de termined light attenuation length of each paddle is used in the simulation This quantity is determined via the standard LAND calibration using the routine called cosmic1 The validity of the simulation is good if the simulated hit multiplicity distribution repro 58 5 1 DETECTOR RESPONSE LEG SIMULATION duces the experimental one A perfect agreement is found within error bars as can be seen in figure 5 1 045 TTT TTT Tt neutron hit multiplicity from simulation OO error from simulation neutron hit multiplicity from experiment error from experiment 0 gt N bin IN otal 0 35 gt io S N n iv 0 1 2 3 4 5 6 7 8 9 10 neutron hit mul Figure 5 1 Shown are the neutron hit multiplicities for a 1n channel once obtained from simulation and once using
14. Furthermore the beam has a microstructure with a repetition period of 45 ns and a pulse duration of 3 7 ns The uncertainty in the latter is the main contribution to the uncertainty in the measured neutron ToF For a visualization of the beam structure see figure 2 3 During the setup phase of the experiment two scintillators were mounted in front of the MRPC holding structure to characterize the beam see figure 2 2 For the final characterization of the prototypes these scintillators have been removed since the impinging neutrons would create charged particles contributing to the background Figure 2 4 shows the scheme of a 2 x 4 gap MRPC prototype built at GSI The readout electrodes are built from massive iron plates which serve as neutron converters at the same time A single ended readout was chosen hence only the middle electrode is structured dividing the detector in 8 strips A thin film of conductive coating on each outermost glass plate was used to apply the high voltage The chamber was operated at 12 kV E 100kV cm Each strip is Tonization Chamber Monitor tTime of Flight 15 CHAPTER 2 THE NEULAND TIME OF FLIGHT NEUTRON SPECTROMETER 2mm Figure 2 4 Schematic representation of the MRPC prototype not on scale Red iron light blue float glass yellow capton foil brown circles spacer fishing line white gas gap The central iron converter is divided into eight read out pick up strips as indicated by the vert
15. In figure B 3 it is schematically shown how a signal is processed on the LAND FEE Here a ideal case is shown in which the protective diode does not cut this happens at 700 mV the signal and the amplifier is not in saturation happens at 200 mV In the following the signal flow will be discussed in detail for one channel The first component on the board is a protective diode HBAT 5402 which in the original design was forseen to protect the amplifier from sparks In the next device Monolithic Amplifier GALI S66 the signal is inverted and amplified The output of the amplifier is then split to do a time and an energy measurement On the time branch the signal is routed to a comparator MAX9601 The volt age of this comparator is kept stable using a linear regulator LP3965EMP ADJ The comparator compares the input signal to a threshold If the input is larger than the threshold a PECL output signal is generated which is routed to the TacQuila and the TRIPLEX board On the energy branch the signal is given to a buffer amplifier MAX2471 which is here used to convert the input to two single ended outputs Normally this device expects a differential input however in this application the second input is connected to ground From these two outputs the negative is send to the TRIPLEX while the positive is routed to the piggyback QDC The gain of the amplifier is nominal 160 5 4 The FOPI FEE had three stages using this ampl
16. O ground state Those two facts make O a candidate for a so called true two neutron decaying nucleus a property which has been discussed in recent publications GMSZ11 PKGR12 and which is found for very few nuclei only 87 CHAPTER 7 DISCUSSION IIT IOI MT TAT ITT WITT TIT T 77 T a e 3 PILAIALIP PILI ALIP PL LDISAIPDALIAS SISA RUAA RRID PIG IA i a gt LEA AORPA LAL PL ARR LARR L AAA LP AOA D A AES ff tL PITA ot 4 Afi 26 LLRI ALASDAIR A 2 e oO FH CAR RAAH R LAHAR A S we oe 20 vie A LN QV 7777 Be oS 10 a KR Opt tp fi ftps film Aas PRR A E LP pI hp pI pI LT LA ESAS ETFS AAI DT HDS A bf pf do 2 Pi EIEN EDEN EEENNGSTESEEN NEST EEESNGSETESSPEES HD P a AL ph hh bt ft tht fit ttt tht f ti tt ti tt fp tl i tt bila a 1 LAPPAR SRRA RRIA RR ARAR AR PR AAA tities LAL AL A P LIDS fbf IARR IAR RAAR PLS EI AG R MULL ACA PPRA PRAA RR AARD AARD AARI AA At bp tf ff ARRAI tf fly Pod 16 Dt hp tl bhpl pi ft hp ft tit bgp filth bf hp pl fp bf bp ff bie d hihi ih ip RLAR bf pops VLLLLELLLLL LLL LLL LLL LGA p a SAL tft ttt th thst i it tt ti tli tt hift AARS ARAIA SL LA LMP A LALA LALA LALA LA LA LL LLL La LAA a NEISSE R ME dol AS Mb bt hp tt hI RAAPA RARR RAR AREIA ILL Ai pp bhp ppp tp lip pip tp bh pp pp tbl fj KI IT FA ae 5 RRA R A AAR AAA RIAA RIAA RA ARRAIA R AARO ARRA AAR A 10 Lip RARR R AARP LARR AAR RAARD AARP AAAA tf tf th tf i fA ge n EL AP LS pif htt bi pt htt bi RARA RAA REEL
17. One has to select a certain USB over IP server and connect it by using the corresponding button Now one can control the settings of the TRIPLEX cards via LabView Opening LabView one gets two windows shown in figure B 7 and B 13 The TRIPLEX control window should be closed The second one figure B 7 is the main TRIPLEX control window Here one sees a scheme of the TRIPLEX chain To start the TRIPLEX control one has to push the arrow in the left upper corner which makes the program run The next step is to initialize the control using the INIT button in the middle of the LabView window For each card a window will open and close again showing that this card is being initialized If the settings e g threshold values look strange meaning e g 123456 this means most likely that the communication between the TRIPLEX tree and LabView does not work One problem could be the device number see figure B 7 field labeled DevNumln at the left upper corner Having two TRIPLEX trees LAND and VETO the correct address has to be chosen here This address is assigned each time the devices are started therefore it is unfortunately not fixed and has to be checked each time To change the settings of a certain card one has to click the R W Read Write button of that card As a result a window like shown in figure B 13 will be opened Here one sees e CardNumber this number identifies the card The Address which is shown in the main TRIPL
18. S M Grimes A Haagsma J E Finck N Frank B Luther S Mosby T Nagi G F Peaslee A Schiller J Snyder A Spyrou M J Strongman and M Thoennessen Evidence for the ground state resonance of 2 O Phys Rev Lett 108 142503 Apr 2012 A Leistenschneider Entwicklung eines Schauererkennungsalgorithmus fiir den Neutronendetektor LAND Master s thesis Johann Wolfgang Goethe Universitat Frankfurt am Main Germany 1997 Y Leifels LAND shower algorithm private communication 2011 M Langevin E Quiniou M Bernas J Galin J C Jacmart F Naulin F Pougheon R Anne C D traz D Guerreau D Guillemaud Mueller and A C Mueller Production of neutron rich nuclei at the limits of particles stability by fragmentation of 44 MeV u 4 Ar projectiles Physics Letters B 150 71 74 January 1985 A M Lane and R G Thomas R Matrix Theory of Nuclear Reactions Reviews of Modern Physics 30 257 353 April 1958 168 BIBLIOGRAPHY May48 M G Mayer On Closed Shells in Nuclei Physical Review 74 235 239 August 1948 MJP 09 K Mahata H T Johansson S Paschalis H Simon and T Aumann Position reconstruction in large area scintillating fibre detectors Nuclear Instruments and Methods in Physics Research Section A Accelerators Spectrometers De tectors and Associated Equipment 608 2 331 335 2009 INC NuSTAR Collaboration Nustar proposals http www linux gsi de wwwnusta tech_report
19. and also supported by the estimated width using formula 5 6 84 oINyeU UBS oY JO you P ysour 1oAeMOY ST IY poyuoseid Jusws nswaUL y UL poArasqo 974S JL OYe4S 47 POHOXO ISIY OY IOJ UOP fe L 9784S POMIXI Oo OY SUTUIOOUOD sUOTJOIperd eoT JoIOSY AL AOA Ul USAIS OFe S N LA JTV OMRI OYJ UL PUNOJ sonfeA I9YJO se om Se YOM STUY UL pouregqo s Nsol fe Jo wnpuodwop T OIqeL 90ZA OLSI 6891 SZET 2l ggg 9819S 9X0 Ogz 9044 0015 9044 i 00S 66HINON 08 t4Ma7 og ST 977 ae gg 9784s PUNO Ogz 90ZA G IG 90ZA zoor 80 aaH 0g 341 05 022 E SyS orga Sres ayes punos Ogg gouelejol I i DU I J I J af SOING NN WF NE NN F I dry 1P 49 STUOUNS 18 42 HOH AIOOY quoutttodxe 1099 qyuoutttIodxe oye s S1970 YIOM STUY 85 CHAPTER 7 DISCUSSION 5 a 0 ground state Energy in MeV s 3 900 800 700 600 500 theo NN theo ep say ON FIN ve Hopp an 3 O lya Energy in MeV Cay n a Hep V206 400 600 b o ground state 400 200 the th ex Ly leo N Na ANN ay waders EN es 2500 2000 1500 1000 lra of uno on et Ti 5 u Z U Ser a se Vog oyn Br ion 06 12 9 2 c me first excited state theo theo exp Vor ON NN wel ag a Fax HN ges V206 a BRog Figure 7 1 Shown are th
20. e 1 445 e N t 4 78 e No 20 5 and formula 6 3 results in an upper limit for the lifetime of the 2 O ground state which amounts to 7 O GS N t 4 78 lt 5 7 ns We conclude that the lifetime of the 7 O ground state must be smaller than 5 7 ns 95 confidence level 82 Chapter 7 Discussion The results obtained in this work are summarized together with results from the litera ture in table 7 1 and figure 7 1 The ground states of 2O and O have also been recently observed at NSCL see ta ble 7 1 making the new measurement a necessary test of the previous experiments The obtained results for the O ground state is in very good agreement with the previous measurement HBB 08 The width T of the O ground state is by approximately a factor of two smaller than observed before however is still in agreement within error bars The single particle width for a l 2 state at E 737 keV is 65 keV this value is determined using a radius of 4 fm and formula 5 6 The calculated single particle width is in very good agreement with the experimentally measured one This can be understood as an experimental confirmation for the l 2 single particle nature of the state The found 6O ground state energy and the previous measurement LDK 12 are also consistent however the upper limit is by the new measurement reduced from 200 keV to 50 keV No excited state of 7 O has been observed before making the observed
21. is shown in figure 2 1 One of the major components amongst the developments is NeuLAND this detector is the successor of LAND which will be described in detail in section 3 3 2 In table 2 1 the design goals for NeuLAND are given as well as a comparison to the existing LAND Based on the experience with LAND during the past 20 years two different design concepts have been investigated during the R amp D phase of NeuLAND 1 MRPC T based detector concept converter based design 2 Scintillator based detector concept fully active design The fully active detector design based on a highly granular plastic scintillator concept has turned out to be superior to the MRPC concept It turned out that its additional calorimetric properties offer a significant advantage concerning multi neutron recogni tion Borl1 Facility for Antiproton and Ion Research Systeme de Production d Ions Radioactifs Acc l r s en Ligne in engl System for Producing Online Accelerated Radioactive Ions Facility for Rare Isotope Beams Michigan State University TLarge Area Neutron Detector A Large Acceptance DIpole magNet Reactions with Relativistic Radioactive Beams TG ALorimeter for the In Flight detection of y rays and light charged pArticles HGSI Large Acceptance Dipole gt Neu engl new Large Area Neutron Detector Successor of LAND TTNuclear STructure Astrophysics and Reactions Research amp Development tT TMulti gap Resisitv
22. lique R Anne C Borcea Z Dlouhy C Donzaud S Gr vy D Guillemaud Mueller M Lewitowicz S Lukyanov A C Mueller F Nowacki Y Oganessian N A Orr A N Ostrowski R D Page Y Penionzhkevich F Pougheon A Reed M G Saint Laurent W Schwab E Sokol O Sorlin W Trinder and J S Winfield Search for 280 and study of neutron rich nuclei near the N 20 shell closure Physics Letters B 409 64 70 February 1997 Michael Thoennessen and Bradley Sherrill From isotopes to the stars NA TURE 473 7345 25 26 MAY 5 2011 UOMH99 Y Utsuno T Otsuka T Mizusaki and M Honma Varying shell gap and Vog08 VZ06 Wam11 Wei YAB 11 deformation in N 20 unstable nuclei studied by the Monte Carlo shell model Physical Review C 60 5 054315 November 1999 E Vogt The Early Days of R Matrix Theory http www jinaweb org events azura08 talks Vogt The Early Days of R Matrix Theory pdf 2008 A Volya and V Zelevinsky Continuum shell model Physical Review C 74 6 064314 December 2006 F Wamers Quasi Free Scattering and One Proton Removal Reactions with the Proton Dripline Nucleus Ne at Relativistic Beam Energies PhD thesis Technische Universit t Darmstadt Germany 2011 H Weick Atima http www linux gsi de weick atima D Yakorev T Aumann D Bemmerer K Boretzky C Caesar M Ciobanu T Cowan Z Elekes M Elvers D Gonzalez Diaz R Hannaske J Hehner M Heil M
23. nine voltages six voltages plus three grounds the board is only supplied with six five voltages plus one ground The voltages 5 analogue and digital are both feed using the Graphical User Interface 129 APPENDIX B NEW LAND ELECTRONICS TACQUILA Figure B 37 Shown is a screen shot of the EPICS GUI for the TacQuila Iv modules same power supply see also table B 8 Furthermore all grounds are connected together For the distribution of the lv to the different devices two different schemes are available 1 A FOPI ly distribution board see figure B 39 is was used for tests 2 For the LAND TacQuila Crates a distribution system using WAGO Rail Mounted Terminal Blocks has been designed The 2 scheme will now be discussed in more detail From the power supplies a master distribution is feed also a sensing to this point is foreseen for the future The cables from the power supplies to the master distribution have a cross section of 16 mm Each TacQuila crate has a lv distribution which is used for the ten TacQuilas in the crate as well as the clock and the 17 channel distribution The connections from the master distribution to those sub distributions is done using cables with a cross section of 10 mm In the sub distributions each voltage is connected to ground several times using 470 uF capacitors to make the voltages more smooth Such a sub distribution is shown in figure B 40 In the shown picture all Rail Mounte
24. 1 200 400 601 801 1000 decimal value set in LabView Figure B 14 Shown is a graphical representation of table B 5 The TRIPLEX board has 2 sections which SSI Darmstadt both hold components Triplex2 SV KK 05 2010 Multiplicity m u7 Mere 9 D NUK PECL toLVTTL part U12 register 1 Figure B 15 On the left side the full TRIPLEX board is shown On the right side it is indicated where on the board one can measure the thresholds For each of the 16 channels there is one field with 4 components On the right upper leg of the left component red arrow one can measure the threshold before the DAC While on the left upper leg of the right part blue arrow the voltage how it is given to the FEE can be measured Furthermore the HEX switch the Multiplexer Part U11 and the two registers Part U12 and Part U13 are shown For details see text 110 B 3 TRIPLEX Digital Value Comparator Value in V TRIPLEX Value in V 0 0 72 4 12 100 0 7 4 200 0 53 3 04 300 0 36 2 07 400 0 19 1 1 500 0 02 0 13 600 0 15 0 83 700 0 31 1 8 800 0 48 2 75 900 0 65 3 72 1023 0 67 3 83 Table B 5 The left column shows TRIPLEX Control see figure B 13 the threshold settings entered via the LabView This is the value which is then used by the DAC on the TRIPLEX board to generate the threshold as seen by the comparator on the LAND FEE Th
25. 1 00 0 57 0 62 0 59 1 07 0 77 Table 2 2 Results of the strip wise efficiency determination Figure 2 8 shows a typical charge spectrum of one MRPC channel The black line represents the raw data For the spectrum represented by the red line a valid time entry is required The sharp rise is attributed to the TDC threshold indicated as a vertical red dashed line Figure 2 8 shows the strong influence of the threshold on the efficiency Different threshold settings for the individual strips are identified as the major contribution to the variations found for the extracted efficiencies as given in table 2 2 For the error estimation the following effects are taken into account e difference in efficiency between left and right readout of the same strip 10 e difference in efficiency between strips due to different effective threshold settings 40 e uncertainty in neutron flux given by n monitors 10 The different errors are added quadratically This leads to an efficiency of 0 77 0 33 19 CHAPTER 2 THE NEULAND TIME OF FLIGHT NEUTRON SPECTROMETER From GEANT 4 simulations an efficiency of 2 5 Ele09 was predicted for an MRPC prototype exposed to 175 MeV neutrons Since the efficiency of the prototype to charged particles is 100 this value is mainly reflecting the conversion of neutrons to charged particles However the simulation does not include experiment
26. 2n relative energy spectrum and the prediction for the first excited state This discrepancy is most likely found because the excited states which are compared do not have the same origin The experimentally found excited state is most likely not the first excited state The two facts that first the ground state of 7 O is more weakly bound than the 2 O ground state and second that the 2 O ground state resonance is rather long lived twelve orders of magnitude compared to 7 7 0 GS make O a true two neutron decaying nucleus a property which has been discussed in recent publications GMSZ11 PKGR12 and which is found for very few nuclei only ii Zusammenfassung In der hier vorgestellten Dissertation wurden die neutronen ungebundenen Grundzust nde von O und O untersucht Dazu wurde das LAND R B Experiment an der GSI in Darmstadt Germany benutzt Relativistische Sekund rstrahlen welche Kerne mit AJZ 3 und Z lt 10 enthielten wurden mit einer Energie von ungef hr 450 MeV u produziert Hierzu wurde Fragmentation eines Ar Prim rstrahls an einem 4 g cm Be Target und anschlie ende Separation im FRagment Separator FRS benutzt Vom FRS aus wurden die Strahlen zum experimentellen Aufbau in Cave C weitergeleitet Ein Protonen Knockout Reaktionen z B X F 60 X p 76O 40 2n wurden an ver schiedenen Targets untersucht Die Zerfallsprodukte der ungebundenen 7 O und O Systeme wurden unter Zuhilfenahme eines komplexen Systems au
27. 5 Using the resonance position E E yields the single particle width The value obtained from formula 5 6 depends strongly on the choice of the channel radius in contrast to the fit to the data which is as pointed out before insensitive to this param eter Vog08 In the literature the following values can be found Hoffman et al HBB 08 state The distribution was found to be insensitive to the size of the channel radius be tween 5 44 and 5 83 fm the origin of these particular values is not explained In the same paper a single particle decay width of 79 keV calculated for the l 2 ground state neutron decay of O at 770 keV is given To reproduce the width for the given reso nance position using formula 5 6 a channel radius of R 4 15 fm has to be used The matter radius of 24O is 3 19 0 13 fm OSTO1 Using the empirical relation to calculate the charge radius from the mass number A R x 1 2 AV 3 fm for O A 25 one obtains 3 5 fm To test the sensitivity of the Breit Wigner function on the channel radius the following parameters have been used e Peak position E 0 737 MeV e Width T 0 072 MeV e Fragment mass Mp 24 0205 u e Neutron mass M 1 0087 u for the O n system The radius has been varied between 3 5 and 6 fm As expected it is found that the line shape is not very sensitive to this value see figure 5 5 Considering this and all the different values found in literature and
28. Clock Distr LEMO FF A 0S 302 CLAK 57 e connector for cable from TRIPLEX to QDC ODU 525 060 035 040 010 e connector for lv cable Phoenix FKC 2 5 10 ST 5 08 RF e connector for 17 channel at SIS Clock Distr LEMO FGG 00 302 CLAD35 The different cable have the following length e GT B cable to next neighbor 15 cm e CLOCK cable from CLOCK Distr to TacQuila 90 cm e TRIPLEX cable to next neighbor 17 cm e TRIPLEX cable to rest e g over next neighbor etc 32 cm e 17 channel cable from SIS Clock Distr to TacQuila 100 cm 143 APPENDIX B NEW LAND ELECTRONICS TACQUILA B 9 Known Issues e Miscounting of clock cycle counter This effect was correlated with an offset for the charge measurement as long as the QDC was un gated this effect is now unfortunately gone The effect is rate dependend here the rate on the FEE is important not the trigger rate This effect is still under investigation and not understood An illustration of the problem is given in figure B 48 The amount of events which encounter this problem which was expected by the developers can be seen in figure B 50 e QDC is not reset after the readout e As a consequence of the above mentioned point the chosen resistance R is im portant since it defines the disintegration time we did some tests using different resistances but there is no final decision taken yet e Amplifier on FEE is in principle not needed if TacQuilas are
29. D en Figure 4 15 The relative angle between two neutrons seen from the breakup of He into an a and two neutrons obtained from experimental data using two different sets of L 1 L L L L 0 0 02 0 04 0 06 standard paramter 6 800 ps and R gt cm cy optimized paramter 6 200 ps and R 20 cm cy L L 0 08 0 1 0 12 0 14 relative angel in rad parameters for the shower algorithm is shown 54 4 2 NEUTRON TRACKING 4500 counts 4000 3500 3000 2500 2000 1500 1000 500 standard paramter 0 800 ps and R 729 cm optimized paramter 6 200 ps and Rro cm I I i I 0 02 0 04 0 06 0 08 0 1 0 12 0 14 relative angel in rad Figure 4 16 The relative angle between two neutrons seen from the breakup of He into an q and two neutrons obtained from simulations using LEG with two different sets of parameters for the shower algorithm is shown 59 CHAPTER 4 IDENTIFYING THE REACTION CHANNEL 56 Chapter 5 Analysis In the last chapter it has been discussed how to select the outgoing fragments and neu trons as well as how to determine their four momenta Using these quantities together with formula 3 6 the relative energy can be reconstructed The obtained energy spec trum is still affected by efficiency acceptance and response of the experimental appara tus These effects have to be accounted for to determine the real resona
30. Lyutostansky M V Zverev D Bazin and W D Schmidt Ott Particle stability of the isotopes 2O and Ne in the reaction 44 MeV nucleon Ca Ta Physical Review C 41 937 941 March 1990 L V Grigorenko I G Mukha C Scheidenberger and M V Zhukov Two neutron radioactivity and four nucleon emission from exotic nuclei Physical Review C 84 2 021303 August 2011 Gonz lez D az D Research and Developments on Timing RPCs Application to the ESTRELA Detector of the HADES Experiment at GSI PhD thesis University of Santiago de Compostela Spain 2006 C R Hoffman T Baumann D Bazin J Brown G Christian P A Dey oung J E Finck N Frank J Hinnefeld R Howes P Mears E Mosby S Mosby J Reith B Rizzo W F Rogers G Peaslee W A Peters A Schiller M J Scott S L Tabor M Thoennessen P J Voss and 166 BIBLIOGRAPHY T Williams Determination of the N 16 Shell Closure at the Oxygen Drip Line Physical Review Letters 100 15 152502 April 2008 HBB 09 C R Hoffman T Baumann D Bazin J Brown G Christian D H Denby P A Deyoung J E Finck N Frank J Hinnefeld S Mosby W A Peters W F Rogers A Schiller A Spyrou M J Scott S L Tabor M Thoen nessen and P Voss Evidence for a doubly magic 740 Physics Letters B 672 17 21 February 2009 HBB 11 C R Hoffman T Baumann J Brown P A Deyoung J E Finck N Frank J D Hinnefeld S Mosby W A
31. Sec B 3 2 and as backup The corresponding TRIPLEX cards have to be interconnected to deliver e g one LAND multiplicity To do so the TRIPLEX tree has been arranged in the way shown in figure B 22 If one wants e g change a TRIPLEX board due to maintenance one has to make sure that the new card has the right HEX address to not destroy the TRIPLEX tree For that reason the Hex Address is also written to each TacQuila system in the crate Furthermore the HEX Addresses are given several times in this document see figure B 21 B 22 B 23 and table B 9 115 APPENDIX B NEW LAND ELECTRONICS TACQUILA HAE MUX T6 26 1 27 HAG MUX 76 1 z H AD MUX 75 25 26 2 H A 7 MUX 77 2 Figure B 21 Ordering of all 40 TRIPLEX cards in the I C bus This is in principle the same scheme then shown in figure B 7 Each box represents one TRIPLEX card H A is the HEX address of the TRIPLEX card while MUX is the address of the multiplexer The two numbers given in the bottom line
32. Strip Detector Dicke engl thick Time of Flight wall Emitter Coupled Logic Experiment Electronic Department Effective Field Theory Electron Linac for beams with high Brilliance and low Emittance ECL NIM level converter embedded in a VME module Experimental Physics and Industrial Control System Effective Single Particle Energy Facility for Antiproton and Ion Research Fast Bus electronics LeCroy digitizers Front End Electronics 159 APPENDIX D ACRONYMS FLNR FOPI FPGA FRIB FRS GANIL GEANT GFI GLAD GND GRP GSI GS GTB GUI HADSHOPOMO HGS HIRe HV HZDR PC ICM IP ISOL JINR KVI LBNL LAND LEG LEVCON Imd lv LVTTL MBS MFP MRPC MSU Mul NeuLAND Flerov Laboratory of Nuclear Reactions FOur PI experiment Field Programmable Gate Array Facility for Rare Isotope Beams FRagment Seperator Grand Acc l rateur National d Ions Lourds GEometry ANd Tracking Gro er FIber detector german for large fibre detecor GSI Large Acceptance Dipole GrouND voltage connected to ground Glass fibre Reinforced Plastic Gesellschaft fiir Schwerlonenforschung Ground State Ger TeBus engl Device Bus Graphical User Interface HADes SHOwer POwer MOnitor Helmholtz Graduate School for Hadron and Ion Research High Voltage Helmholtz Zentrum Dresden Rossendorf Inter Integrated Circuit is a multi master serial single ended computer bus Ionization Chamber Monitor Internet Proto
33. again for being a office mate at TUD during the hard times while writing this document thanks a lot i guess this is the meaning of Geteiltes Leid ist halbes Leid My appreciations also go to all the members of the R B Collaboration out of those which have not been mentioned so far the following should be named in particular Dr Yuliya Aksyutina BIBLIOGRAPHY Leyla Atar Dr Aleksandra Kelic Heil Dr Tudi Le Bleis Sebastian Altstadt Bastian L her Alina Movsesyan Dr Olga Ershova Dr Deniz Savran Dr Klaus S mmerer G nter Ickert Dr Branislav Streicher Dr Gerhard Schrieder Dr Alexander Ignatov Dr Jonathan Taylor and all those which i unfortunately forgot to mention here The guidance and support received from all the members who contributed and who are contributing to this project was vital for the success I am grateful for their constant support and help Since at this point in live the institutional part of the education is completed i would also like to take a few lines to thank those who made a huge impact on that way Eric Endre should certainly be mentioned here helping with physics homework and exams during school and undergrad time in particular answering thousands of phone calls is appreciated a lot For the nice time at NSCL which brought me into nuclear physics i would like to thank the charge exchange group In particular Prof Dr Remco Zegers and Dr George Perdikakis Special thanks go al
34. board SAM5 Via GTB cables the readout of up to 60 TacQuila boards is done ENV1 is a 16 Channel NIM ECL NIM level translator packaged in VME module It converts in both directions between NIM and ECL logic families VULOM1 VME Universal LOgic Module is here used to do the dead time blocking Figure B 41 Shown is the LVTTL to PECL converter The device has to be supplied with 5 V Having one LVTTL input twelve PECL outputs are available 134 B 7 ADDITIONAL ELECTRONICS Figure B 42 Picture of the 40 MHz clock The device needs 5 V The clock can be distributed to up to 10 TacQuilas Scaler TRIDI Clock u large u kaa Gk lerne Figure B 43 Picture of the LEVCON It is indicated how this module is used in the LAND setup at the moments 135 APPENDIX B NEW LAND ELECTRONICS TACQUILA NI JADDAL Figure B 44 Shown is exemplary how the VME crate for a TacQuila readout system could look like For details see explanations in the text 136 B 7 ADDITIONAL ELECTRONICS In the following paragraph it will be discussed how the TacQuila VME crate has to be cabled the used numbers labels refer to figure B 44 In the common LAND DAQ we have four trigger those can be connected to the ENV see red box in figure B 44 The ENV delivers the triggers to the VULOM cable 1 which then does together with the TRIVAT I O interconnections cable 2 and cabl
35. channel It can clearly be seen that in the area in which in figure 4 12 the values drop left lower corner the probability increases in figure 4 14 As mentioned earlier the parameters have to be chosen such that as few 2n events as possible are shifted into the 1n channel Following this argument that shifting to a higher neutron multiplicity is acceptable but shifting events to lower multiplicity is forbidden one would simply take the smallest shower volume possible But this would lead to the effect that one uses each hit as a primary hit and gets therefore wrong primary hits meaning wrong parameters describing the neutrons To find the optimum parameters LAND Event Generator l CHAPTER 4 IDENTIFYING THE REACTION CHANNEL the following ratio was used P 2n 2n P 2n 1n Where P 2n 2n is the probability to detect a real 2n as 2n and P 2n 1n is the probability to detect a real 2n as In The ratio is shown in figure 4 13 P 2n 2n 0 5 10 15 20 25 30 35 40 Rey incm Figure 4 12 Probability to detect two neutrons for two generated neutrons P 2n 2n as function of the two parameters o and Rey used in the shower algorithm As a compromise between shifting as few events as possibly to the In but on the other hand also not to create fake primary hits R yj 20 0 cm and o 0 2 ns were used this point is also highlighted in figure 4 13 The full set of used parameters of the shower algorithm is shown in table 4
36. detector in the fragment branch is the TFW This detector is build out of 14 horizontal scintillator paddles in the first plane and a second plane having 18 vertical paddles Each horizontal paddle has the dimension 196 6x10 4x0 5 cm while the vertical paddles have a dimension of 154 6 10 4x0 5 cm3 All 32 paddles are read out using a PMT on each side The time AF as well as position of each hit is measured Having the ToF between target and TFW and knowing the length of the trajectory gives the velocity of the ion while the deposited energy determines the charge The neutrons are detected in LAND BEET 92 The characteristics of this detector are also given in table 2 1 The detector covers an area of 2x2 m and is 1 m deep It consists of 10 planes and every plane contains 20 paddles which have the dimensions of 200x10x10 cm3 The detection of the neutrons is based on the use of inactive converter materials in which the neutrons create charged particles via nuclear reactions Those secondary particels are then detected with plastic scintillators To not stop too many of the created secondary charged particels in the converter material itself the design of the detector is based on a sandwitch structure using thin iron layers as converter material GroBer FIber detector german for large fibre detecor Time of Flight Wall 34 3 4 DATA ACQUISITION One paddle has eleven iron and ten scintillator sheets of 5 mm thickne
37. experimental data Both histograms are normalized to one The results of the simulations for four important channels In 1n In 2n 2n 1n 2n 2n where the notation means generated tracked are shown in figure 5 2 and figure 5 3 respectively Projecting these two dimensional spectra over the full range onto the y axis results in the one dimensional efficiency x acceptance curve such as e g shown in figure 6 1 In the upper panel of figure 5 3 2n 1n two correlation regions can be seen For small energies the two neutrons interact in LAND so close to each other that they are sorted in the shower algorithm into one shower volume At large energies one of the neutrons has such a large angle that it does not hit LAND this is the so called acceptance cut In the one dimensional projection figure 6 1 the decrease of efficiency at approximately Ere 3 5 MeV can be seen reflecting the acceptance cut which depends on the neutron kinetic energy only The acceptance is 100 up to Eye 3 5 MeV for higher energies a fraction of neutrons does not hit LAND since a higher neutron kinetic energy means that the neutron has a larger angle relative to the fragment beam axis In the lower panel of figure 5 3 2n 2n it can be seen that the correlation bends for low energies away from the diagonal This reflects again the fact that neutrons which come too close to each other cannot be resolved The overall efficiency of LAND shown in figure 6 1
38. given above a channel radius of R 4 fm has been used during the analysis This relation is here only used to get a figure of merit 63 CHAPTER 5 ANALYSIS S0E 7 455 3 50 fi H R 4 00 fi 40 E R 4 50 f I 35 E R 5 00 f 4 H R 5 50 fi J 30 R 6 00 f H Breit Wigner with R 4 fm 25 E after folding with response matrix E 20E iS Fal 10E SE o5 i E y ET z 0 2 0 4 1 2 1 4 1 6 1 8 2 E in MeV rel Figure 5 5 Breit Wigner line shape as given in the one level approximation for fixed E and I The radius R has been changed between 3 5 and 6 0 fm It can be seen that the line shape is not very sensitive to the channel radius The dotted line represents the Breit Wigner line shape for R 4 fm folded with the response matrix shown in the upper panel of figure 5 2 It is shown that a substantial part of the experimentally observed width is due to the experimental response and not due to the original width of the state The folded function is normalized to the area of the original function 64 5 3 CHI SQUARE x AND LIKELIHOOD METHODS 5 3 Chi Square x and Likelihood Methods Having a test function on one side and the experimental data on the other side it has to be determined how good this test function describes the experimental data In an analysis framework like e g ROOT this is commonly done using standard x methods In general several different approaches are a
39. hier vorgestellten Ergebnisse die Schlu volgerung nahe dass die Neutronen sowohl im O n als auch im 40 2n System einen reinen d Wellen Charakter besitzen Die theoretischen Vorhersagen und die hier pr sentierten experimentellen Werte stim men f r die Lage des jeweiligen angeregten Zustandes in 7 O nicht berein Dies ist h chswahrscheinlich der Fall da es sich bei dem im Experiment gefundenen Zustand nicht um den ersten angeregten Zustand handelt Die zwei Tatsachen dass der Grundzustand von 7 O weniger stark gebunden ist als der von O und das zweitens der O Grundzustand eine eher lange Lebensdauer hat die Lebensdauer dieses Zustandes ist zw lf Gr enordnungen l nger als die des 2 O Grundzustandes machen 260 zu einem Kern welcher Wahren Zwei Neutronen Zerfall aufweisen k nnte Dies ist eine Eigenschaft welche in aktuellen Publikationen diskutiert wird GMSZ11 PKGR12 und welche nur f r sehr wenige Kerne auftritt Contents 1 Introduction 2 The NeuL AND Time of Flight Neutron Spectrometer 2 1 MRPC based Neutron Detector Concept 2 2 2 2 22 nn nenn 2 2 Readout Electronics for NeuLAND 2 2 2 En nenn 3 Experimental Method and Setup 3 1 Invariant Mass er 8202 I bb an sn 3 2 RIB Production GSI and FRS LK En nn nenn 3 3 LAND RB Setup at Cave see Se engen 3 3 1 Identification of Incoming Particles aoa aaa 3 3 2 Detection of the Reaction Products 2 2 aaa 34 Data Acquisition
40. in figure 3 3 The energy loss of an ion passing through matter depends following the Bethe Bloch formula on Z and 8 Using this fact the charge number Z is derived from a AE measurement using a PSP in front of the target The resulting two dimensional PID plot is shown in figure 3 4 and depicts that this method allows for a very clean identification of the incoming particles The ions of interest 6F and F can be chosen for further analysis using two dimensional cuts The beam spot on the target can be fine tuned using active slits ROLU This detector system consists of four movable plastic scintillators which are each read out with one PMT Every ion detected in this detector will be disregarded The incoming angle is measured using two DSSSD s in front of the target The coordinate system used in the analysis presented here is labeled in the following convention The z axis points in beam direction the x axis points to the left looking with the beam and the y axis Position Sensitive Pin diode Rechts Oben Links Unten german for right up left down Double Sided Silicon Strip Detector 31 CHAPTER 3 EXPERIMENTAL METHOD AND SETUP incoming Z TIT 10 a re N J 2 6 2 7 2 8 2 9 3 3 1 3 2 incoming A Z Figure 3 4 Incoming PID for the most neutron rich setting A Z 3 of the s393 exper iment The cuts on 2 F and 2 F are indicated as red ellipses points to the top One side of
41. keV 107 gt 156 A 29 42 722 a9 344 536 B 108 74 707725 203 20 155 67 742155 Pre D 91 67 110 2 0 85 74215 Var E 95 78 742733 gt Table A 1 Given are the results analyzing the 7 O data using a Breit Wigner line shape and methods A to E It can be seen that XA method A gives a very low value for the x at the minimum this is the case since for empty bins an error of 1 was assigned and this method is the only one using this large errors The reduced x is only given for method D this method uses a Poisson distribution for the errors taken from the test function which is the most reasonable assumption NDF is the number of bins in the fit range minus the number of free parameters in the input function 91 APPENDIX A FIT TO O DATA USING R 6 FM AND A gt gt 0 k p T in MeV T in MeV 0 9 1 E in MeV E in MeV 2 XBANL T in MeV 0 9 E in MeV Figure A 1 2 dimensional x spectra are shown for methods A to E applied to the 2 Q data using a Breit Wigner line shape as described by formula 5 1 As red solid line the oe 1 contour is indicated in each spectrum the yellow horizontal and vertical lines represent the errors determined using this contour The exact values are given in Tab A 1 92 Appendix B New LAND Electronics TacQuila The goal of this document is to introduce the user of NeuLAND to its new readout electronic
42. leads to enhanced stability because the energy gap to the next available shell is large In reality the location of a certain energy level depends on the occupation number of other energy levels This effect is included in the ESPE s which can explain changes of nuclear structure as a function of the A to Z ratio In figure 1 2 the ESPEs of neutrons are shown as a function of the proton number Z from oxygen to calcium for the N 20 isotones It can be seen that while for the stable nucleus Ca N 20 is a magic number the magic number is shifted for the unbound neutron rich 780 to N 16 This effect is due to the fact that for nuclei far from stability the highest occupied neutron v and proton 7 orbitals have very different quantum numbers If a stable nucleus has neutrons which occupy the vd3 2 shell there are also protons which fill the mds5 2 proton shell This very attractive md5 2 vd3 2 interaction is missing for the neutron rich oxygen isotopes This can be summarized citing the following sentence from Jensens 2009 Nature article Jan09 it seems that as soon as protons occupy the Ods 2 orbital as happens when going from O to F the gap between the neutron 151 2 and 0d3 2 shells decreases significantly an indication that a tensor force an especially attractive spin dependent force between protons and neutrons is providing the additional binding The following paragraph reviews briefly how the inclusion of three body fo
43. likely value 25 25 keV Prob Func determined using exp data 2 MeV lt E lt 7MeV most likely value 4250 MeV 7 8 Ea in MeV Figure 6 6 Shown is the relative energy of the 74O 2n system on the x axis and the probability that the given energy is the real energy of the state on the y axis For the ground state almost 80 probability are reached within the first bin first 50 keV For a excited state of 2 O the probability peaks at 4200 4250 keV The error is determined such that 68 one c of the area have to be under the curve The theoretical predictions are indicated at the top border of the panel for the ground state as well as the first excited state 80 6 3 LIFETIME ESTIMATES 6 3 Lifetime Estimates For a given angular momentum the decay width can be calculated as a function of the resonance position if a single particle structure is assumed e g done using formula 5 6 The connection between the lifetime of a decaying nucleus and the width of the state or resonance is given by rer 6 1 where the lifetime 7 is the time constant in the exponential decay law N t No e 6 2 For the 7 O ground state a width of T Ar keV has been observed Transform ing this into the liftime using formula 6 1 leads to 1 20 GS 978 10712 ns For 260 the width of the ground state resonance could unfortunately not be determined But using the ToF of the fragment from the target to the center o
44. method A gives a very low value for the x at the minimum this is the case since for empty bins an error of 1 was assigned and this method is the only one using these large errors The final result is determined using method D for that reason the reduced x is only given for this method Method D is chosen over the other methods since it uses a Poisson distribution for the errors taken from the test function which is the correct parent distribution Furthermore this method is least sensitive to the chosen binning of the experimental data as has been observed during the analysis The result obtained for the 7 O ground state is E caer keV and T fate keV method x at min E at min in keV T at min in keV 81 F273 A 22 71 707 te 228 167 B 54 93 arte Dr C 93 62 737 TE Par D 52 85 55 2 0 997 737 95 Fr E 57 22 768 24 I Table 6 1 Given are the results analyzing the 7 O data using a Breit Wigner line shape and methods A to E For method D the NDF is given in brackets which is the number of bins in the fit range minus the number of free parameters in the input function In figure 6 1 two different theoretical predictions for the O ground state are indi cated Those theoretical calculations have been performed by Holt et al Holl12 and Simonis et al Sim12 Chiral EFT including three body 3N forces have been used theory and experiment will be compared in chapter 7
45. of the In and the 2n channel are as described above treated together The results of these fits using method A to F are shown in figure 6 4 No real minimum can be determined in these spectra using finer steps for the chosen test func tions introduced statistical fluctuations The result was interpreted such that the lowest value gives the best description of the experimental data The error was determined via a linear interpolation between the two lowest points and using the value at y 1 For the high energy part of the 2n channel the group of events found between 2 and 7 MeV in this channel were analyzed separately Again a flat 50 keV wide distribution was used as test function The results are shown in figure 6 5 All methods show the expected behavior meaning that the x distribution around the minimum follows the shape of a parabola A fit using a polynomial of second order was done to describe the range close to the minimum indicated as red dashed lines in figure 6 5 Using the parameters of this fit the position of the minimum is used as peak position and the value at x 1 is used as error The low energy part of the In channel is described together with the low energy part of the 2n as explained before A second contribution here comes from real In events namely 2 O These two contributions are shown in the lower pad of figure 6 3 as a solid blue line 2n and a dashed purple line 1n To determine the final results of the analysis of t
46. ones The mode is selected via the LabView settings There is no QDC measurement for these events only the comparator will be fired by changing the threshold from its nominal voltage to 5V or 5V Doing this the threshold crosses the baseline and the comparator gets triggered The pulser input has to be a LVTTL signal Furthermore it has to be terminated using a lemo T and an additional 50 Q resistor Otherwise one can get problems with a floating baseline e Mul output signal which represents how many channels fired in the whole tree The width can be individually changed for each channel changing the setting of the corresponding stretcher The height can be changed for each card via a po tentiometer In figure B 9 an example how a Mul signal looks like is shown in Inter Integrated Circuit is a multi master serial single ended computer bus I C communication did not work reliable Low Voltage Transistor Transistor Logic Lemo is the name of a company producing mainly push pull connectors in the context of this document the name lemo is used as a synonym for the in nuclear physics experiment commonly used 1 pin connectors of type FFA 00 250 CTAC29 102 B 3 TRIPLEX this example five TRIPLEX cards have been connected in one tree Bach of them had a different width for its Mul signal The height of the Mul signal was adjusted such that it was 40 mV fired channel so the picture shown here represents one channel fir
47. perfect agree ment with the experimentally determined values for the ground states of both oxygen isotopes Two body NN and three body 3N interactions have been taken into ac count For the three body part the following contributions have been included two neutrons are in the PO core and one is in the valence space effectice one body part one neutron is in the 16O core and two are in the valence space effectice two body part and all three neutrons are in the valence space so called residual part Hol12 Sim12 The O and 7 O unbound resonances had recently also been investigated by other collaborations HBB t08 LDK 12 The location of the O ground state resonance given in HBB 08 and the one presented here agree very well The width determined in this analysis is smaller by factor of two than the one given in HBB 08 but it agrees very well with the single particle width determined using the Breit Wigner line shape For 260 only the resonance position of the ground state had been observed before while in that experiment an upper limit of 200 keV LDK 12 has been determined The data presented here reduces this limit to 50 keV It furthermore supports strongly that the emitted neutrons from the 74O n as well as in the O 2n system have a pure d wave character A disagreement between theoretical calculations and the here presented analysis can be seen in the position of the excited state in the experimental 740
48. range for which two correlations are visible is in the order of 1 ns which is 4 of the TAC range 146 B 9 KNOWN ISSUES TACQUILA with 17 channel and counter 7 150 events counter 9 events counter 8 120 4 4 e J e as Ca E un ee a 1005 counter 9 Eu Tu J e a 100 o of i Ba nn um oo o a n o oo 50 4 c P e et n a ao 80 ee En J 2 e u 2 J i 7 04 gt 604 u Data 2 amp 10 times smoothed gt 1 eo p E n 4 e 7 z counter 8 20 mee 100 5 7 or L e ki o 4 0 150 i T T T T T y T i T T 1 1840 1860 1880 1900 1920 1940 1960 1840 1860 1880 1900 1920 1940 1960 ADC channel ADC channel 2500 4 EE events ner o EVENS counter 9 gaussian fit 2000 4 S 4 1 amp o 66 ps 3 counter p Daten Datat1_D 1500 Gleichung 90 AW sq PU2 expt Z berc 2 L Gewicht ta y Keine Gewichtung Chn2IDoF 26251 66186 J RI2 0 91212 7 5221 419 97597 s 1895 58327 20 22079 r 10 9832 20 47096 340003 2608578093 1079 45734 Pa 5 4 gt H 0 3 gt uncertainty of counter value 0 26 500 0 1840 1860 1880 1900 1920 1940 1960 ADC channel Figure B 50 Shown is the percentage which is expected by the developers for the amount of events which have a wrong clock counter 147 APPENDIX B NEW LAND ELECTRONICS TACQUILA 148 Appendix C Data Sheets of MRPC Prototypes
49. seems to be at a first glance too low The nominal LAND efficiency for the detection of one neutron at 500 MeV is 95 BGI 03 However when using a rather small shower volume see table 4 1 a part of the true 1n events are seen as 2n This issue is overcome in the analysis in the 59 CHAPTER 5 ANALYSIS following way The O channel is analyzed using as condition LAND track multiplicity Ntmul gt 1 This can be done since for incoming 7 F and outgoing O there can be a maximum of one neutron In the O channel very little real In events are expected Furthermore their position in energy is known from O which allows us to look for this signature No large contamination from this channel is found as discussed later see figure 6 3 lower pad In summary one can say that as stated before shifting to higher neutron multiplicity is acceptable for the present analysis The slow increase in efficiency seen below Epe 3 5 MeV can be explained as follows the larger the relative energy the larger the angle and therefore the larger the irradiated area in LAND This effect reduces the impact of dead zones high thresholds or broken channels as a function of the relative energy e 10 10 ur 10 z b O gt 0 2n 103 2 E 107 a oe i 10 6 7 8 MeV E rel generated Figure 5 2 Shown are the response matrices of LAND for the 1n gt 1n a and In 2n b channels Meaning for both spectra o
50. seice deee Gl egau in deck Ve a ana ah ae n a 4 Identifying the Reaction Channel 4 1 Fragment Mass Identification via Tracking ooo a 4 2 2 Neutron Tracking 04 eed Bie salat nr Ben ern dar G 5 Analysis 5 1 Detector Response LEG Simulation 2 5 2 Breit Wigner Line Shape 0 000002 eee eee 5 3 Chi Square x and Likelihood Methods 2 2 2 22 222222 6 Results 6 1 Analysis of the O Chamnel 2 u s234 3 u 5er 6 2 Analysis of the 2O Channel ooa Be ea 6 3 Lifetime Estimates oaa eee ee eee 7 Discussion A Fit to O Data using R 6 fm and A 12 20 25 26 27 30 31 33 35 39 42 48 57 57 62 65 71 71 75 81 83 91 CONTENTS B New LAND Electronics TacQuila 93 B 1 Software TacQuila MBS sn er a ae een 93 Bi AND FEE se 2 2 0 0 a bog ei ar en a 97 B 3 TRIPLEX soe eek ORE wer aaa eee ee AG Se hae 102 B 3 1 TRIPLEX I C Address Tree of LAND 3 2 2 2 2 2 ee ad 112 B 3 2 TRIPLEX I C Address Tree of VETO 119 BAr Te la Send oracle eh Ged ees wees ok ee eA ad Se bs 120 BS ODD Cosi es teak Set AGE Oe BO BA Sis aa a Dal es We det ok a 125 B 6 Low Voltage sx doci 20a a aa nn ele ee en le 129 B 7 Additional Electronics aoao ee 133 B 8 Miscellaneous sri amara Ausser Shi ae Rh ee ee aw ae AS 140 B 8 1 Needed Cabling 2 000000 2 2 eee 140 B 8 2 Used TacQuila Systems 2 22 2 on nn 142 B 8 3 Used Material 2 ss pon 3282 S
51. that even for the very low statistics of the present experiment BAker s Nominal Likelihood 68 5 3 CHI SQUARE x AND LIKELIHOOD METHODS all methods give the same result within error bars see next chapter However the different analysis procedures are kept to proof the robustness of the analysis It has to be noted that for method C and F the absolute value of x has now meaning These methods deliver the correct shape and can therefore be used to determine the minimum x but the reduced x gives no information on the goodness of the fit 69 CHAPTER 5 ANALYSIS 70 Chapter 6 Results In the last two sections it has been discussed how a test function is generated including e g detector response as well as how to compare this function to the experimental data The described procedures will now be applied to the obtained relative energy spectra 6 1 Analysis of the O Channel Analyzing the 2 F O 4O0 n reaction the relative energy spectrum of O and therefore its mass is reconstructed see figure 6 1 The relative energy of the 74O n system depicts the data in the 0 to 8 MeV range one sharp resonance can directly be seen at approximately 700 keV The acceptance cut is at roughly 3 5 MeV and does therefore not influence the observed shape at all This is one disadvantage of the previously performed measurement at NSCL The setup used there has the following property In particular the wi
52. the time constant defines the limit for which pileup is not negligible e The second stage in the QDC channel is the baseline restorer which cancels out slow us variations of the baseline e The last stage is an ADC which converts the current to a 12 bit data word connector for cable to TRIPLEX Te a ja 58200 E i ar ca en an Bjc iss QDEZ2 K Koch gsi de 6 May gsi de LE 5 2006 ces a Figure B 32 Shown is the QDC2 piggyback board To get a QDC calibration several pulser measurements were done Some of the results are shown here to see the basic characteristics of the QDC board for more details see https elog gsi de tacquila test 273 The shape of the QDC channel vs SAM is not done perfect 125 APPENDIX B NEW LAND ELECTRONICS TACQUILA injected charge spectrum is like the one of a saturation curve see figure B 33 This was a feature which FOPI wanted to have since they want higher resolution for small charges The charge measurement is in the FOPI case only needed to perform a walk correction The decrease of the resolution with higher injected charge can be seen in figure B 34 For the experiments s393 s306b and s389 the NTF was read out with FaBu electronics and TacQuila electronics at the same time to see the correlations between the two read out systems One result is shown in figure B 35 Again the non linearity of the TacQuila QDC measurement can be seen Prior to the
53. this point the beam has been bend by 7 5 which is the minimum deflection angle to be 81 CHAPTER 6 RESULTS outside the acceptance of LAND The spectrum shown in figure 4 8 was produced using a CB trigger only to not reduce the statistic further If one requires the two above mentioned characteristics to have no neutron in coincidence and A gt 24 one possible candidate remains leading to an upper limit of N t 1 This one count can either be background or a real O event Here it is assumed to be real to define the upper limit To determine the error of this one count a Poisson distribution is used Figure 6 7 shows the probability to measure one count as a function of the given mean for a initial Poisson distribution The one and two sigma regions are determined such that integral reflects 68 3 and 95 4 probability respectively This yields N t 4 87 for a 95 confidence level gt in 0 0 45 A 150 Result using 1 0 1933 0 4 Result using 2 0 11 0 3 nn VEREORERERUEREEDERTEREPERSEPERLELTERLERSOSEPERDELEROEERELLUAERLLTIELEERELUERBORFLIRREERELSETEERTTELLERELDERE time Una Ka Da a Poisson Probability of Occurrence of x 1 2 3 4 5 6 7 Mean used in Poisson Probability Figure 6 7 Shown is the Poisson probability for the occurrence of x 1 as function of a given mean for a initial Poisson distribution Using the values discussed above e t ToF 11 8 ns
54. to Z different isotopes will experience a different energy loss Furthermore the energy loss depends on the trajectory of the ion since the depth of the material changes according to the radius of that trajectory Therefore unwanted isotopes are then bend out of the beam due to the different energy loss they experienced before while passing the second Bp stage of the FRS Of course a degrader has on the other hand the drawback that it decreases the intensity The beam setting used for the analysis presented here had low intensity For that reason the incoming ions did not have to be restricted to a certain species only and consequently no degrader was used Therefore many ions with similar A Z ratio were present in the so called cocktail beam The FRS beam line has been equipped with two 3 mm thick scintillator paddles Those detectors are needed to perform a incoming ToF measurement over a long distance FRS to Cave C for each ion One scintillator paddle was placed at the middle focus the so called S2 position and the second was situated behind the FRS so called S8 As the scintillator at the mid plane of the FRS S2 about 136 m upstream of the reaction target was overloaded with the intense beam we have been using the scintillator at the intermediate focal plane S8 leaving us a 55 m flight path to Cave C 29 CHAPTER 3 EXPERIMENTAL METHOD AND SETUP 3 3 LAND R B Setup at Cave C After production the secondary beam
55. used to readout a PMT which is coupled to a scintillator A new FEE version is currently being developed e Energy measurement is missing if we do not have the time Hakan s request for cross triggering e At the edges of the TAC spectrum there is a non linearity see figure B 27 and figure B 49 e At the moment the three cables used to connect the mother clock to the daughter clock modules do not have the same length this adds of course a phase shift Theoretically this should not create problems However it would be nice to keep the system as simple as possible so this should be changed in the future e Change slow control from LabView to EPICS use multiplexing at the moment only K Koch knows how this works e Because VV and VV have been inverted the labeling and color code of the cables is here wrong This has to be changed capacitors have already been replaced e Write cabling documentation e tacset txt contains two parameters which are not needed threshold and gener ator this causes only confusion e Heiko and Michael reported that the minimum for the clock cycle counter is 8 smaller values are not possible could that be connected to the 8 clock cycles delay 144 B 9 KNOWN ISSUES between the trigger at the comparator on the FEE and the readout of the QDC since this is also 8 cycles maybe something went wrong in the FPGA code e A scaler for the 17 on the TacQuila board would be helpf
56. value the first such measurement The theoretical calculations to which the results of this thesis have been compared to have been performed by Holt et al NN 3N Hol12 and Simonis et al NN 3Nres Sim12 Microscopic 2N and 3N forces from chiral EFT have been used to derive a po tential which is then used for shell model calculations Including three body forces due to A excitations one can explain the oxygen anomaly as detailed in chapter 1 Includ ing the normal ordered 2 body part of 3N forces leads to repulsive interactions between valence neutrons as shown in figure 7 2 Where effective 2 body part of the interaction means that one neutron from the core interacts with two valence neutrons Calculations done using this type of interaction are here labeled NN 3N Contributions from three valence nucleon interactions are in general suppressed by Nyatence Neore FS11 those contributions will therefore only be important in the most neutron rich nuclei Making 83 CHAPTER 7 DISCUSSION this effect significant for the oxygen isotopes discussed here This is why also contribu tions from interactions amongst three valence neutrons have been recently included in the theoretical predictions this interaction is called NN 3Nres within this work The results of the two different theoretical calculations performed by Holt et al NN 3N and by Simonis et al NN 3Nres are given in table 7 1 While the calculatio
57. was generated requiring a CB sum trigger only to not decrease statistics even further 46 4 1 FRAGMENT MASS IDENTIFICATION VIA TRACKING Mi eee ar ee REE ig i i240 250 20 te ie N t ee ae Mm counts JJ tt Lt tl js eae j 1 SEN BER BERN 22 23 24 25 26 27 28 29 30 corrected Bp in a u Figure 4 8 Fragment mass distribution for incoming F and outgoing oxygen isotopes Z 8 derived from tracking of the fragment through the magnetic field The effect of the so called odd even staggering meaning even nuclei are more stable due to pairing is visible The cut on O is indicated by the blue dashed lines The spectrum was gen erated requiring a CB sum trigger only to not decrease statistics even further Applying the condition that no neutron is detected in LAND one event remains in the 7 O oxygen and zero in the 7 O oxygen gate the middle of the target Using ATIMA Wei the energy loss which is itself velocity energy dependent is taken into account and a look up table is generated which converts a measured ToF into a velocity at the center of the target The energy and p respectively at the center of the target for the two incoming species of interest were e 26F E 442 MeV u 3 E 0 735 e F B 414 MeV u G E 0 722 ATomic Interaction with MAtter 47 CHAPTER 4 IDENTIFYING THE REACTION CHANNEL 4 2 Neu
58. 0 21 fragment mass Figure 4 5 Fragment mass distribution for incoming PO and outgoing oxygen isotopes derived from tracking of the fragment through the magnetic field A gaussian fit to the unreacted O peak shows that a mass resolution of 0 9 is achieved In the inlay a zoom on lower masses is shown to present that also in this mass region the separation is very good in figure 4 6 The mass distributions for incoming F and F and outgoing ions with charge number Z 8 oxygen isotopes are shown in figure 4 7 and figure 4 8 Applying the mass cut shown in these two spectra completes the selection of the reaction channel as far as fragments are concerned One remark has to be made All three tracking methods suffer from wrongly recon structed positions on the GFIs Meant are the events which e g in figure 4 4 are visible as the vertical band at GFI2x 10 But combining two methods like shown in figure 4 6 these events are not correlated and can therefore be excluded from the analysis The consequence of the potential 7 O and O in the fragment mass spectra shown in figure 4 7 and figure 4 8 is discussed in chapter 7 Besides the mass and the angle also the velocity of the fragment has to be known this quantity can be determined from the measured ToF Since the fragment passes through material air and detectors on its way from the target to the TFW the velocity is not constant therefore the measured ToF can not directly
59. 1 parameter name used value Rimar 20 0 cm Tras 5 0 ns Ot 0 2 ns Zinat 20 0 cm Zinin 12 0 cm Reyi 20 0 cm UFermi 8 0 cm ns Table 4 1 parameters used for the LAND shower algorithm Using a too large shower volume in the shower algorithm leads to the fact that small relative angles between the neutrons are not possible Figure 4 15 shows the relative 52 4 2 NEUTRON TRACKING P 2n 2n P 2n In 1 0 5 10 15 20 25 30 35 40 Rey incm Figure 4 13 Ratio of P 2n 2n P 2n 1n used to optimize the shower parameters The optimum is indicated as light blue box P 2n 3n 1000 in ps N oo 800 700 600 500 0 18 400 300 200 0 5 10 15 20 25 30 35 40 Rei incm Figure 4 14 Probability to detect three neutrons for two generated neutrons P 2n 3n as a function of the two parameters o and Rey used in the shower algorithm 53 CHAPTER 4 IDENTIFYING THE REACTION CHANNEL angle between two neutrons seen from the breakup of He into an a and two neutrons The same for the He decay between S n and 2 MeV In both cases it is clearly visible that the large shower volume cuts the small relative angles and therefore small energies in the Epe spectrum making this set of shower parameters unusable for the physics channels is shown in figure 4 16 but for a simulation which uses a flat distribution analyzed here 60 counts 50 40
60. 29 tac 30 R3 30 read_meb util paraload tacset txt done R3 30 read_meb util hstart 6 done File util tacquila_big_2010 hex loaded 98032 bytes total 98032 bytes File util tacquila_middle_2010 hex loaded 59215 bytes total 157247 bytes R3 30 read_meb util tacload 6 util tacquila_big_2010 hex util tacquila_middle_2010 hex done File util tacsam5 m0 loaded 2950 words R3 30 read_meb util hpistart2 6 util tacsam5 m0 done R3 30 read_meb R3 30 read_meb wait for GTB TACQUILA address initialization R3 30 read_meb this needs a few seconds R3 30 read_meb on SAM5 with id 6 GTB 0O 6 TACQUILAs found R3 30 read_meb R3 30 read_meb finished TACQUILA initialization R3 30 collector acquisition running This means the following three files have been loaded e tacset txt contains some of the TacQuila settings e tacquila_big_2010 hex code for first TacQuila FPGA e tacquila_middle_2010 hex code for second FPGA on the TacQuila And 6 TacQuilas have been initialized by MBS This number should of course match the current setup Some settings of the TacQuilas can be done in the above mentioned tacset txt which can be found in the MBS directory This file is only loaded during the startup of the DAQ so one has to do a stop acquisition and start acquisition to activate changes The following can be changed using this file 1 SAM id parameter which will be
61. 5 and therefore allowing only boron fragments To be able to understand this structure in more detail the generic tracker was used as simulation tool to reproduce the hit positions and investigate the dependence on certain parameters The tracker can be run using the sim option giving A Z x0 yo dx and dy Where xp and yo denote the postion on target while dz and dy are the angle at which the ion enters the magnetic field The locations of the detectors in Cave C are given as determined by the photogram metry see table 3 1 The magnetic field of ALADIN has been measured before for different currents these field maps are used here to simulate the magnetic field For the given parameters of the setup detector positions and magnetic field the track of the defined ion A Z 3 and incoming angle is fully determined and the tracker gives hit positions in detector coordinates as in the experimental data 42 4 1 FRAGMENT MASS IDENTIFICATION VIA TRACKING GFI2x GFIlx a u geen las eb vr Elsa la ma aaa lin la nn 15 10 5 0 5 10 15 20 25 GFI2x a u Figure 4 3 Shown is the x position on the second GFI vs Ar between the two GFIs A which is essentially the angle a tana x x Ar The here shown data represents z incoming B and outgoing boron isotopes This test case was chosen since the structure can clearly be seen due to higher statistics To investigate conditions closer to the reaction channel
62. 5 10 15 20 25 GFI2x in cm Figure 4 4 Shown is the x position on the second GFI vs Ax between the two GFIs The data shown in color represents incoming PO and outgoing O isotopes On top the results of a simulation are shown in red For more details see text than in the simulation as a result of the reaction and straggling On the basis of the correlation shown in figure 4 4 the fragment mass can be re constructed using an empirical formula Besides the dependence on the angle a two further corrections have been included covering the correlation to 8 as well as the target position xo resulting in the following formula for the fragment mass Ar F a ua GFl 4 1 p Ap a Gr 4 2 Ar Ap 43 2X0 4 3 The coefficients a have been determined empirically using the experimental data The result obtained using the empirical formula for the fragment mass Ar see formula 4 3 is exemplary shown in figure 4 5 In this spectrum the peaks were shifted and scaled to match the mass number However this is not done for the analysis of the O and O channel To check that the empirical mass formula is correct it is compared to the results obtained from the generic tracker A good agreement is found for the test case shown 44 4 1 FRAGMENT MASS IDENTIFICATION VIA TRACKING 3 12000 3 mass resolution 0 9 S mean 20 0 i 10000 mr 1 o 0 178 8000 6000 4000 2000 15 16 17 18 19 2
63. 7 x G x 9 gW 007 x OG X OT gW 007 X OT x OT suorsuowq pPped 4d vuop 10 US ydoou0D Od uN ANVT CANV TOON 11 CHAPTER 2 THE NEULAND TIME OF FLIGHT NEUTRON SPECTROMETER R3B Start version 2016 RIB from Super FRS I NeuLAND R B Si TRACKER Heavy fragments Protons CALIFA R B GLAD Figure 2 1 Shown is the start version of the R B setup The figure is taken from Bor11 After NeuLAND has been completed spectroscopy of certain neutron rich unbound systems beyond the neutron drip line will be possible for the first time Amongst those 280 is one of the key nuclei which could be populated via proton knockout from a F beam This beam will be available with high intensities at FAIR The experimental challenge is to reconstruct the relative energy of the 5 body decay which requires the detection of four neutrons which are emitted with small relative energy NeuLAND with its high 4n efficiency and its high resolving power will enable these kind of studies One main part of the here presented work was performed within the NeuLAND working group In particular MRPC prototype tests and implementation of new readout electronics have been carried out These two subjects will be briefly discussed in the following two sections 2 1 MRPC based Neutron Detector Concept In the R amp D phase one possible design of NeuLAND consisting of a layered structure of passive converting material i
64. 90s by J Keller and is called tracking algorithm KM91 The first shower algorithm was written by Th Blaich Attempts to improve it by using for example the deposited energy Lei97 did not lead to further improvement The main difference between the tracking and the shower algorithm is that the tracking algorithm has the better performance concerning charged particle identification Leill Sim11 Ros11 A recent review covering this topic can be found in PBL 12 The shower algorithm uses HIT level data see figure 4 1 as input Various conditions and sorting methods are applied on the data to identify neutrons photons cosmic muons and charged particles The output of the shower algorithm contains 0 and 8 of the neutron instead of the original information present at HIT level time position and energy this format is more useful for the subsequent physics analysis The different conditions which are applied to identify neutrons will be discussed here counts neutrons velocity c in cm ns Figure 4 10 Shown is the reconstructed velocity of particles detected in LAND minus the speed of light c No incoming cut is applied meaning reactions from all different isotopes in the cocktail beam are allowed The data represents events recorded using a carbon reaction target The y peak which serves to calibrate the time axis is clearly visible its width is a 450 ps The resolution for the ToF measurements of the
65. ADIN which is filled with helium gas The mag netic field of ALADIN bends the charged fragments but leaves the neutrons unaffected on their straight trajectories All detectors behind ALADIN are operated in air In the following we will first focus on the detection of the fragments which is done using the so called fragment branch or arm of the experimental setup The fragment branch is oriented such that the central position at each detector is at 15 with respect to the incoming beam axis It consists of 3 detectors see figure 3 3 two fiber detectors GFI 1 and 2 and the TFW Each GFI is build of 480 vertical fibres covering in total an area of 50x50 cm A position measurement in x direction horizontal with a resolution of 1 mm MJP 09 is done using this type of detectors Having these two position measurements behind the magnet and the ones done by the DSSSDs in front of the magnetic field the trajectory of the ions can be reconstructed Different isotopes are deflected to different angles in the magnetic field of ALADIN according to their different mass to charge ratio see for mula 3 8 Therefore the Bp of an ion is measured by reconstruction of the track of this ion through the magnetic field The Bp value determines the mass of the ion un ambiguously if the charge is already known see section 4 1 Detailed information on the GFIs can be found in CSG 98 and the calibration procedure is described in MJP 09 The last
66. D C1t7 18 1535 508 2508 2 LAND C1t8 14 1511 515 2515 1 LAND C1t9 15 1519 516 2516 E LAND C1t10 6 1515 505 2505 E LAND C2t1 SN11 1512 503 2503 5 LAND C2t2 SN4 1513 512 2512 9 LAND C2t3 A 1529 537 2537 E LAND C2t4 10 1518 528 2528 E LAND C2t5 8 1514 oll 2511 2 LAND C2t6 8 1531 504 2504 9 LAND C2t7 14 1517 506 2506 4 LAND C2t8 9 1532 518 2518 2 LAND C2t9 6 1533 510 2510 9 LAND C2t10 SN3 1526 538 2006 4 LAND 142 B 8 MISCELLANEOUS position TRIPLEX FEE TacQuila QDC HEX used for C3tl SN6 1520 513 2513 4 LAND C3t2 11 1503 533 2533 5 LAND C3t3 A 1506 522 2527 7 LAND C3t4 SN14 1540 529 2539 D LAND C3t5 SN12 1504 519 2519 6 LAND C3t6 T 1530 540 2536 8 VETO C3t7 22 1528 523 2540 3 VETO C3t8 5 1508 520 2011 4 VETO C3t9 T 1524 527 2529 3 C3t10 15 1502 531 2531 6 Table B 9 The SN of all used electronic boards are shown It is also shown were each TacQuila system is located and which Hex Address the TRIPLEX has to have to fit into the TRIPLEX tree B 8 3 Used Material In this section the names of the used material will be mentioned this might be use ful if somebody in the future is looking e g for a certain connector for upgrades or maintenance of the system e connector for GTB cable ERNI 024403 SMCB 50F AB VV 3 01 e connector for TRIPLEX cable ERNI 214346 SMCB 26F AB VV 6 01 e connector for CLOCK cable at
67. E O CHANNEL 6 2 Analysis of the O Channel The relative energy spectrum of 6O is shown in figure 6 3 Predicting the shape of this spectrum with hand waving arguments one would naively assume that it should be similar to the E 1 O 7 O is a bit more exotic larger A Z but on the other hand the system should gain binding energy due to pairing effects experimental data 27 26 24 F gt O0 gt 0 2n simulated response to flat distr between 0 and 50 keV simulated response to flat distr between 4200 and 4250 keV Efficiency x Acceptance a 2 PEeTTTTeTETTTTTI counts 200keV Efficiency x Acceptance M A N experimental data a gt ITITTEITTTTTETT counts 200keV simulated response to to flat distr between 0 and 50 keV gt Breit Wigner o os r 1 r e a 9 0 2 3 4 5 6 7 8 E a in MeV Figure 6 3 The In and 2n channel of the 2 O data are shown upper and lower pad The experimental data is shown as a solid black line In blue the best fit to the low energy part of the In and 2n channel is shown In purple other contributions to the In as well as 2n are indicated In the upper pad the solid purple line reflects a excited state of 2 O while in the lower pad the dashed purple line depicts the contribution to this channel originating from real 1n events In the upper pad the LAND efficiency x acceptance is shown in red
68. EX control window can be used by several cards as long as they are on different branches of the tree The Addressing of the cards will be discussed in detail later see Sec B 3 1 e PutData pushing this button will write the shown values to the TRIPLEX card e GetData pushing this button will read the current settings from the TRIPLEX card e stop will close the window if one uses the X to close the window like most users might intuitively do the LabView GUI will freeze 108 B 3 TRIPLEX e LoadFile one can load old settings from a file In the values box on the right side of the TRIPLEX control window one can set the following e threshold and stretcher values for each of the 16 channels of this TacQuila The stretcher value sets the width of the logical signal used for the Mul see also fig ure B 9 While the threshold is used for the comparator The threshold is applied after inversion and amplification of the signal at the comparator see figure B 3 To be able to estimate a reasonable threshold setting see table B 5 e Read Channel Reg i i 1 2 expects a eight digits binary value Read Channel Reg is used for the Multiplexer Here one can put one of the x 16 used channels to the Mux output of the TRIPLEX interface With x being the number of TRIPLEX cards in the tree This was never used by me For details ask K Koch or A Ignatov The Read Channel Reg2 is u
69. Gentil S Leray J Lukasik J L hning W G Lynch U Lynen Z Majka M Mocko W F J M ller A Mykulyak H Orth A N Otte R Palit S Panebianco A Pullia G Raciti E Rapisarda D Rossi M D Salsac H Sann C Schwarz H Simon C Sfienti K S mmerer M B 169 BIBLIOGRAPHY Tsang G Verde M Veselsky C Volant M Wallace H Weick J Wiechula A Wieloch and B Zwieglinski Neutron recognition in LAND detector for large neutron multiplicity eprint arXiv 1203 5608 March 2012 PKGR12 M Pf tzner M Karny L V Grigorenko and K Riisager Radioactive PPP71 RCa RCb R0s09 Ros11 SC81 Sch11 Sch12a Sch12b Sim11 decays at limits of nuclear stability Reviews of Modern Physics 84 567 619 April 2012 V V Parkhomchuck Y N Pestov and N V Petrovykh A spark counter with large area Nuclear Instruments and Methods 93 269 1971 R3B Collaboration Direct reactions of light ex otic beams measured in complete kinematics at R3B http gsi de informationen wti library scientificreport2010 PAPERS PHN NUSTAR NR 04 pdf R3B Collaboration Technical Proposal for the Design Construction Com missioning and Operation of R3B A universal setup for kinematical com plete measurements of Reactions with Relativistic Radioactive Beams http www gsi de forschung kp kr R3B Technical Proposal pdf D M Rossi Investigation of the Dipole Response of Ni
70. IAI E EET E a 102 w Der ee 10 O 2n O n 0 A 2 2n A 1 n A Figure 7 4 Shown is a level scheme of the known states in O as solid black lines on the left as well as the new results on 2 O and O shown as blue solid lines on the right The experimental errors are represented by the light blue bands 24O has no bound excited state the neutron separation energy Sn at 4 1 MeV is indicated as horizontal black dotted line as well as the first excited 2 state above it at 4 7 MeV values are taken from HBB 11 26O and O have in the experiment presented here been populated via proton knockout reactions from F and F respectively For O exclusively the ground state has been populated which decays directly to the O ground state solid gray arrow In the 6O case the excited as well as the ground state can be populated From the ground state only a direct decay to the 74O ground state is possible black solid arrow The connection between the 2 O ground state and the O ground state is indicated by the light grey arrow showing that this decay is energetically not possible Sn gt 0 For the first excited state a direct decay black solid arrow as well as a sequential decay gray dotted and gray solid arrow via the O ground state is possible 89 CHAPTER 7 DISCUSSION 90 Appendix A Fit to O Data using R 6 fm and A method x at min E at min in keV T at min in
71. Kempe V Maroussov O Nusair H Simon M Sobiella D Stach A Wagner A Zilges and R3B Collaboration Prototyping and tests for an MRPC based time of flight detector for 1 GeV neutrons Nuclear Instruments and Methods in Physics Research A 654 79 87 October 2011 171 BIBLIOGRAPHY YBR12 D Yakorev D Bemmerer and RB Collaboration Work on a large area MRPC based time of flight detector for high energy neutrons Journal of Physics Conference Series 337 1 012035 February 2012 ZHN 97 M Zinser F Humbert T Nilsson W Schwab H Simon T Aumann M J G Borge L V Chulkov J Cub T W Elze H Emling H Geissel D Guillemaud Mueller P G Hansen R Holzmann H Irnich B Jonson J V Kratz R Kulessa Y Leifels H Lenske A Magel A C Mueller G M nzenberg F Nickel G Nyman A Richter K Riisager C Scheiden berger G Schrieder K Stelzer J Stroth A Surowiec O Tengblad E Wa jda and E Zude Invariant mass spectroscopy of Li and Li Nuclear Physics A 619 151 176 February 1997 172 Acknowledgement Physics is becoming so unbelievably complex that it is taking longer and longer to train a physicist It is taking so long in fact to train a physicist to the place where he understands the nature of physical problems that he is already too old to solve them EUGENE WIGNER I don t want to argue with Wigner of course and i certainly still know very little but I gues
72. Lycron Spray anode iron steel plate thickness 4 mm anode strip distance 0 3 mm gas in between outer electrode thickness 2 mm comments all 8 strips the same longer than active area and with triangular strip end all strips are coupled with transformator to readout pictures results time resolution 89 ps efficiency voltage 833 6 75kV cross talk at n n 30 amplitude of crosstalk 6 APPENDIX C DATA SHEETS OF MRPC PROTOTYPES prototype name GSI 3 tested during beam time HZDR 15 06 09 19 06 09 number of strips structured anode 8 anode strip width 25 mm strip length 40 cm active area 800 cm number of gaps 2 4 gap size 0 3 mm glass plate thickness 0 5 mm HV Lycron Spray anode iron steel plate thickness 4 mm anode strip distance 0 6 mm gas in between outer electrode thickness 2 mm comments all 8 strips the same longer than active area and with triangular strip end all strips are coupled with transformator to readout pictures 152 results time resolution 74 ps efficiency voltage 77 6 75kV cross talk at n n x16 amplitude of crosstalk z5 prototype name GSI 4 tested during beam time HZDR 30 08 09 02 09 09 number of strips structured anode 8
73. Peters W F Rogers A Schiller J Snyder A Spyrou S L Tabor and M Thoennessen Observation of a two neutron cascade from a resonance in 740 Physical Review C 83 3 031303 March 2011 Hol12 J Holt Theoretical calculations for neutron rich oxygen isotopes using nn 8n interactions private communication 2012 HV11 M Heine and V Volkov Photogrammetry private communication 2011 Jam F James http wwwasd web cern ch wwwasd cernlib me genbod html Jan09 Robert V F Janssens NUCLEAR PHYSICS Unexpected doubly magic nucleus NATURE 459 7250 1069 1070 JUN 25 2009 Joh06 H T Johansson The DAQ always runs Master s thesis Chalmers University of Technology Goteborg Sweden 2006 Joh10 H T Johansson Hunting Tools Beyond the Driplines PhD thesis Chalmers University of Technology Goteborg Sweden 2010 JPSTO1 R K Jain A V Prokofiev A N Smirnov and L Tommasino Measurement of high energy neutrons by fission reactions Radiation Measurements 34 129 132 2001 Proceedings of the 20th International Conference on Nuclear Tracks in Solids JSH49 J H D Jensen H E Sue and O Haxel Modellm ige Deutung der ansgezeichneten Nucleonenzahlen im Kernbau Naturwissenschaften 36 155 156 May 1949 KHS 05 K Koch H Hardel R Schulze E Badura and J Hoffmann A New TAC Based Multichannel Front End Electronics for TOF Experiments With Very High Time Resolution IEEE Tra
74. Ps tp ti bpp ftp AIAR RIAD ARAR ARAR ti ig fi ftp ti ttt ftp tlt 2 12 HIPS RAR RIARI AD PIP I PSAP AIA SS Php pI PLE LEER 410 LLL AE DL AE LL AIRA AG LL AE SS IG fl ti A A A Mss LLL BARRAR RRRA RRIA RR AARD AAA RI AIDIA RR AAA R AAR RAA PEAN PARRA ESE SESE ARAA ARAA ORARAA RARA RIAA RRA ARIA ATARA s LILLIE WHHL 7 107 10 10 10 1 E MeV Figure 7 3 Graphs are adapted from GMSZ11 The x axis depicts the position of the resonance the y axis on the left gives the width T while the y axis on the right shows the corresponding lifetime 7 The colored lines show the theoretical correlation between width lifetime and resonance position assuming different single particle states The experimental results are shown on top of those theoretical predictions In the 7 O case pad a the determined limits grey lines define a allowed region which is shown as a blue hatched area In the lower pad b the result for the width T and resonance position E of the O ground state are indicated as grey lines and their errors are represented by the blue area For further details and discussion see text 88 gt 3 3 E amp F 10 L EEIEER EESE ESE EPE EEEE EEEE EET rr 3 E SEPETE C 25 a SE e g LL m 1 Er PEE L LL TIIE FE PIPE AE EEIE PEE RE T IEEE LEST ATIA LETER SAEN RIAAL ATAN IRETE AIET TIIE A EE E EEIE AET E LIV 10 pa R EEEN LIIE tls Meee eh alin AAS SOPIE SEE AEE IERE TALR EEEE E
75. Rdtxt02 6 TAC measurement PMT2 Rdtxt03 6 counter measurement PMT1 Rdtxt04 6 counter measurement PMT2 Rdtxt05 6 17 channel TAC measurement Rdtxt06 6 17 channel TAC measurement Rdtxe01 6 QDC value PMT1 Rdtxe02 6 QDC value PMT2 Table B 2 Shown is the structure of TacQuila data in land02 RAW level Using tcal it is possible to create SYNC level data within the land02 analysis frame work Besides land02 also a UCESB version for the TacQuilas is available see e g u land unpackexps tacquila17 or u land unpackexps s393 This is e g usefull to setup an unpacker or watcher for a small test setup Unpack and Check Every Single Bit 96 B 2 LAND FEE Sdtxmul same as in RAW level Sdtxil6 same as in RAW level Sdtxt01 6 reconstructed time of PMT1 in ns Sdtxt02 6 reconstructed time of PMT2 in ns Sdtxt03 6 counter of PMT1 in ns Sdtxt04 6 counter of PMT2 in ns Sdtxt05 6 17 TAC in ns Sdtxt06 6 17 TAC in ns Sdtxe01 6 QDC PMTI randomized over bin width Sdtxe02 6 QDC PMT2 randomized over bin width Sdtxe03 6 empty Sdtxe04 6 empty Sdtxe05 6 empty Sdtxe06 6 empty Table B 3 Shown is the structure of TacQuila data in land02 SYNC level B 2 LAND FEE The LAND FEE is an adaption of the FOPI FEE CSC 07 Due to its shape it is also called Hammer Hai engl hammerhead shark see figure B 1
76. SIS 18 having an energy of 11 5 MeV u For the future upgrade of GSI FAIR the SIS 18 will be used as injector Figure 3 1 shows a schematic layout of the existing GSI accelerator on the left blue beam line and the planned upgrade FAIR on the right red beam line The numbers in the names of the synchrotrons depict the magnetic rigidity in Tm so from the naming one can immediately see the ratio by how much the FAIR synchrotrons are larger see e g figure 3 1 Leaving the SIS 18 the 4 Ar ions have been accelerated to an energy of 490 MeV u and the primary beam is guided onto the production target at the entrance of the FRS A 4011 mg cm thick Be production target was used The primary beam had an intensity of 6 10 ions spill and it was used for the proton rich as well as the neutron rich secondary beam settings The beam composition delivered to the experimental cave depends on the FRS settings only The s393 experimental campaign took place in Cave C highlighted by a green box in figure 3 1 A detailed description of FRS can be found in GAB 92 In addition to the FRS setting magnetic rigidity also the spill structure of the beam was changed depending on the rate of the secondary beam The number of ions per spill is fixed as well as the maximum rate of ions for which the experimental setup is able to record all events As a consequence long spills and a slow ramping mode are used for high intensity beam settings and short spi
77. _1 GTB_6_0 and GTB_6_1 out of which we currently use three One GTB chain is used for each TacQuila crate The connection is done in the following way e GTB_60 TacQuila crate 1 top ECL NIM level converter embedded in a VME module VME TRigger Synchronizing Module VME TRigeer Distribution Module 137 APPENDIX B NEW LAND ELECTRONICS TACQUILA e GTB_6_1 TacQuila crate 2 middle e GTB 50 TacQuila crate 3 bottom This means all 20 modules on GTB_6_0 and GTB_6_1 are used for LAND The modules one GTB_5_0 are split into two groups The first five are also used for LAND while the last five are used for Veto and as spare modules see also Sec B 3 1 and B 3 2 in particular figure B 23 shows how the modules are arranged This information is needed to do the mapping from TacQuila channels to detector channels If the full list for the mapping of each individual channels is needed please contact D Rossi The TRIDI and TRIVA trigger bus connectors have to be terminated see red circles which are labeled with T in figure B 44 On the TRIVA bus the following signals are present 1 physics trigger 2 off spill trigger 3 clock 4 tcal On the TRIDI bus the triggers are ordered in the following way 6 clock 7 tcal 8 physics To trigger the 17 channel for physics events the trigger should never be taken from the TRIDI bus but from the original source using a direct cabl
78. al effects such as thresholds of the readout electronics counts 60000 50000 40000 30000 threshold 20000 10000 L 0 5000 10000 15000 20000 25000 charge in ch Figure 2 8 A typical charge spectrum of one MRPC channel is shown The black line represents the raw data while for the data shown in red a valid time is required The TDC threshold is indicated as red dotted vertical line The inlay represents a zoom depicting the threshold Although NeuLAND will finally be build from fully active scintillator material the response of MRPC neutron detector prototypes to fast neutrons will be further investi gated using quasi monoenergetic neutrons at various energies derived from the quasifree break up reaction of a deuteron beam on a CHg target in an GSI experiment scheduled for autumn 2012 Efficiencies as well as time resolutions shall be studied and a final characterization of the MRPC prototypes will be performed 2 2 Readout Electronics for NeuLAND Besides the detector itself also new readout electronics have been developed to improve the performance of NeuLAND The new readout electronic is based on the TacQuila board The TacQuila board is an electronic readout board developed by GSI K Koch for the FOPI experiment KHS 05 The most remarkable characteristic of the board is GEometry ANd Tracking 20 2 2 READOUT ELECTRONICS FOR NEULAND the exce
79. all detectors to extract physics from the measured data However in the analysis presented here the High Voltage 57 CHAPTER 5 ANALYSIS experimental response of the full setup is mainly determined by LAND since in the full analysis a neutron trigger was required and no acceptance cut is seen in the fragment distributions Furthermore the CB which is normally the other main contribution to the response is not needed in the presented analysis To determine the LAND response matrix real neutron data from a calibration ex periment serve as input This data is stored in a database from which hit patterns are taken and those are shifted to the location of the neutron interaction which has been simulated by the LEG For the digitizer the thresholds have to be given for each in dividual channel and the data reconstruction is done using the earlier described shower algorithm this entire procedure is combined in the so called LEG The LEG simula tions presented here were done by D Rossi The particle decay was simulated using the GENBOD CERN library N Body Monte Carlo Event Generator Jam Since 4O does not have bound excited states see e g figure 7 4 no y s have to be included in the simulation The efficiency of LAND is directly influenced by broken detector channels During the s393 experimental champaign the LAND paddles listed in table 5 1 could not be used for the data analysis plane paddle 1 1 20 19 6 13
80. area on the detector in cm e Nyp number of beam micropulses per second Combining formulae 2 1 and 2 2 leads to N N ee ME 2 3 Ni A Fn Nmp is given by the beam structure and amounts to 3 3 MHz The neutron flux Fn at the detector position is obtained using the information of the neutron monitors A is for each strip derived from the beam profile at the MRPC detector see figure 2 7 The vertical position of a hit is directly given by the fired strip however in the analysis smeared out over the 2 5 cm width of the strip Using the time difference between the left and right signals the horizontal position of the hit is determined and using both the beam profile is reconstructed A signal velocity of 20 cm ns is used for this purpose The result is shown in figure 2 7 The reconstructed beam profile is in good agreement with the expected beam diameter of 10 cm defined by the collimator 18 2 1 MRPC BASED NEUTRON DETECTOR CONCEPT counts 1400 1200 y position in cm 1000 800 600 400 200 0 20 15 10 5 0 5 10 15 20 X position in cm Figure 2 7 Beam profile as reconstructed from the position information of the MRPC The result of the efficiency analysis is summarized in table 2 2 The efficiency was determined for the lower five strips ranging from 0 6 to 1 1 The three upper ones were not sufficiently irradiated see figure 2 7 strip 1 2 3 4 5 average efficiency in
81. as mainly set to the maximum however during a real experiment it should of course be set to the shortest value reasonable to minimize the dead time of the system In the LAND DAQ the TacQuila events are marked with the following numbers type 94 subtype 9400 So one can do a type event in the DAQ screen e g X86G 10 ty ev v and look for the 94 9400 events to see if the TacQuilas deliver data For an event measured with TacQuila readout electronics the result of three individ ual measurements is written to the data file 1 TAC value 2 clock cycle counter 3 QDC value On the MBS side the data format of the TacQuilas looks like shown in table B 1 Unpacking Imd files with landO2 one gets the parameters shown in table B 2 and B 3 The structure shown here is from an example of the DTF x paddles the unpacking was done using the option paw ntuple RTS RAW SYNC DTF so one has DTF data in RAW and SYNC level To have times in ns from the data delivered by a TacQuila system one should use land02 SYNC level data For RAW level data the time information is split into the fine and coarse measure R Plag integrated the calibration routine explained in Sec B 4 into the tcal program The tcal routine is one of the standard programs of the land02 package For more information on the land02 package see www linux gsi de rplag land02 These values are not TacQuila hard coded they are e g defined in u land lynx land
82. at high statistics are needed But on the other side all data can be used for the calibration No special events e g tcal trigger are needed The calibration parameters should be monitored over time since the TAC measurement shows a weak temperature dependence see figure B 28 121 APPENDIX B NEW LAND ELECTRONICS TACQUILA 1200 counts 1000 800 600 400 200 a RERJTZRSREENRERER L l I l i 0 500 1000 1500 2000 2500 3000 3500 2000 TAC in channels 2500 counts 2000 1500 1000 500 SE p pr T 0 5 10 15 20 25 30 TAC in ns Figure B 27 In the upper pad a raw TAC spectrum is shown The non linearity at the edges can be seen This particular shape is an intrinsic property of the TAC chip see also section B 9 N Kurz suggested to cut these events and during the test phase this was done to have a cleaner sample of events The cuts are indicated by the red dotted lines The lower panel shows the calibrated time spectrum in ns after applying the cuts Counts TAC Channels Figure B 28 Shown is schematically how the TAC spectrum changes with temperature Only the right edge moves The change is in reality much smaller than shown in the picture 122 BA TACOUILA To test the resolution of the TacQuila system the time difference between two chan nels is used Here first a measurement is presented which highlights a specific feature o
83. ation for Mul signal Baseline of analogue sum Amplification for analogue sum I C address 104 B 3 TRIPLEX TRIPLEX FrontEnd Figure B 7 Screen shot of the LabView Main panel to control the TRIPLEX Visualizes also the tree like order of the TRIPLEX cards TRIPLEX Interface prototype Mul i Pulser In d Y Figure B 8 Shown is the front Panel of the TRIPLEX interface prototype the different connectors are indicated In addition to the here shown connectors on the front a connector for the I C to the first TRIPLEX in the tree can be found on one side normally the connector J5 is used for this purpose For more details see text 105 APPENDIX B NEW LAND ELECTRONICS TACQUILA J 400ns 2 Vf2 5005 75 1 96 0mV 10k 10k points Waa Jun 2000 17 40 58 Figure B 9 A screen shot from the scope of the TRIPLEX Mul signal is shown The sparks visible here have been removed for the currently used TRIPLEX boards by adding capacitors For details on other aspects of the signal shape see text Bl JAHT USB MFP Server Control Center HER 9 Server Information 8 TCP IP Status S Supported Protocol Status Figure B 10 Shown is the menu of the MFP server At the moment there are 3 USB over IP server used for TacQuilas at GSI The one called EE should not be touched The ones called land usb001 and land usb002 are used to control the two TRIPLEX chains which are currently i
84. attered forward elastically This is done by comparing the momentum which connects the two hits to the momentum one would get if a elastic scattering had taken place The described seven parameters of the shower algorithm have to be optimized such that the performance of the algorithm is ideal for the analyzed physics channel In the 50 4 2 NEUTRON TRACKING analysis presented here it is crucial that as few 2n events are shifted to the In channel as possible To investigate the optimum shower parameters simulations using the LEG were done A three body decay fragment and two neutrons was simulated the LAND shower algorithm was used to reconstruct neutrons and the two most crucial parameters Rey and o were changed Having a real 2n event one will with a certain probability detect In 2n 3n and so on The probability to detect In for a generated 2n event as a function of R y and o is shown in figure 4 11 P 2n In 0 5 in ps 0 5 10 15 20 25 30 35 40 Rey incm Figure 4 11 Probability to detect one neutron for two generated neutrons P 2n 1n as a function of the two parameters o and Rey used in the shower algorithm The probability to detect 2n for a generated 2n event as a function of R y and o is shown in figure 4 12 In this figure two effects are clearly visible having a large shower volume the events are shifted to the In channel while having a very small shower volume shifts events to the 3n
85. be converted into the velocity at 45 CHAPTER 4 IDENTIFYING THE REACTION CHANNEL A generic tracker counts 50 SSP a a a a Zu a a a a a a a a a T bi K i s a h ker n aj e 1 1 1 i l 1 1 i l ei 1 1 16 i7 10 10 ee 1 s I 1 1 ke L 1 1 18 19 20 2 A empirical mass Figure 4 6 Two dimensional fragment mass distribution for incoming PO and outgoing oxygen isotopes The fragment mass is once determined using the generic tracker and once using the empirical mass formula counts 50 40 30 20 23 24 T T T T T T T T T T T T T T T T T T T T IT T T TI T T T K E25 u ie O E I a C 4 gt ee nee EE be aa RE EEE l Hape u H f x C h l a C h SRRSSHANEN ES ceed Lanes Gre A EWR le Red EAE Ae Ce PE eR ne nm a pi x Pi Pi I i ae k i J p N eee ieee ee jr ie S LF I RR METER Du i 4 Cir il tle o H L L L L L L L i L L L L L L L I i L L L L i L L 1 1 ki dhn l n 27 28 29 25 26 corrected Bp in a u Figure 4 7 Fragment mass distribution for incoming 7 F and outgoing oxygen isotopes Z 8 derived from tracking of the fragment through the magnetic field The gate on 4O is indicated by the blue dashed lines it is also indicated in which region heavier oxygen isotopes would be expected The spectrum
86. cQuila card of 10 ps is achievable see figure B 31 Measuring a time difference using channels from two different TacQuila systems worsens the resolution slightly For completeness of this section the calibration trigger will be discussed as the last topic The TacQuilas accept two different trigger types see also Sec B 1 the physics trigger trigger 1 and the calibration trigger trigger 2 The calibration trigger comes always at 80 of the width of the TAC spectrum So it could for example be 123 APPENDIX B NEW LAND ELECTRONICS TACQUILA 12000 counts 10000 8000 o 24ps 6000 4000 2000 pert N 1525 153 1535 154 15 45 15 5 15 55 15 6 TAC TAC ns Figure B 30 Shown is the time difference between two calibrated channels in ns For these two channels inputs with a fixed delay have been used In black the whole data is shown The green and blue peaks show the contributions of the two different event types see figure B 29 111 Ke i 0 9 Ch 10 5 ps 70000 counts 60000 50000 40000 30000 20000 10000 0 25 30 35 40 45 50 TAC TAC Ch Figure B 31 Shown is the difference between two RAW TAC measurements In this example the full TRIPLEX LAND tree was triggered using the TRIPLEX pulser The width of a TAC spectra is 2150 channels see e g figure B 27 this reflects the 25 ns of the Clock Therefore the bin width is a
87. ckel Isotopes in the Presence of a High Frequency Electromagnetic Field PhD thesis Johannes Gutenberg Universitat Mainz Germany 2009 D Rossi LAND calibration and reconstruction private communication 2011 R Santonico and R Cardarelli Development of resistive plate counters Nuclear Instruments and Methods in Physics Research 187 2 3 377 380 1981 H Scheit Experimentelle Untersuchungen der Kernstruktur weitab der Sta bilitat Vortrag im physikalischen Kolloquium der TU Darmstadt 2011 A Schwenk Neutron rich matter in the laboratory and the cosmos Vortrag DPG Fr hjahrstagung Mainz 2012 2012 A Schwenk Theoretical calculations for neutron rich oxygen isotopes includ ing three body forces private communication 2012 H Simon LAND calibration and reconstruction private communication 2011 170 BIBLIOGRAPHY Sim12 SLN 99 TAA 97 TS11 J Simonis Theoretical calculations for neutron rich oxygen isotopes using nn 8nres interactions private communication 2012 H Sakurai S M Lukyanov M Notani N Aoi D Beaumel N Fukuda M Hirai E Ideguchi N Imai M Ishihara H Iwasaki T Kubo K Kusaka H Kumagai T Nakamura H Ogawa Y E Penionzhkevich T Teranishi Y X Watanabe K Yoneda and A Yoshida Evidence for particle stability of F and particle instability of N and SO Physics Letters B 448 180 184 February 1999 O Tarasov R Allatt J C Ang
88. col Ion Source On Line Joint Institute for Nuclear Research Kernfysisch Versneller Instituut Lawrence Berkeley National Laboratory Large Area Neutron Detector LAND Event Generator programmable LEVel CONverter list mode data low voltage Low Voltage Transistor Transistor Logic Multi Branch System GSI Data AcQuisition software Multi Function Peripheral Multi gap Resisitve Plate Chamber Michigan State University Multiplicity Neu engl new Large Area Neutron Detector Successor of LAND 160 NDF NIM NSCL NTF NuSTAR PAW PECL PhD PID POS PPC PMT PSP QDC R amp D R B RIB RIBF RIPS RIKEN ROLU RPC SAM SEETRAM SINP SIS SISSI SN SPE SPIRAL TAC TDC TFBC TFW ToF TRIDI TRIVA Number of Degree of Freedom Nuclear Instrumentation Module standard National Superconducting Cyclotron Laboratory New Time of Flight wall Nuclear STructure Astrophysics and Reactions Physics Analysis Workstation Positive Emitter Coupled Logic Philosophiae Doctor Particle IDentification POSition detecor which is nowadays used for the incoming timing Parallel Plate Chamber Photo Multiplier Tube Position Sensitive Pin diode charge Q to Digital Converter Research amp Development Reactions with Relativistic Radioactive Beams Radiactive Ion Beam Rare Isotope Beam Factory RIKEN Projectile Fragment Separator The Institute of Physical and Chemical Research japanese abbreviation Rechts Obe
89. cussed later The heaviest unbound nucleus for which a resonance has been measured so far is F CFA 12a CFA 12b Very recently Lunderberg et al LDK 12 populated the O ground state via proton knockout from F and they found that O is unbound by 150159 keV placing its ground state below the 50 ground state D 2 g 3 Z l e e A F 9 Oo 8 N 7 stable nuclei Cc 6 B 5 bound proton rich nuclei Be 4 bound neutron rich nuclei Li 3 He 2 unbound neutron rich nuclei H 1 neutron drip line 0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 Neutron Number Figure 1 1 Shown is a section of the chart of nuclides Stable nuclei are indicated by black squares Neutron rich nuclei by blue and proton rich by red squares The exper imentally verified neutron drip line is shown as a yellow solid line up to the oxygen isotopes while the status for fluorine isotopes which has not been determined in ex periments yet is indicated as a yellow dashed line Unbound neutron rich nuclei are represented by cyan squares The two nuclei of particular interest within the framework of this thesis namely 2 O and 7 O are marked by grey dots The classic magic numbers are highlighted by the solid black vertical and horizontal lines and bold numbers As discussed in OSH 10 shell model calculations based on a microscopic nucleon nucleon NN force predict
90. d Terminal Blocks for one crate are mounted but only the cables for one out of the ten TacQuila systems are connected Name of the company which produces the parts 130 B 6 LOW VOLTAGE lt q Connector gt Power supply TacQuila label on TacQuila needed voltage GND VV GND yy STY VV 5 7 V GND GND analog GND 8 8V 5 5V analog unused 5 5V analog GND digital 5V digital Figure B 38 It is schematically shown how the lv distribution for the TacQuilas is realized The color code is shown Furthermore the values of the voltages which are needed on a TacQuila board are given as well as how they are labeled 131 APPENDIX B NEW LAND ELECTRONICS TACQUILA 2 pin connectors e g for clock Capacitors rd location of connector 7 to lv supplies 10 pin connectors for TacQuila Figure B 39 Shown is the FOPI lv distribution board There are two types of connectors on the board Two 2 pin and ten 10 pin connectors The 2 pin connectors are used to connect the clock and the LVTTL to PECL converter Here one has to be careful since since it easy to reverse the polarity The 10 pin connectors are used for the TacQuilas Figure B 40 Shown is one WAGO sub distribution during the production All Rail Mounted Terminal Blocks are fixed to the rail The cables for only one TacQuila system are connected 132 B 7 ADDITIONAL ELECTRONICS B 7 Additional Electro
91. d particles created in the iron is too low to pass the iron and reach the active volume of the MRPC From the good reproduction of the neutron energy distribution for the fast neutrons Glass fibre Reinforced Plastic Time to Digital Converter TDC TAC ADC charge Q to Digital Converter Field Programmable Gate Array 16 2 1 MRPC BASED NEUTRON DETECTOR CONCEPT 5000 counts 4000 3000 2000 1000 ol je ap golf open A NE eg ee pe ET t ao 6 100 120 140 160 180 200 220 240 Energy in MeV 6 Figure 2 5 The neutron energy reconstructed from the time of flight is shown Contin uous line measured blue circles adapted theoretical spectrum from PBB 07 it is concluded that an MRPC may be used as a time of flight wall for high energetic neutrons N nl n 2 n 3 counts m Micro Pulse Period 45 in 4000 3000 2000 1000 0 1000 2000 3000 4000 time in ch Figure 2 6 Mean time spectra of one RPC strip The width of one channel is 25 ps therefore the range represents exactly the 200 ns time window of the TDC The four observed peaks correspond to four beam micropulses within the 200 ns The detector tests reported here were done in parasitic mode Due to that the neutron beam intensities could not be adjusted to the data taking capabilities Therefore 17 CHAPTER 2 THE NEULAND TIME OF FLIGHT NEUTRON SPECTROMETER the dead time of the da
92. diagonal 2 E E 23500 Heimis Beeren Bemmmmneteren Bemmnssirgen te nn ien enenen eennertnon 2 H 3 Q a 4 2000 Leese J ere ret Tagine eae ee TER E VERDI EE EE 2 i E S a t 15ns t 100ns 9 i i 1500 N E wuusnunenun en po ta22NS ta One Poe a ee EE A SEEN ee ASETE PE oe 500 been a E EEEE ae L igs 0 L 1 1 1 i 1 L 1 1 i 1 1 1 1 i 1 1 1 I I 1 ii L 1 L 1 1 1 I 1 1 i 1 I 1 0 0 5 1 1 5 2 5 5 4 amplitude of injected pulse V Figure B 36 Shown is the correlation between injected and measured charge For details see text and table B 7 128 B 6 LOW VOLTAGE B 6 Low Voltage The lv distribution for the TacQuilas is done using five TDK LAMBDA power supplies GEN 8 180 LN Each of them can provide 0 8 V and 0 180 A The access and control of those power supplies is realized using EPICS The first power module master has an ethernet connection The other four modules are daisy chained to the master via RS485 standard The EPICS GUI can be found in u land LANDLV execute the following shell script to start it tacquila_startGUI sh The TacQuila lv GUI can also be opened via the standard LANDHV GUI There one will find a button for the TacQuilas at the right bottom A window like shown in figure B 37 will be opened This can be used to switch the supplies on and off and to monitor the voltage and current of each device it i
93. dth of the measured data is almost entirely due to experimental resolution and the shape of the data above 0 8 MeV is dominated by the limited acceptance at higher relative energies CFA 12a The data has been analyzed using a Breit Wigner line shape described by for mula 5 1 with a channel radius fixed at R 4 fm and A 0 T as well as E have been used as free fit parameters The amplitude is chosen such that the integral of the fit function and the experimental data match Methods A to E have been applied the 2 dimensional x distributions are shown in figure 6 2 For empty bins an error of 1 was used The width T has been changed between 0 and 500 keV the peak position E was varied between 500 and 1000 keV for both quantities 5 keV steps have been used The results obtained from the different fits are shown in figure 6 2 and given in table 6 1 An overall agreement on the peak position E and the width T can bee seen within error bars for the results obtained by the different methods The uncertainty in determining the width is mainly due to the fact that the observed width is dominated 71 CHAPTER 6 RESULTS by the experimental resolution see e g figure 5 5 Since it reduces the obtained errors dramatically A 0 has been used The results obtained including A and R 6 fm which is comparable to the Breit Wigner line shape used by Hoffman et al HBB 08 is shown in appendix A It can be seen that x
94. e The signals coming via the TRIDI or TRIVA bus have a jitter since they are synchronized to the internal clock of the corresponding module Furthermore one has to remember that the output trigger type which the SAM delivers to the TacQuilas has always to be 1 this means one has to do the mapping accordingly To see the current mapping which is used for a certain SAM one has to log on to the RIO which is controlling the VME crate of that SAM and do e g R3 30 land usr land tridis tridis2_rio3 a3 s1 if one is using a RIO3 and the HEX address of the TRIDI is 3 for more information see the help function of this program The SAM needs a certain firmware to work in conjunction with a TacQuila readout how to change this firmware can e g be found here u land lynx landexp SAM5 HowTo_flash_a_SAM5 txt The here presented VME setup explains all possible connections which might not all be used at the same time In stand alone mode the TRIDI and TRIVA bus will e g 138 B 7 ADDITIONAL ELECTRONICS not be connected while if the TacQuilas are readout in the main LAND DAQ the local VULOM is not needed In a minimal setup one could operate the TacQuilas having only a RIO TRIVA and SAM in the VME crate 139 APPENDIX B NEW LAND ELECTRONICS TACQUILA B 8 Miscellaneous In the following chapter the needed cabling will be summarized for a full system the SN of all systems which are currently in use will be given and furthermore
95. e 3 the dead time blocking The dead time free triggers are then in the here shown example connected again to an ENV cable 4 to be able to monitor them on a scope From the ENV the signal is routed to the TRID cable 5 The TRIDI delivers the trigger to the SAM cable 6 which then distributes it via the GTB to the TacQuilas For the interconnection between the TRIDI and the SAM one has to be a bit careful The SAM has always to see a trigger 1 this can be achieved by either changing software wise the mapping of the TRIDI or one does the same hardware wise using e g a FAN In FAN Out Furthermore one has to be aware that the mapping between the TRIDI inputs and the outputs on the TRIDI Ctrl Bus is done in a cross like manner TRIDI IN Ctrl Bus 1 t o NO OB W WD WM HM OND CELIT This is only important if one wants to look at signals in between the TRIDI and the SAM Since the mapping on the Ctrl bus inputs of the SAM are adapted to this cross mapping the signal routing works if one inter connects TRIDI and SAM using a standard flat 8 pin cable The VME crate used for LAND holds at the moment two SAMs see figure B 44 These two SAMs are used to have at least three GTB chains for the three LAND TacQuila VME crates As a consequence different HEX addresses have to be used for the device addressing on the VME bus for those SAMs Currently this is 5 for the first SAM and 6 for the second This offers four GTB connections GTB 5 0 GTB_5
96. e Plate Chamber 10 SIUI puy JOLT quouttodxe Tq MOA SIOZIPSIp OAV SITUO1I99 2 sng yseq aqny uM 040d sd114s oplM W G g ur poyuouIses ore s ppped OLIBUOIS SIY 0 SPUO SOLIOD 9I9U USATS NTLA AL 1 110g yuswsery jryp ford AAvoy wm pou e o p ds or yya AY 00G JO ASIoUO JALJUJOI V YPM POYEIHUIF 919M SUOIJNOU I93 1EI OY 0 WU GGT JO IIULJSTP e ye poye90 sem ANVTON APN 009 YAM poyemwms 9I9M SUOIJNON SUOLNIPUOI SULMOT OJ OY YAM OUO 3 9 SUOTJL NUIIS SUISN POTLIOA 98 NV TINON JO speo3 usisop aseo Up O UL 0 spuodso11o9 pue 869g OA user st GNVT JO onea oq dnyos feyusurmdxo at JO sIeY9P oY Pure oeus UOISSTUIO UOIMEU y UO ATsuo1s spu d p UOLITUF0991 UOLMOU TIMNU UL STeO3 usSIsep oy Juosa da ANY TNON 1037 U9AIZ sonJeA JL ANY TNON JO s 3doouoo omy oY 0 poreduiod are NYT SUTISTXO IJ JO SOTISTIEYEIEUD ALL T Z ALL W GE GT Ae Aed WZ amp YZILL IOIIIPA SOueIsIq Up 10 09 Up IO YOT UOTPUSOSOY UOLMON ININ APN 007 S6 APN 007 06 Kduon yq UT WD GT xo Ww 2 o UOTJNOSoY uorysod sd ocT gt 0 sd 0 amp 7 2 uUornJosoy OWL Ul GGT peru 08 Ul gI peru 08 UW gI peru 08 soueydoooy rem3uy wer WEEN wg 1099999 M4 Jo yydaq eIMOoeL qaqdaNv1 LNd eIMORL HHApIdOA oN get LNA mopeoy Joyejjtyursg HIYN wur p Iosu uo Toyefyurg wu G 1919AUOJ UOIT fellsyeN 0009 0091 007 Spuueqp jo IqunN gu 09
97. e dashed lines are the same as the dotted once shown in figure 1 3 a while the shaded area highlights how the neutron ESPE changes if three body 3N forces are included The calculations shown here include contributions from chiral low momentum interactions View x including the changes due to A excitations The three body mechanism due to A excitations can be pictured in the following way one nucleon virtually excites a second nucleon to the A resonance which is deexcited by scattering off a third nucleon Figure 1 3 d represents a visualization of those three body interactions one nu cleon belongs to the O core and the other two are valence nucleons Those 3N forces amongst two valence neutrons and one nucleon in the 60 core give rise to the repulsive interactions between the valence neutrons which make the 7 O unbound Including three body interactions amongst three valence neutrons is a topic of current research Contributions from three valence nucleon interactions are in general suppressed by Nyatence Neore FS11 those contributions will therefore only be important in the most neutron rich nuclei Making this effect significant for the here discussed oxygen isotopes New calculations performed by Holt et al Hol12 and Simonis et al Sim12 will be discussed in chapter 6 and 7 see e g figure 7 2 Those calculations include contributions from interactions amongst three valence neutrons and predict the O ground stat
98. e energy correctly CHAPTER 1 INTRODUCTION Chapter 2 The NeuLAND Time of Flight Neutron Spectrometer The investigation of short lived radioactive nuclei has to be done in inverse kinematics since targets for scattering experiments cannot be produced This makes it necessary to produce radioactive beams with sufficient intensities Two techniques are used to produce RIB s ISOLT and in flight The in flight technique was invented in the 1970s at LBNL Ct05 Since the 1990s more neutron rich nuclei are produced via in flight fission or projectile fragmentation of high energy heavy ion beams due to higher energy accelerators TS11 The main in flight RIB facilities and their separators nowadays which produce high energy beams via fragmentation are for example FRS at GSIT Germany RIPS at RIKEN Japan A1900 at NSCL USA SISSI at GANILS France Acculinna at FLNR Russia PKGR12 As discussed in the previous chapter there is an urgent need to investigate nuclei far from stability To be able to cover these exotic parts of the nuclei chart several second generation facilities are being constructed The first of these facilities the RIBF started operation in 2007 in Japan having BigRIPS as the successor of the above mentioned RIPS Other research centers which are currently in the development stage are Radiactive Ion Beam Ion Source On Line Lawrence Berkeley NationalLaboratory SFRagment Seperator TGe
99. e experimental results obtained in this work and the theoretical values from Simonis et al Sim12 and Holt et al Hol12 in red together with results from the literature in light blue Theoretical values are shown as lines while experimentally determined values are depicted as squares with error bars The error bars represent the experimental error and not the width of the state All theoretical values depicted in panel c represent calculations for the first excited 2 state The nature of the experimentally observed state is most likely different 86 16 O core Figure 7 2 Visualization of three body 3N forces The left figure A shows 3N inter actions between valence neutrons The right figure B depicts 3N interactions between two valence neutrons and one neutron from the core Figure is taken from Sch12a The obtained lifetime of the 7 O ground state is indicated in figure 7 3 b The values obtained from the measurement are indicated by grey Those values limit the available range to the area indicated as blue striped area A schematic level scheme depicting the determined energy levels of 2 O and 0 relative to the known states of 24O is shown in figure 7 4 Populating the O ground state only a direct decay to the O ground state is possible by passing 7 O Furthermore the lifetime obtained for the 7 O ground state is rather long This value is twelve orders of magnitude larger than the lifetime found for the
100. e middle column shows the voltage which defines the threshold for the comparator While the right column shows the voltage how they are measured on the TRIPLEX card These values are helpful because they can be measured on the board while everything is mounted and running See figure B 14 for a graphical representation of the table TRIPLEX Interface IN pulser OUT Mul AMux DMux analogue sum OR R W I2C on Flat cable I2C via round connector 12C to USB converter Network USB over IP server Figure B 16 A schematic of the TRIPLEX setup is shown One can see how the TRIPLEX tree is connected to the network via the I2C to USB converter and the USB over IP server 111 APPENDIX B NEW LAND ELECTRONICS TACQUILA To get an idea about the timing of the TRIPLEX chain a measurement with 1 card was done The delay between a pulser input connected to the FEE and the lemo Mul output of the TRIPLEX interface amounts to 42 ns This measurement was done using a TacQuila system whose TRIPLEX card was at the topmost position in the TRIPLEX tree Using a card which is in lowest 6 level of the tree adds 40 ns of delay Possible Settings and connections on the TRIPLEX board itself are shown in figure B 6 and B 15 The TRIPLEX board has several LEDs At the moment only the first two are used The first LED indicates that one of the corresponding FEE channels fired while the second one shows w
101. each in beam DSSSD delivers a position measurement in x direction and the second side gives the y position To deduce the angle 0 of the incoming ion also the distance in z coordinate between the two detectors has to be known to a very good precision The exact positions of all detectors shown in figure 3 3 have been determined using photogrammetric techniques The results of this procedure are shown in table 3 1 HV11 Having x y and z position 0 of the incoming ion can be deduced The next material which the ions pass is the actual reaction target The targets are mounted on a remote controlled target wheel offering the possibility to change 32 3 3 LAND R B SETUP AT CAVE C the targets in vacuum without beam breaks The used targets for the here presented beam setting are given in table 3 2 detector name distance to center error due to photogrammetric of ALADIN in cm technique in cm DSSSD 1 266 0 0 05 DSSSD 2 263 3 0 05 target 259 6 0 05 DSSSD 3 248 5 0 05 DSSSD 4 246 0 0 05 GFI1 196 6 I GFI 2 351 1 Il TEW 874 0 2 LAND 1268 0 5 Table 3 1 Distance in z direction of each detector along the nominal beam trajectory relative to the center of ALADIN target area density in mg cm CH 922 C 935 Pb 2145 Table 3 2 Reaction targets as used in experiment s393 Since the reaction mechanism is not of major impact here data take
102. eded to get a correct deposited energy of the shower A hit has to fulfill the following conditions to be considered as a secondary hit The distance in space to the primary hit has to be smaller than Rmaz The distance in time to the primary hit has to be smaller than Tmaz Due to the time resolution of LAND it is not possible to discriminate hits which come too close to each other For that reason all hits which are inside a sphere determined by the time resolution and the speed of light and in which therefore causality cannot be checked are assigned as secondary hits A cylindrical cut is applied The cylinder is defined by a depth which is given by a distance in beam direction Zmaz and against beam direction Zmin and as third parameter the radius R y Within this cylindrical cut it is checked if the location of the hit can be reached from the primary hit by a velocity smaller than the speed of light If this condition is fullfilled the hit is treated as a secondary A further condition called backward Fermi tests two requirements First the velocity connecting two hits has to be smaller than vYprermi which is the velocity corresponding to the fermi momentum of a nucleon inside the nucleus And second it is tested if the relative distance between the two hits is negative meaning that the potential secondary hit is backward relative to the beam direction The last condition checks if the neutron was sc
103. ee i I EN E 143 B 9 Known Tssties i 2 22 2 ae ee ee oe a na 144 C Data Sheets of MRPC Prototypes 149 D Acronyms 159 Bibliography 163 Chapter 1 Introduction In May 1911 Ernest Rutherford published his discovery of the atomic nucleus and founded thereby the research field of Nuclear Physics In 1932 the discovery of the neutron by Chadwick proved that a nucleus is build of neutrons and protons In the past hundred years scientists have addressed several questions concerning these build ing blocks of nuclei and many experiments have led to a deeper understanding of the nucleus and the nuclear force but there are still questions which have not been fully answered yet TS11 such as e How forces hold together protons and neutrons e How many neutrons can be bound in a nucleus for a given number of protons To address these questions it is of particular interest to investigate nuclei with large pro ton to neutron asymmetries In this thesis neutron rich oxygen isotopes are investigated Their particular importance is highlighted within the following two citations 1 The limit of neutron rich nuclei the neutron drip line evolves regularly from light to medium mass nuclei except for a striking anomaly in the oxygen iso topes OSH 10 This feature of nuclear structure was first pointed out 1999 in a Letter by Sakurai et al SLNT99 in which the sudden change in stability from oxygen to fluorine was reported
104. ee of VETO As mentioned already thirty TacQuila systems are mounted currently at the LAND setup Twentyfive of those are used for LAND itself and are connected to one TRIPLEX tree see Sec B 3 1 The remaining five systems are connected to a second TRIPLEX tree This is here called Veto Tree where Veto refers to the additional scintillator plane mounted infront of the actually LAND detector which is used as a veto for charged particles Out of the five TacQuila systems three are used for this Veto detector while the other two cards are used for test systems and as backup TRIPLEX connector names J6 Figure B 24 Shown is which parts of the address space for the TRIPLEX tree is used for Veto and as backup Where the cards can be found in the crate can be seen in figure B 23 119 APPENDIX B NEW LAND ELECTRONICS TACQUILA B 4 TacQuila The original TacQuila board developed for FOPI has 16 channels Bach of them mea sures a time relative to a external clock at the moment a 40 MHz clock is used As a result the time measurement is limited to a time smaller than 25 ns The R B TacQuila boards have an additional 17 channel This is why the board is also called TacQuila Using this additional channel the time measurement is divided in a fine measurement relative to the clock and a coarse measure The coarse measure counts the amount of clock cycles between
105. el after the lithium target by a dipole magnet The experimental area extends from about 3 to 15 m behind the neutron production target The relative neutron beam intensity is monitored by the integrated proton beam current at the beam dump and by TFBC JPSTO1 Thin Film Breakdown Counters 14 2 1 MRPC BASED NEUTRON DETECTOR CONCEPT mounted in the neutron beam Another monitoring option is provided by an ICM Both monitors utilize neutron induced fission of SU with the cross section adapted as neutron flux standard PBB 07 A typical neutron rate at the detector position was 4 kHz cm The long neutron flight path in the experimental area allows for several irradiation positions The permanently installed MEDLEY DABT00 and the MRPC setup used the neutron beam simultaneously The MRPC setup was placed such that the first detector was 11 m behind the lithium target 1 Macro Pulse Period 5 4 ms L 2 Macro Pulse Width 0 81 ms expanded view tee aAa LEES Micro Pulse Width 3 7 ns gt 4 Micro Pulse Period 45 ns Figure 2 3 The time structure of the primary proton beam which defines also the time structure of the secondary neutron beam is schematically shown The time structure of the neutron beam is defined by the time structure of the proton beam from the cyclotron The beam has a macro structure with a repetition rate of 185 Hz and a beam pulse duration of 810 us
106. er IP server e The TRIPLEX cards have to be connected among each other see Sec B 3 e The pulser input should be connected to the TRIPLEX interface The Triplex pulser events can be used to determine the pedestal This is the only way to create a zero charge event white cable brown cable empty Figure B 46 The connector for the 17 channel on the TacQuila board is shown It is indicated how to connect the 3 pin cable which is used to feed the 17 channel The developers had up to 30 systems in one chain but we faced data transfer problems having more then 10 141 APPENDIX B NEW LAND ELECTRONICS TACQUILA Figure B 47 The connector for the clock on the TacQuila board is shown It is indicated white cable 2 e ground Fi ground brown cable how to connect the 4 pin cable which is used for the clock B 8 2 Used TacQuila Systems Building up the new LAND readout electronics 30 TacQuila systems have been mounted in 3 Crates In this section the SN of all sub modules will be given This information should in the future be integrated into the LAND cabling documentation position TRIPLEX FEE TacQuila QDC HEX used for Cltl ae 1539 514 2514 8 LAND C1t2 x9 1509 430 2530 3 LAND C1t3 14 1534 509 2509 3 LAND C1t4 3 1516 526 2532 5 LAND C1t5 9 1501 502 2502 E LAND C1t6 2 1536 507 2507 2 LAN
107. es it bends slightly Lower right pad generated energy vs resolution of the tracked energy is shown 61 CHAPTER 5 ANALYSIS 5 2 Breit Wigner Line Shape After discussing in the last section 5 1 how the detector response matrix is obtained the next step is to convolute it with an input function In the final step this test model is then compared to the experimental data using x methods described in section 5 3 The standard function to describe a resonance is the so called Breit Wigner res onance This line shape as given in the one level approximation is a function which depends on the relative energy E and two fit parameters e resonance energy Er e reduced width y For the amplitude an additional factor is later on used which scales the integral of the function to match the experimentally found integral The Breit Wigner line shape has the following form LT58 T E A E 1 4 T f E Er 9 5 1 with e T 2P E 9 o A S E S E The energy E and angular momentum l dependent functions P E and S E can be taken from BM69 For the angular momentum 2 is used due to the following argument The Breit Wigner line shape is used to describe the 25O resonance The additional neutron of O compared to O is in the 0d 2 Shell while for 240 the highest occupied neutron energy level is 1s 1 2 see figure 1 2 Therefore one finds for the O deca
108. example the eight fold cables of LAND pin 1 on cable 1 would be TacQuila channel 8 101 APPENDIX B NEW LAND ELECTRONICS TACQUILA B 3 TRIPLEX The initial name of the TRIPLEX board during the development stage was FEE pig gyback board the board is shown in figure B 6 The nowadays used name could be interpreted as TRlIgger and multiPLEXer board The TRIPLEX offers the following features e individual threshold for each of the 16 channels on the FEE e Mul signal of the full TRIPLEX tree e analogue sum of the full TRIPLEX tree e Or signal of the full TRIPLEX tree e A pulser to fire the timing branch of an individual several or all channels in the TRIPLEX tree e A multiplexer to look at an individual signal as well as e g at Mul signals of parts of the TRIPLEX tree The TRIPLEX cards are connected in a tree like structure via I C see e g figure B 7 The most top module in the tree is connected to the so called TRIPLEX interface which is a modified TRIPLEX board in a housing which holds several additional in and outputs see figure B 8 The cable used to connect the first TRIPLEX card to the TRIPLEX interface should be as short as possible Using e g a 2 m long cable caused severe problems in the test phase The in and outputs of the TRIPLEX interface prototype shown in figure B 8 are e Pulser input the here connected pulser can be distributed either to all TacQuila channels in the tree or to individual
109. exer PCA9547 is shown This part can be found on each TRIPLEX board labeled as PART U11 see e g figure B 15 0 1 Oo 0 A2 Al AO RW on pe FIXED HARDWARE SELECTABLE Figure B 20 The device addressing for the 8 bit IC I O port PCA9554 is shown Two of these devices are mounted on each TRIPLEX board labeled as PART U12 and PART U13 see e g figure B 15 HEX switch multiplexer addr register 1 addr register 2 addr 0 70 20 20 1 71 21 20 2 72 22 21 3 73 23 21 4 74 24 22 5 75 25 22 6 76 26 23 7 77 27 23 8 70 20 24 9 71 21 24 A 72 22 25 B 73 23 25 C 74 24 26 D 75 25 26 E 76 26 27 F 77 27 27 Table B 6 In this table the addresses of the multiplexer and both registers on the C bus are shown All numbers are in HEX The combinations of addresses are a direct consequence of the chosen address on the HEX switch Since for the HEX switch addresses 0 and F both registers on one card have the same address these HEX addresses can never be used so one has to consider that on each path each address can only be used once Currently we use a tree which allows to combine 40 TRIPLEX cards to one TRIPLEX interface this tree is shown in figure B 21 At the current LAND setup three TacQuila crates are in operation each crate holds ten 114 B 3 TRIPLEX TacQuilas twentyfive of those are used for LAND itself The remaining five systems are used for Veto see
110. exp aug2010 tacquila multi_land setup usf in For the s394 AsyEOS experiment e g 1 10 was chosen Dicke engl thick Time of Flight wall Since the time calibrations of the TacQuilas is done in a different way than for all other systems in the LAND setup the program looses it s generic features here If problems occur please contact R Plag 95 APPENDIX B NEW LAND ELECTRONICS TACQUILA For each GTB on each SAM a 32 bit header word were the bits mean the following 31 28 27 24 23 20 19 16 15 12 8 0 SAM id GTB id local event counter trigger type as seen by SAM 1 physics 2 calibration trigger type as seen by TacQuila 0 physics 1 calibration data length on this SAM GTB in 32 bit words excluding this header word For each fired TacQuila channel two time and charge 32 bit words first word time 31 27 26 22 21 20 19 12 11 0 TacQuila GTB address channel number on TacQuila card 0 16 0 indicating first TacQuila channel data word 0 physics 1 calibration trigger clock cycle counter hit time TAC second word charge 31 27 26 22 21 20 11 0 TacQuila GTB address channel number on TacQuila card 0 16 1 indicating second TacQuila channel data word 0 physics 1 calibration trigger QDC data Table B 1 TacQuila data structure in MBS Rdtxmul how many paddles fired Rdtxmuli 6 which paddle fired Rdtxt01 6 TAC measurement PMT1
111. f a time measurement which is done using a random common stop like it is the case for the TacQuilas The input at one of the channels is delayed relative to the other one by 12 ns amp half a clock cycle From such a measurement one gets two types of events like shown in figure B 29 In case A both TAC values are measured relative to the same clock rising edge But in case B relative to two successive ones The time difference t is calculated using ti gt tea ta t te B 3 ti lt t2 ta tak t1 t2 B 4 with tag clock cycle 25 ns case A case B reference clock reference clock el e2 el e2 Figure B 29 Shown are the two types of events which one can get if two events el and e2 with a fixed delay 12 ns are used as an input Either they are measured relative to the same rising edge of the clock case A or they are measured versus the rising edges of two successive clock cycles case B Looking at a time difference spectra see figure B 30 one gets two contributions which reflect the two types of events shown in figure B 29 For this example the resolution is 24 ps it was chosen to illustrate the two types of events The here presented way in which the time difference between two events is limited to 25 ns is exactly how the TacQuila readout is used by FOPI As mentioned already in the introduction a time resolution on one Ta
112. f ALADIN an upper limit on the lifetime can be estimated Including the relativistic Lorentz factor y one gets from formula 6 2 t y T inNo N Se which means that we can obtain the lifetime 7 if the number of surviving nuclei N t at time t can be determined The distance from the target to the middle of ALADIN is given in table 3 1 and amounts to 256 cm The beam velocity is given on page 47 and is 8 0 722 Using these values results in a ToF of 11 8 ns for the 6O ions from center target to center ALADIN To estimate the lifetime using formula 6 3 N t and No have to be known in addition to the ToF No can be obtained from the observed number of events belonging to the 7 O ground state and the known efficiency x acceptance Six events can be attributed to the 2 O ground state looking at the 2n channel see figure 6 3 using the simulation to obtain the efficiency xacceptance a initial number of No 20 5 is found N t ToF 1 is used due to the following argument If the 7O did not decay at the time were it reaches the center of ALADIN two characteristics have to be present for such an event 1 It has to be at A gt 24 in the mass identification plot 2 Furthermore no neutron should be observed in LAND since the decay happens after the ion has been bend off the zero degree line far enough to be outside the acceptance of the neutron detector The center of ALADIN has been chosen as threshold exactly for that reason At
113. g are commonly applied while analyzing data taken with the LAND R B setup 1 generic tracker 2 reference tracking The first method e g described in Wam11 uses a tracking technique based on the knowledge of the magnetic field and absolute detector positions and can be employed using a standalone program developed by R Plag The second approach e g described in Ros09 uses the trajectory of a known beam A Z as reference and translates deviations from this trajectory into A Bp and therefore mass values using a first order transport matrix For the light nuclei used in the presented experiment a third method called here empirical mass becomes applicable ZHN 97 It will be described in detail in this section The reference tracking is mostly used if outgoing and incoming reference ions have the same charge number and hence is not suited for the physics channels analyzed here The empirical mass formula has two advantages compared to the generic tracker the usage of an external program is not needed and it easier to use since the procedure of calibrating the tracker is not needed By looking at the position of one ion on one GFI vs the angle of its track relative to the beam axis measured with both GFIs like shown in figure 4 3 the different masses can be separated already without applying full tracking To prepare the shown figure an incoming cut on B was applied as well as a cut on the outgoing charge Z
114. geregter Zustand E 4225 227 keV iv 4 T 0 GS 9485 10712 ns und 5 r 260 GS lt 5 7 ns Theoretische Berechnungen auf Grundlage von chiral Effective Field Theory EFT stimmen sehr gut mit den experimentell bestimmten Werten f r die Grundzustandsener gien beider Sauerstoff Isotope berein Zwei K rper NN und drei K rper 3N Wechselwirkungen wurden hierbei ber ck sichtigt F r den 3N Anteil wurden folgende Beitr ge ber cksichtigt Zwei Neutronen befinden sich im 60 Kern und eines im Valenz Raum effektiver Ein K rper Anteil ein Neutron befindet sich im 160 Kern und zwei im Valenz Raum effektiver Zwei K rper Anteil alle drei Neutronen befinden sich im Valenz Raum Residual Anteil Hol12 Sim12 Die ungebundenen Resoanzen von O und 7 O wurden k rzlich auch von anderen Kollaborationen untersucht HBB 08 LDK 12 Die bestimmte Lage der 2 O Grundzu stands Resonanz von HBB 08 und die Lage welche in dieser Doktorarbeit bestimmt wurde stimmen sehr gut berein Die Breite der 7 O Grundzustands Resonanz ist in der vorliegenden Arbeit um einen Faktor 2 kleiner als in HBB 08 angegeben Die hier bestimmte Breite stimmt gut mit der Ein Teilchen Breite berein welche mit Hilfe der Breit Wigner Resonanz bestimmt wurde F r die 7 O Grundzustands Resonanz wurde zuvor eine obere Grenze von 200 keV LDK 12 bestimmt Das hier pr sentierte Expe riment reduziert diese obere Grenze auf 50 keV Au erdem legen die
115. gets The decay products of the unbound O and 2 O systems have been measured in inverse kinematics using a complex detector system which allows for kinematically complete measurements including y ray detection Both neutron rich oxygen isotopes decay via the emission of neutrons which were detected in the LAND detector The present analysis concentrates on the relative energy between neutrons and fragments Structure has been observed in the In continuum of the 74O n system as well as in the O 2n relative energy spectrum The 24O n relative energy spectrum is described using a single resonance reflecting the ground state In the 4O 2n relative energy spectrum two resonances are used to describe the ground state resonance as well as a excited state For both ground states the lifetimes have been extracted for the O ground state the lifetime was directly measured via the width of the state which has been obtained from a fit to the data using a Breit Wigner line shape in the one level approximation For the 6O ground state an upper limit on the lifetime was determined using the Time of Flight ToF from the target to the ALADIN magnet In summary the obtained results are 1 50 ground state Er ala keV and OE keV 2 260 ground state E 25 keV 3 260 excited state E 42251221 keV 4 T 0 GS 974 1071 ns and 5 r 260 GS lt 5 7 ns Theoretical calculations using chiral Effective Field Theory EFT give a
116. h four momenta P E c pj is given by the following equation Mi j 2 3 n 3 1 from which follows Fj E Min 2 m J 1 Be cos lt i j 3 2 ij tAj where lt i j is the angle between the momentum vectors of particle 2 and 7 Using Ean e 3 3 and Pi Yi Mi Bic 3 4 equation 3 2 can be written as follows Mi 2m gt ymm 1 bib cos lt fi j 3 5 ij from which the relative energy Erel can then be obtained by subtracting the masses of the decay products at rest from the invariant mass Miny of the system Evel 2m gt yyy 1 bib cos lt fi j CO 3 6 ifj Applying this formula to the 2 O case one obtains for example Epel r i m2 2 yao nMauoMnll Baron cos lt 4 240 n Mzao mn e 3 7 26 3 2 RIB PRODUCTION GSI AND FRS 3 2 RIB Production GSI and FRS The s393 experimental campaign has been performed at GSI where the RIB is pro duced via the so called in flight technique meaning the radioactive ions are produced and separated in flight This beam production will be briefly explained in the follow ing paragraph The production mechanism of the RIB starts with a stable so called primary beam At GSI the ions of choice are injected into the UNILAC from an ion source In the experiment described here Ar ions have been used as primary beam From the UNILAC the 4 Ar beam is injected into the
117. harge measurement for each event The TacQuila measures the time relative to an external clock Each channel starts its own time measurement and the next rising edge of the clock serves as a common stop Furthermore the time measurement is split into a fine and a coarse measure This means in addition to the measurement relative to the next clock cycle also a measurement how many clock cycles passed between the event and the DAQ master trigger is done For that reason the 17 channel is a copy of the DAQ master trigger and allows therefore to measure times longer than one clock cycle for a schematic of the time measurement see figure 2 10 this figure is also discussed in appendix B 4 reference clock channel 1 channel 2 g u channel 17 en 7 lt gt tac_17 trigger on GTB Figure 2 10 Shown is how the TacQuila timing works for details see text The QDC has a virtual gate meaning a delay is software wise applied in a way that the QDC is read out eight clock cycles 200 ns after the corresponding comparator fired For more details on the QDC see appendix B 5 The individual TacQuila systems are connected via a GTB chain and then to a VME module called SAM which is readout using an MBS based DAQ A charge measurement is only available together with a valid time meaning that if the threshold of the comparator has not been crossed also the charge is not read ou
118. he O data the probability func tions derived while applying method F see section 5 3 are directly used The most likely value is used as peak position E The error is determined using the maximum probability found in the distribution and lowering it such that 68 of the area is under neath the function The probability distributions for the two found 7 O states are shown in figure 6 6 For each state the individual distribution is normalized to one For the ground state 80 probability is reached within the first bin For the excited state the 68 range is indicated as a blue dashed area The found value of A005 oe MeV is in agreement with the values observed from the parabola fits to the y distributions 76 6 2 ANALYSIS OF THE O CHANNEL method x7 at min E at min in keV A 8 0 25 58 B 78 1 25 9 C 46 6 25 24 D 32 8 48 1 0 7 25 24 E 34 4 25 13 F 0 5 25 16 Table 6 2 Given are the values obtained describing the low energy part of the 7 O In and 2n channel using a flat test distribution having a width of 50 keV NDF is the number of bins in the fit range minus the number of free parameters in the input function The error is obtained using a linear interpolation between the two lowest points and using the energy at x 1 method x7 at min E at min in MeV A 5 3 4 21 0 65 B 38 7 4 40 0 08 C 36 8 4 27 0 24 D 22 8 60 1 0 39 4 27
119. he relative energy of the fragment plus neutron s system Therefore the four momenta of both neutron s and fragment have to be measured To be able to reconstruct those the following parameters have to be known 1 Fragment velocity Particle IDentification 30 3 3 LAND R B SETUP AT CAVE C Fragment charge number Z Fragment mass number A Fragment trajectory Neutron velocity Neutron trajectory How these quantities are derived using the LAND R B setup will be explained in the following section 3 3 1 Identification of Incoming Particles To identify the incoming ions the mass to charge ratio A Z and the charge number Z are measured From formula 3 8 one sees that A Bp 1 3 9 Z const By 3 9 Therefore determining A Z means measuring 8 and Bp The value of Bp is known from the FRS setting The velocity of the incoming ion 8 is measured using a plastic scintillator paddle at S8 as start detector This detector is read out with two PMTs one at the top and one at the bottom As stop detector a square shaped 2 5 cm x 2 5 cm plastic scintillator with thickness of 1 mm at the entrance of Cave C is used This detector is called POS and is read out with four PMTs one for each side The distance between these two detectors is about 55 m which results in a very good velocity resolution The S8 detector is shown in figure 3 2 while POS is the first plastic scintillator shown to the left
120. hen connected to a scaler to monitor the clock frequency See also www gsi de informationen wti ee elekt_entwicklung clockdstr html to get further details on the clock modules e LEVCON see http www gsi de informationen wti ee elekt_entwicklung levcon html for details on the module In the LAND setup it is currently used to feed all three SIS Clk Distributions from the same source and to connect the clock to a scaler See figure B 43 The used LEVCON has been specially programmed to be suited for the here shown task In addition to the above mentioned modules a NIM crate is located close to LAND it holds several modules which are e g used to distribute a pulser or construct a trigger In the NIM Crate the following modules can e g be found programmable LEVel CONverter Bunch phase Timing System Nuclear Instrumentation Module standard 133 APPENDIX B NEW LAND ELECTRONICS TACQUILA LA8000 Level Adapter Converts from NIM to TTL Is needed for the pulser inputs of the TRIPLEX interface CF8000 Constant Fraction Is used to discriminate the Mul signal delivered by the TRIPLEX interface Furthermore some VME modules are necessary to complete the readout electronics see figure B 44 RIO3 computer in VME crate MBS runs for example on this ma chine TRIVA5 distributes DAQ trigger tells the DAQ at what time it has to read out the data TRIDI1 distributes FEE trigger here to TacQuila
121. hen the TRIPLEX pulser is used The I C to USB converter and USB over IP server might in the future be replaced by a modified version of the HADSHOPOMO which will be used as a I C to Ethernet con verter Furthermore to simplify the Slow Control of the TacQuilas the Triplex settings will be integrated into EPICS B 3 1 TRIPLEX TC Address Tree of LAND To understand the cabling scheme of the TRIPLEX tree which is used for LAND first the addressing within this tree has to be explained Bach TRIPLEX board holds two registers Part U12 and Part U13 see figure B 15 The settings of a TRIPLEX board e g thresholds are written into these registers To do this one has to be able to address them This is done via the I C bus which is directed using multiplexers on each TRIPLEX card The information flow on one Y cross in the I C bus is exemplary shown in figure B 17 the full TRIPLEX tree is build out of many such crosses The way how this is realized at the moment creates some limitations for the available address space A nicer solution is shown in figure B 18 Using this scheme for the information routing not each register is on the main path and therefore the addresses could be used several times within one branch of the tree The addresses of the multiplexer and both registers are defined via the HEX switch on the TRIPLEX board How the device addressing has to be done for the multiplexer and the register is shown in figure B 19 and figure B 20 res
122. hor D Bazin B A Brown C M F III A Gade T N Ginter M Hausmann M Mato D J Morrissey M Por tillo A Schiller B M Sherrill A Stolz O B Tarasov and M Thoennessen Discovery of Mg and Al suggests neutron drip line slant towards heavier isotopes NATURE 449 1022 1024 October 2007 S Baker and R D Cousins Clarification of the use of CHI square and like lihood functions in fits to histograms Nuclear Instruments and Methods in Physics Research 221 437 442 April 1984 Th Blaich Th W Elze H Emling H Freiesleben K Grimm W Henning R Holzmann G Ickert J G Keller H Klingler W Kneissl R K nig R Kulessa J V Kratz D Lambrecht J S Lange Y Leifels E Lubkiewicz M Proft W Prokopowicz C Sch tter R Schmidt H Spies K Stelzer J Stroth W Walus E Wajda H J Wollersheim M Zinser and E Zude A large area detector for high energy neutrons Nuclear Instruments and Methods in Physics Research Section A Accelerators Spectrometers Detec tors and Associated Equipment 314 1 136 154 1992 K Boretzky A Gr nschlo S Ilievski P Adrich T Aumann C A Bertu lani J Cub W Dostal B Eberlein T W Elze H Emling M Fallot J Holeczek R Holzmann C Kozhuharov J V Kratz R Kulessa Y Leifels A Leistenschneider E Lubkiewicz S Mordechai T Ohtsuki P Reiter H Simon K Stelzer J Stroth K S mmerer A Surowiec E Wajda and 163
123. hosen as test functions to describe the low energy part of the 6O In and 2n channel The minimum is for all methods found at the lowest value 25 keV the extracted results are given in table 6 2 78 6 2 ANALYSIS OF THE 0 CHANNEL az 13 7 6 5 6 5 5 5 Er ll 0 0 0 Ds a 3000 3500 4000 4500 5000 5500 6000 3000 3500 4000 4500 5000 5500 6000 center of orig distribution in keV center of orig distribution in keV eee z an ag 5 60 55 50 45 40 35 Ds l ss lsssalsssalsssylsun alas td 3000 3500 4000 4500 5000 5500 6000 3000 3500 4000 4500 5000 5500 6000 center of orig distribution in keV center of orig distribution in keV a Q a a am Nee ee ree es ee en Be a 3000 3500 4000 4500 5000 5500 6000 3000 3500 4000 4500 5000 5500 6000 center of orig distribution in keV center of orig distribution in keV Figure 6 5 Shown is x for the different methods as a function of the central energy of the test input distribution Flat distributions with a width of 50 keV are chosen as test functions to describe the O 2n channel in a range between 2 and 7 MeV The extracted results are given in table 6 3 and discussed in the text 79 CHAPTER 6 RESULTS 1 Kernen i menu nenn theory prediction ground state i theory prediction first excited state Probability NN 3N NN 3N residual Prob Func determined using exp data 0 MeV lt E lt 2 MeV most
124. ical black lines The blue arrow symbolizes an inci dent neutron which creates a secondary proton green arrow via an reaction in the iron converter 2 5 cm wide and 40 cm long The strips are separated by 1 mm thick GRP material A standard gas mixture was used 85 CaHaF4 10 SFe 5 iso C4Hjo For a more detailed description of the MRPC prototypes see Bor11 The FOPI FEE card CSC 07 was used together with a V1290N CAEN TDC a V965 CAEN QDC and a V1495 CAEN FPGA used as scaler and trigger logic to read out the prototypes As a first analysis step the neutron energy spectrum was reconstructed using the ToF method The result is shown in figure 2 5 In an earlier experiment at TSL PBB 07 the neutron energy spectrum was measured at 142 7 MeV and compared to model cal culations folded with the experimental response This theoretical curve has been ex trapolated to 175 MeV see blue circles in figure 2 5 and used for comparison to the measurement presented here neglecting the differences in the experimental response e g time resolution Overall a good agreement between the spectra is found The difference in the low energy part can be explained with two effects e The ToF measurement is limited to a maximum time by the required coincidence between the detector and the beam signals e The efficiency of the counter decreases for lower energies This is the case because the energy of a substantial part of the secondary charge
125. ifier and a total gain of 160 Measuring the signal using a probe infront and 97 APPENDIX B NEW LAND ELECTRONICS TACQUILA Input 1 to 8 Input 9 to 16 IN9 16 1 47085665 AI Ef h 2 30 BERETS BRRRRRRERT BRRBERR gt GRCERRREREE Ji tacquilal7 analogue signals to TRIPLEX and QDC digital signals to TRIPLEX thresholdsettings from TRIPLEX to FEE Figure B 1 The LAND FEE is shown On the left side the input signals coming e g from a detector have to be connected The TRIPLEX board goes on top on the three connectors while the TacQuila board is connected using the connector on the right side On these connectors the FEE is also supplied with lv input amplitude mV output amplitude mV 7 20 15 40 37 90 75 140 100 160 200 180 gt 200 180 190 Table B 4 Correlation between amplitude of input and output signals of the amplifier behind the amplifier gave different results see table B 4 The amplifier is designed such that it delivers a very fast rise time to achieve the optimal time resolution but this results on the other hand in a small linear range see table B 4 and figure B 4 this means the amplifier goes very fast into saturation How the currently used amplifier for example affects the shape of a too large input signal is visualized in figure B 4 98 B 2 LAND FEE
126. in the order of 1 and data acquisition rates are limited This leads to the fact that the unbiased triggers are downscaled These downscale factors have to be known very precisely to be e g able to calculate a cross section later on The most important triggers as well as their downscale factors for the most neutron rich beam setting are given in table 3 3 The information on the trigger type is written to the data in a bit pattern The experiment described here was the first LAND R B experiment which used a VULOM 4 based trigger logics called TRLO IT This means the trigger matrix coin cidences were set via FPGA software Besides the physics triggers which all include a spill ON condition two calibration triggers are used Namely one trigger which is used for time calibration tcal of all TDCs and a second trigger which issues a zero charge measurement for all QDCs called clock trigger For more details on the LAND Cave C DAQ the reader is referred to Joh10 and Joh06 VME Universal LOgic Module 36 3 4 DATA ACQUISITION input to TRLO ja g 5 la ziz 3 o 2 lt Oo Par pe ape Se trigger bit objective a oo downscale 1 minimum bias x x 26 2 fragment x x x 24 5 CB sum x x x 20 8 LAND x x x 20 Table 3 3 A part of the trigger matrix is shown to present the most important triggers The full matrix contains 20 i
127. ing on each of the five cards AMux and DMux are the multiplexer outputs The AMux is used to look at an individual signal after the amplifier While the DMux delivers analogue sum OR and Mul signals of the tree but only up to a certain card which can be selected via slow control settings The analogue sum signal of the full tree The OR signal of the full tree The two lemo connectors without any label are used to program which of the connectors J5 or J7 is used as the start for the TRIPLEX tree This is done by sending a logical signal to these inputs These inputs should be terminated The default setting is that J5 is used The I C bus is used to communicate in both ways get data write settings to the TRIPLEX 103 APPENDIX B NEW LAND ELECTRONICS TACQUILA PC DC connector J5 connector J7 Baseline of Mul Lj Amplification of Mul gy Sur j Baseline of analogue sum GSI Darmstadt Triplex2 4910 SV KK 05 2010 Multiplicity Rai m 10 ug A MUX D MUX PECL toLVTTL j Qs Rz Amplification of analogue sum Connetor for cable to QDC a Fe HEX switch Li EFEEEEEFEEEEE u EEEE J6 f il i I2C connector J6 Figure B 6 On the left side the full TRIPLEX board is shown On the right side some components are highlighted to show the possible connections and where on the TRIPLEX card one can set the following Baseline of Mul signal Amplific
128. is transported to the experimental hall Cave C where the reaction target is located The cocktail beam allows simultaneous measure ments using a variety of different incoming ions but it also makes an event by event incoming PID mandatory A schematic drawing of the detector setup as used for the analysis presented here is shown in figure 3 3 For the incoming PID the following two quantities have to be known event by event for each ion 1 velocity 8 2 charge number Z Furthermore the incoming angle is measured for each event as well The magnetic rigidity Bp has for the experiment described here not been determined event by event This is the case since the FRS yields a precision of Bp 1 which is good enough to resolve the light ions investigated here As a consequence the nominal value is used for all ions this value is defined through the chosen FRS setting Nal sphere t x y z AE beam direction Plastic Scintillator Position Sensitive A Pin Diode g dipole magnet tx y AE Fiber Detector N DSSSD Figure 3 3 The experimental setup in Cave C as used during the s393 campaign is shown Note that only detectors are shown which are of interest for the analysis presented here In the figure it is also indicated which quantities are extracted from which detection system The final goal of the analysis is to reconstruct t
129. l is measured with pickup strips separated from the HV foil The HV has to be applied through a non perfect conductor in order to be transparent to the induced signal The first two gap RPC was developed in 1988 CSdL88 leading to an increase in efficiency and being the first so called MRPC Using MRPCs several test experiments have been conducted during the NeuLAND R amp D phase at e g KVI Groningen using proton beams Ros09 HZDR Dresden using electron beams ELBE YBR12 YAB 11 and TSL Uppsala using neutron beams CAB 12 Prototypes have been built in different member institutes of the NeuLAND working group namely GSI HZDR and SINP Results on the GSI MRPC prototypes tested at ELBE Dresden can be found in appendix C A short description of the experiment performed at TSL Uppsala in November 2009 will be given in the following paragraph A quasi monoenergetic neutron beam with an energy E 175 MeV was directed onto a 2 x 4 gap prototype operated at E 100 kV cm The neutron energy of 175 MeV used during this experiment is at the lower limit of the energy range 0 2 1 GeV for which the future detector is designed for The experiment had two goals on the one hand it aimed to proof that an RPC can be used to detect high energy neutrons at all this has never been done before and on the other hand the neutron detection efficiency of the MRPC prototypes should be determined In total eight different detectors were tested
130. l_slope u tcal_offset u rear En phasel cosmicl velocity lt time pos DHIT time energy pos data level reconstruction Figure 4 1 Shown are the different data levels as well as the routin 77 gt input data calibration routine parameter es and parameters which are involved in transforming the data from one to the next level The different levels contain data in the following status e RAW Values are given in channels and are stored as integer e TCAL Times are converted to ns and for energies the pedestal is subtracted Data is from now on stored as floating point e SYNC A time offset synchronizes all channels energies are synchronized using a gain factor e DHIT Reconstructed hits are given in a detector specific coordinate system e g indices of a matrix defined by a position sensitive PMT e HIT The position of a hit is given in a detector internal coordinate system with its origin normally at the center of the detector Position time and energy are given in physical units 40 e g charge and mass have to be applied to the data The two following reactions are of particular interest here 1 F T SO S740 n 2 TF T 30 440 2n Where T can be any of the target nuclei given in table 3 2 First the appropriate incoming isotope has to be selected this is done via a 2d elliptical cut on A Z vs Z as shown in figure 3 4 Knowi
131. llent time resolution of 10 ps sigma A modified version called TacQuila be cause of its additional 17 channel is from now on used in the LAND R B experimental setup The TacQuila electronic is composed out of four different electronic boards These boards replace a whole set of other electronics like e g preamplifier splitter multiplexer sum units TDC and QDC TRIPLEX piggyback QDC lt LAND FEE Pa TacQuila Figure 2 9 Upper part Photographs of the four individual components of one TacQuila readout system are shown namely TRIPLEX FEE TacQuila and QDC The center indicates schematically how the four components have to be connected Lower Part One full assembled TacQuila readout system is displayed The four TacQuila components are shown in figure 2 9 At the moment these are e LANDFEEF2 TacQuila3 e QDC2 e TRIPLEX2 21 CHAPTER 2 THE NEULAND TIME OF FLIGHT NEUTRON SPECTROMETER The numbers appended to the name of each board refer to the version of the corre sponding electronic board The modularity of the TacQuila system offers the usage of this readout electronics for various different signal types Exemplary MRPCs and PMTs can both be readout using the TacQuila system changing only the FEE A much more detailed description of the TacQuila system can be found in appendix B Here only the main features will be explained Each TacQuila channel delivers a time and a c
132. lls and a fast ramping mode are used for low intensity beam settings In the framework of this thesis only the most neutron rich setting A Z 3 is of interest In this setting a fast ramping mode and a short spill have been used due to the low intensities The reaction products of the nuclear fragmentation of the incoming Ar beam impinging on the Be target make up the so called cocktail or secondary beam A large variety of elements with masses smaller than the one of the primary beam is produced Out of those the beam composition is then selected by means of the Bp AE Bp method which is applied in the FRS see figure 3 2 Bp selection means that the ions are separated according to their mass to charge ratio This can be shown using UNIversal Linear ACcelerator Schwer Ionen Synchrotron 27 CHAPTER 3 EXPERIMENTAL METHOD AND SETUP S1S100 300 Plasma Physics Atomic Physics Figure 3 1 Shown is a schematic layout of the GSI accelerator complex as it exists today to the left its beamline is depicted in blue On the right also the planned FAIR facility red beam line is shown the following argument The Lorentz force equals the centripetal force which keeps the particle of charge number Z and mass number A on a circular orbit with radius p leading to the following condition A Bp const Z By 3 8 Formula 3 8 is only valid for high energy beams for which the ions a
133. n Links Unten german for right up left down Resistive Plate Chamber Steuerungs und Auslese Modul engl Control and readout module SEcondary Electron TRAnsmission Monitor Saha Institute of Nuclear Physics Schwer Ionen Synchrotron Superconducting Intense Source for Secondary Ions Serial Number Single Particle Energy Systeme de Production d Ions Radioactifs Acc l r s en Ligne in engl System for Producing Online Accelerated Radioactive Ions Time to Amplitude Converter Time to Digital Converter TDC TAC ADC Thin Film Breakdown Counters Time of Flight Wall Time of Flight VME TRigger Distribution Module VME TRIgger Synchronizing Module 161 APPENDIX D ACRONYMS TRLO TRigger LOgic TSL The Svedberg Laboratory TTL Transistor Transistor Logic UCESB Unpack and Check Every Single Bit UNILAC UNlIversal Linear ACcelerator VME or VMEbus Versa Module Europe VULOM VME Universal LOgic Module VV Vor Verstarker engl Preamplifier 162 Bibliography Aum05a Aum05b Aum07 BABt07 BC84 BEE 92 BGIt03 T Aumann Nuclear structure at the dripline Nuclear Physics A 752 289 298 April 2005 T Aumann Reactions with fast radioactive beams of neutron rich nuclei European Physical Journal A 26 441 478 December 2005 T Aumann Prospects of nuclear structure at the future GSI accelerators Progress in Particle and Nuclear Physics 59 3 21 July 2007 T Baumann A M Amt
134. n use LAND and Veto see Sec B 3 Figure B 11 Shown is a prototype of the I C to USB converter 106 B 3 TRIPLEX Figure B 12 Shown is the used USB over IP server 10100000 fe S J 10100000 J 1010010C 1010010C 1530 1530 530 1530 1530 soo 1500 500 1500 500 1500 1500 500 m Me Figure B 13 Shown is the TRIPLEX control window of one card For explanations see text 107 APPENDIX B NEW LAND ELECTRONICS TACQUILA Besides the TRIPLEX interface two more modules are needed to control the TRIPLEX over the network the I C to USB converter and a USB over IP server see figure B 11 and figure B 12 The USB over IP server can be remotely reset using a network con trollable power plug This can be accessed via a web interface www landpwr002 de usr name and password are the standard land ones The TRIPLEX cards are at the moment controlled via LabView In the next paragraph some details on the LabView settings as well as detailed instructions on how to read and write those will be given LabView is e g installed on KRAPC003 a windows machine currently located in Cave C The user can either use this machine locally or connects to it using a Remote Desktop Connection To be able to communicate with the TRIPLEX one has first to connect the USB over IP server to that computer this is done using a program called MFP server The GUI of this program is shown in figure B 10
135. n using the NN 3N interaction predicts the 2 O ground state to be bound a perfect agreement is found using the calculation which includes the residual interactions NN 3Nres This interaction predicts correctly that the 6O ground state is unbound The ground states of both oxygen isotopes are perfectly reproduced by theory using the NN 3Nres interac tion It should be pointed out that going from the NN 3N to the NN 3Nres interaction both ground state predictions get closer to the experimental observed ones The theoret ical predictions for the first excited state of 2 O and the experimentally observed excited state do not match This discrepancy however is most likely found between theory and experiment because the states which are compared here are not of the same nature The lifetimes obtained here are discussed in context with values given in GMSZ11 Graphs showing the relation between the width and the resonance position for different angular momenta l have been adapted from this publication and are shown in figure 7 3 Unfortunately the text of this publication uses the term lifetime 7 while the axis label indicates that the half life 7 3 However the values given within this thesis are all lifetimes 7 For the 7 O ground state the measurement presented here is indicated in figure 7 3 a The blue box represents the measured values including their error It can clearly be seen that the state has angular momentum l 2 as expected
136. n with all three targets has been combined for this analysis 3 3 2 Detection of the Reaction Products Before coming to the question how the reaction products are detected a few general things concerning the reaction shall be discussed The high beam energy is advantageous for several reasons The kinematical forward focussing of the reaction products makes full acceptance measurements possible having detectors which cover in the lab system a considerably smaller solid angle than 47 Furthermore the high beam energy allows the usage of relatively thick targets which makes the investigation of exotic nuclei possible even if such isotopes can only be produced with very low rates Aum05b Aum05a The outgoing fragments are detected in two DSSSDs directly behind the target their charge is determined via a AFE measurement and the outgoing angle 0 is measured as well The target is surrounded by four additional DSSSDs the so called box detectors and a 4r Nal sphere named CB The CB is segmented into 162 crystals which are Crystal Ball 33 CHAPTER 3 EXPERIMENTAL METHOD AND SETUP all equipped with a high gain readout for y rays and in the forward hemisphere also an additional low gain readout for protons This detector is in the analysis presented here used to tag the proton knockout via the CB sum trigger The beam line was evacuated up to the target chamber After DSSSD 4 the reaction products leave the vacuum and enter AL
137. nal of all combined TacQuila cards is available via this TRIPLEX tree The TRIPLEX tree offers also the functionality of a multiplexer meaning that each individual signal of each FEE can e g be directed to a scope Experimental Physics and Industrial Control System tMultiplicity 23 CHAPTER 2 THE NEULAND TIME OF FLIGHT NEUTRON SPECTROMETER 24 Chapter 3 Experimental Method and Setup In the thesis presented here neutron unbound states of neutron rich oxygen isotopes are investigated The experimental technique which is used to study neutron unbound states is the invariant mass method BST12 In this method the four momenta of all decay products in the laboratory frame are measured to reconstruct the unbound states Since the invariant mass Mjny is as suggested by the name invariant to Lorentz transformations it gives also information on the mass of the unbound state before the decay meaning the mass of the unbound system in its rest frame Before the invariant mass can be reconstructed first the unbound state has to be populated This is due to the inverse kinematics done at beam velocity and the decay products are therefore emitted at approximately beam velocity These decay products have to be detected in coincidence which requires at least two detection systems one for the charged particles heavy fragments and one for neutrons In order to detect the neutrons and heavy fragments which stem from in flight breakup
138. nce position and width The individual steps from the measured data to the final interpretation will be described in the following sections 5 1 Detector Response LEG Simulation Analyzing data taken with any detector rarely gives directly the original physics quan tity The response of a detector to a monoenergetic source will e g distribute the energy over many channels according to the gain and energy resolution of the detector and readout electronics This resolution function is usually relatively complicated and depends on the input distribution as well as on the detector settings threshold HV etc and the analysis procedures in case of LAND e g on the shower parameters see section 4 2 Since the final goal of the analysis is to reconstruct the real energy it is con venient to table the response of the detector while the real energy serves as a parameter This lookup table is simply a matrix and gives the data set the name detector response matrix Often the detector response matrix is normalized to the number of simulated events A detector response matrix is commonly determined by using known values as input for a simulation and generating the output including all individual steps e g neutron conversion to charged particles scintillation process light transport through the paddle etc Using a complex experimental setup like the LAND R B setup which is composed of several detectors one would have to combine the response of
139. ncident particle The working principle of a QDC requires a gate signal to define the integrating window This is why a delay line on the signal path to the QDC input is necessary The input pulses have to be matched with the gate which arrives with some latency in respect to the analog pulses that produced it The QDC gate and the TDC start or stop signal are issued by the so called trigger The term trigger refers here to a signal issued by the DAQ which delivers the gates starts and stops to the corresponding digitizers but also causes all digitizers to deliver the data to the data stream During the analysis of the data it is very important to know how the data has been recorded in particular the trigger conditions have to be known Having a signal in only one detector channel refers most likely to noise To avoid collecting noise in the data stream a trigger is constructed requiring coincidences between several signals One step is for example to combine signals from one detector building a detector trigger But also more complex triggers can be build combining the information from several detectors The least restrictive coincidence which is required in the LAND DAQ is the so called minimum bias see table 3 3 which ensures that the ion impinges on the reaction target To trigger on reactions additional conditions to the minimum bias are required e g a LAND trigger The reaction rate of incoming ions in the reaction target is
140. ne neutron has been generated The upper panel represents the response matrix for which this one neutron has been reconstructed correctly In the lower panel the response matrix is shown for which a second fake neu tron has been reconstructed by the analysis procedure The dotted black line represents the diagonal Assuming that the charge Z and mass A identification is to 100 correct 60 5 1 DETECTOR RESPONSE LEG SIMULATION gt E j Z a O gt O n 10 sesula anlegt I e lt 2 E eee ee 10 u 10 a 10 E 10 a 104 8 Ee generated MeV Figure 5 3 Shown are the response matrices of LAND for the 2n 1n a and 2n 2n b channels The dotted black line represents the diagonal 0 08 0 06 0 04 0 02 gt T f k m I i D i 2 i S l 5 l re E es E Feed ee i i 0 1 2 3 0 1 2 3 4 5 4 5 generated E i in MeV generated E N in MeV Figure 5 4 Upper pad Eleven projections of the response matrix shown in the upper panel of figure 5 2 are depicted in different colors The corresponding generated energy to the eleven shown spectra is ix500 keV where 2 is 1 to 11 The individual distributions are fitted using a gaussian The peak position as well as the sigma is determined Lower left pad The generated vs tracked energy is shown The correlation follows almost perfect the diagonal black dotted line only for small energi
141. neutrons can not be directly determined but this value is smaller than the resolution determined from the y peak since the light output is larger for neutrons 49 CHAPTER 4 IDENTIFYING THE REACTION CHANNEL The magnetic field of ALADIN guarantees that only uncharged particles reach LAND photons 7 can be distinguished from the neutrons immediately using the ToF see fig ure 4 10 Since there is a small probability that a y produces secondary particles the shower algorithm uses a small cone behind the first hit as well as a short time window to look for further hits which could be assigned as secondaries Ebe98 When analyzing a In channel the further tasks are rather trivial The algorithm sorts all hits according to their hit time The first one is used as primary hit and all others are assigned as secondary hits But since for the 6O channel the detection of two neutrons is necessary the procedure has to be more elaborated To find all primary hits which characterize a neutron the following procedure is applied event by event 1 2 3 4 Choose the first hit in time as primary hit Check all remaining hits if they can be secondaries of the first hit Continue at 1 until all hits are either declared secondary or primary hits If several neutrons have been found check the distance between each first hit and all secondary hits and assign the secondary hits again to the closest primary hit This step is ne
142. ng at anytime Thanks a lot to the other PhD students i spend most of the time with at GSI especially Dr Valerii Panin Marcel Heine Matthias Holl Dr Felix Wamers Philipp Schrock Vasiliy Volokov Dr Christoph Langer thanks for allways being supportive during work and willing to help Thanks go also to Dr Ralf Plag and Dr Hakan Johansson for writing and main taining the landO2 analysis framework which is one of the most important tools for the analysis It was a great pleasure to work with Dr Heiko Scheit Thanks for all the interesting input during the analysis especially your knowledge about all the statistical methods was very valuable I would also like to acknowledge the interest in the physics covered by this thesis and the introduction to R Matrix theory and the Breit Wigner line shape of Prof Dr Leonid Chulkov Great thanks go to Prof Dr Achim Schwenk who is in charge for all the theoretical calculations presented in this thesis and who was always interested in discussions about the neutron rich oxygen isotopes in general Thanks go also to Dr Dmytro Kresan for doing the R B ROOT simulation concerning LAND I also wish to express my gratitude to Prof Dr Rene Reifarth for the discussion about physics being always interested in the work and giving advises for career planning Special thanks again to Philipp Schrock who is the only one who was brave enough to read and correct the full TacQuila manual To Valerii i want to thank
143. ng the charge and the mass of the incoming ion the next step is to identify the fragment This procedure is split in two parts First the charge number Z is identified and second the mass has to de determined The charge number can be identified via AE measurements As shown in figure 4 2 two AE measurements are applied one directly behind the target and a second one at the end of the fragment branch counts AE DSSSD 3 a u AE TFW a u Figure 4 2 Energy loss in the first DSSSD behind the target DSSSD 3 vs energy loss at the end of the fragment branch in the TFW The incoming cut on F has been applied already The most intense peak is the unreacted F On the diagonal one can see that isotopes of all lower charges are produced in the target The horizontal band contains products of breakup reactions behind the target e g in air The small inlay in the right lower corner depicts the cut which is applied to select oxygen Z 8 isotopes Knowing the charge the mass is fixed as soon as one knows the mass to charge ratio which on the other hand can be determined via the trajectory of the fragments through 41 CHAPTER 4 IDENTIFYING THE REACTION CHANNEL the magnetic field of ALADIN The procedure of determining the mass of the reaction products will be described in the following chapter 4 1 Fragment Mass Identification via Tracking Two standard ways for fragment mass Ar identification via trackin
144. nics In addition to the TacQuila systems themselves some other electronic modules are needed to complete the readout electronics e A LVTTL to PECL converter SIS Clk Distribution which distributes the 17 channel to the TacQuila boards this module is shown in figure B 41 Three such devices are currently used to distribute the signal to all 30 TacQuilas This three modules are connected from only one source using a LEVCON If one wants to monitor the output of the LVTTL to PECL converter on a scope one has to be aware of how to measure a PECL signal A nice FAQ concerning this topic can e g be found here http www pulseresearchlab com faqs ecl_ques ecl_q13 q16 htm and a printout of this webpage can be found here u ccaesar tacquila FAQ ECL_PECL pdf e The clock Distribution At the moment an external 40 MHz clock is used as a reference for the time measurement of the TacQuilas The module is shown in figure B 42 This will in the future be replaced by BuTiS The clock needs a voltage of 5 V Up to ten TacQuila boards can be connected to one clock module To have the same reference clock for all 30 TacQuila systems at LAND the clock shown in figure B 42 is used as a master clock and three further modules are used as daughter modules Each of those is placed behind one TacQuila crate and used to distribute the clock signal to ten TacQuilas The master clock is also connected to the LEVCON there the signal is converted to TTL and t
145. nput signals and 16 different generated triggers as output Requirements for coincidences x between input signals lead to different triggers The left column shows the trigger bit which is set if this trigger is issued The second column holds the name of the trigger The middle part separated with the double vertical lines shows how the inputs are connected to build a certain trigger a blank field means that this input is not considered The last column holds the used downscale factors for the different triggers 37 CHAPTER 3 EXPERIMENTAL METHOD AND SETUP 38 Chapter 4 Identifying the Reaction Channel Signals delivered from the detectors are as described in the previous chapter stored as event wise data words into so called Imd files MBS data format developed at GSI While this file format is very handy to write the data to during the experiment it is not suitable for the analysis later on The process going from a Imd file to e g a ROOT Cer file is called unpacking This process is done using the land02 software package written by H Johansson Without calibrations the data delivered by this step is present in RAW level see figure 4 1 this means a quantity has commonly a value between 0 and 4095 since the data is stored as 12 bit words this data format is referred to as having a value in channels To enable a subsequent physics analysis those values have to be converted into physical quantities like for exam
146. nsactions on Nuclear Science 52 745 747 June 2005 167 BIBLIOGRAPHY Kie81 KM91 KNP 09 Lan12 LB09 LDK 12 Lei97 Leil 1 LQB 85 LT58 J Kieffer A Modular Stand Alone Monitor and Control System Samac IEEE Transactions on Nuclear Science vol 28 issue 1 pp 646 650 28 646 650 1981 J G Keller and E F Moore Shower Recognition and Particle Identication in LAND GSI Scientic Report 1991 p39 Darmstadt 1992 1991 R Kanungo C Nociforo A Prochazka T Aumann D Boutin D Cortina Gil B Davids M Diakaki F Farinon H Geissel R Gernh user J Gerl R Janik B Jonson B Kindler R Kn bel R Kr cken M Lantz H Lenske Y Litvinov B Lommel K Mahata P Maierbeck A Musumarra T Nilsson T Otsuka C Perro C Scheidenberger B Sitar P Strmen B Sun I Szarka I Tanihata Y Utsuno H Weick and M Winkler One Neutron Removal Measurement Reveals O as a New Doubly Magic Nucleus Physical Review Letters 102 15 152501 April 2009 C Langer Coulomb Dissociation of Cl and Ar constraining the rp pro cess PhD thesis Johann Wolfgang Goethe Universit t Frankfurt am Main Germany 2012 T Le Bleis Experimental Study of Collective Electric Dipole Mode in Neutron Rich Nickel Nuclei PhD thesis University of Strasbourg France 2009 E Lunderberg P A DeYoung Z Kohley H Attanayake T Baumann D Bazin G Christian D Divaratne
147. nsformators pictures results Not analyzed yet 156 prototype name GSI 8 tested during TSL Uppsala 03 11 09 13 11 09 HZDR 25 01 10 28 01 10 number of strips structured anode 8 anode strip width 25 mm strip length 40 cm active area 800 cm number of gaps 2 4 gap size 0 3 mm glass plate thickness 1 0 mm HV Lycron Spray anode iron steel plate thickness 4 mm anode strip distance 1 0 mm GFK material glued outer electrode thickness 2 mm comments all 8 strips the same triangular strip ends rectangular coupled transformators pictures results For the analysis on this prototype see section 2 1 and ref CAB 12 APPENDIX C DATA SHEETS OF MRPC PROTOTYPES 158 Appendix D Acronyms ADC ALADIN ATIMA a u BANL BAPL BEPL BuTiS CALIFA CB clk CPLD DAC DAQ DSSSD DTF ECL EE EFT ELBE ENV EPICS ESPE FAIR FaBu FEE Analog to Digital Converter A Large Acceptance DIpole magNet ATomic Interaction with MAtter arbitrary units BAker s Nominal Likelihood BAker s Poisson Likelihood BEvington s Poisson Likelihood Bunch phase Timing System CALorimeter for the In Flight detection of y rays and light charged pArticles Crystal Ball clock Complex Programmable Logic Device Digital to Analogue Converter Data AcQuisition Double Sided Silicon
148. of each box are the addresses of the two registers 116 B 3 TRIPLEX TRIPLEX connector names Figure B 22 Shown is which parts of the address space for the TRIPLEX tree is used for the LAND electronics The tree is split such that one has only one interconnection between the three crates How this is arranged into the crates can be seen in figure B 23 117 APPENDIX B NEW LAND ELECTRONICS TACQUILA LAND TRIPLEX Interface ee ee Cltl Clt2 Cl Clt4 HEX 8 3 1 31 5 i SEIS Se eon ee Seeger Ngache Q N gt a P N N Q N amp Q N gt R C2t5 C26 C2t7 C2t8 C2t9 C2t10 4 219141219 crate 3 Veto TRIPLEX Interface I l I I I l I i 1 J 1 1 1 I 1 i C3t5 C3t6 C3t7 C3t8 C3t9 C3t10 61813141315 C3tl HEX 4 61 7 D Figure B 23 Shown is the cabling of the TRIPLEX cards which build the LAND TRIPLEX tree and the Veto TRIPLEX tree The following things are indicated 1 place in the crate e g C1t1 means Crate 1 TacQuila 1 2 needed HEX Address at this position 3 which connector one has to use for the cabling J5 or J7 118 B 3 TRIPLEX B 3 2 TRIPLEX TC Address Tr
149. of the neutron unbound state simultaneously a magnetic dipole field is used to deflect the charged particles from the path of the neutrons Future developments which will allow spectroscopy of neutron rich nuclei beyond the drip line with very extreme A Z ratios have been discussed in chapter 2 In this chapter the present experimental apparatus will be described The results presented in this work are obtained from data taken during the s393 experimental campaign This experiment was conducted successfully by the R B colla boration in August 2010 aiming at the investigation of light neutron rich nuclei with the aid of kinematically complete measurements of reactions at relativistic energies with the LAND R B setup at Cave C Using six different settings of the FRS a very wide A Z range was covered For the oxygen isotopic chain for example all existing bound isotopes from the proton drip line nucleus O to the neutron drip line nucleus 740 1 9 lt A Z lt 3 have been created RCa In this chapter first the invariant mass formula will briefly be reviewed In the 25 CHAPTER 3 EXPERIMENTAL METHOD AND SETUP subsequent sections the experimental apparatus and technique will be discussed in detail Starting from a description of the GSI facility and the FRS mainly the configuration of the detectors in Cave C namely the LAND R B setup will be detailed 3 1 Invariant Mass The invariant mass of a system of i masses wit
150. one connects detector signals to the LAND FEE one has to be careful since the numbering of the used eight fold connectors and the channel numbering on the board is not mapped in a one to one way see figure B 5 99 APPENDIX B NEW LAND ELECTRONICS TACQUILA rusis TLLAT N proso XTIdIAL 9 Joyeredurog LIEN 0 eusis TO ad IAU Pail fl pue eporp 2q 91 AL eusts sod aanoajoid ayz duny a ha Andino enuarayip Uy Ep ei ILPcXVWN oydwy 134mg T096XVN Joyereduiod 7 XFldIaL 1 l 99S ITVD rayydury I teuis Sou TOPS LV GH aporq 0 qa uo syusuoduiod yeusis ndur Figure B 3 It is schematically shown how a signal is processed on the LAND FEE for 100 details see text B 2 LAND FEE A o Amplifier fl ive Tveite ef Amplifier iva Inverter C Amplifier and Ly Inverter Figure B 4 Three different input pulses and the response of the amplifier to these are schematically shown This schematic visualizes that the amplifier goes into saturation if the input is too large and how it affects the shape of the pulse LAND FEE top view Figure B 5 Shown is a schematic view of the LANDFEE The connectors are indicated It is illustrated how the mapping of the detector signals to the electronic channels has to be done If one uses for
151. onico A di Biagio and A Lucci Progress in resistive plate counters Nuclear Instruments and Methods in Physics Research A 263 20 25 January 1988 CSGT98 J Cub G Stengel A Griinschlo8 K Boretzky T Aumann W Dostal B Eberlein Th W Elze H Emling G Ickert J Holeczek R Holz mann J V Kratz R Kulessa Y Leifels H Simon K Stelzer J Stroth A Surowiec and E Wajda A large area scintillating fibre detector for rel ativistic heavy ions Nuclear Instruments and Methods in Physics Research Section A Accelerators Spectrometers Detectors and Associated Equipment 402 1 67 74 1998 Ct05 J C Cornell and the SPIRAL Collaboration Radioactive beam facili ties in Europe current status and future development eprint arXiv nucl ex 0501030 January 2005 DAB 00 S Dangtip A Ata B Bergenwall J Blomgren K Elmgren C Johansson J Klug N Olsson G A Carlsson J S derberg O Jonsson L Nilsson P U Renberg P Nadel Turonski C L Brun F R Lecolley J F Lecolley C Varignon P Eudes F Haddad M Kerveno T Kirchner and C Lebrun A facility for measurements of nuclear cross sections for fast neutron cancer therapy Nuclear Instruments and Methods in Physics Research A 452 484 504 October 2000 DB10 P Descouvemont and D Baye The R matrix theory Reports on Progress in Physics 73 3 036301 March 2010 165 BIBLIOGRAPHY Ebe98 EK03 Ele09 FS11
152. output amplitude in mV fee Pe a ee le ee ge Tepe ee 200 300 400 500 600 input amplitude in mV Figure B 2 A graphical representation of the values given in table B 4 is shown The values are represented by the blue markers and line The black dashed line represents the diagonal The red vertical line represents the limit between the two working regimes of the amplifier The following versions of the LAND FEE do exist at the moment e FEE LEMO version e FEEI1A SAMTEC version e FEE2 HammerHai which do only differ by the used connector types Since the TacQuila readout was designed to be used with RPC detectors but is now used in combination with a plastic scintillator which is coupled to a PMT the FEE needs some modifications to work optimal for these larger signals While in conjunction with the RPC the TacQuila has to work with very small input signals deliver a very fast timing and the energy measurement is of minor interest for the use together with a scintillator based detector the input signals are much larger the time measurement is not as critical and the energy measurement is also important To cope with these requirements the following modifications are planned for the FEE3 The protective diode will be removed and the amplifier will be replaced by an component which does only an inversion such like e g the general purpose pulse transformer murata 78601 This development is currently ongoing If
153. pectively Here one sees that only three bits in the figures labeled as HARDWARE SELECTABLE can be changed for the address of each part These three bits and therefore the addresses of the registers are directly defined by the HEX switch of the corresponding TRIPLEX card The HEX address of a TRIPLEX board is four bits long e g HEX address A binary 1010 Out of these four bits the first three are used to generate the address of the multiplexer and one register while for the second register the three last bits are used All combination of addresses for the multiplexer and both registers which one can gener HADes SHOwer POwer MOnitor 112 B 3 TRIPLEX Figure B 17 Information Flow for one of the Y crosses which appear within the TRIPLEX tree This is a schematic of the situation how it is realized at the moment Figure B 18 Optimized information Flow for one of the Y crosses which appear within the TRIPLEX tree If this scheme would be used the addresses of the registers would not constrain the TRIPLEX tree This should be applied for future developments ate by changing the HEX switch on the TRIPLEX board are shown in table B 6 Using the 14 sets of available addresses see table B 6 one has to build a I C Tree While doing 113 APPENDIX B NEW LAND ELECTRONICS TACQUILA 1 1 1 O A2 A1 AO RW aa FIXED HARDWARE SELECTABLE Figure B 19 The device addressing for the 8 channel I C bus multipl
154. ple times in ns Furthermore the individual channels of one detector have to be synchronized to each other the detectors have to be synchronized amongst each other and if present the time dependence of these synchronization and conversion parameters has to be corrected for Synchronizing channels refers to the procedure in which times are shifted using an offset in a way that all channels have a common zero and energies are scaled using gain factors such that they have the same range for all channels For all these higher data levels calibration steps are needed and for that purpose the land02 software package offers routines as shown in figure 4 1 To determine the time dependence of calibration parameters a set of Perl scripts was added to this software package by D Rossi The different data levels steps required and programs involved in going from one to the next level are shown in figure 4 1 The calibration synchronization procedures applied for the detectors of the LAND R B setup have been described before see e g R0s09 Joh10 and LB09 they will not be detailed here The calibration parameters used for detectors in front of the target incoming PID were obtained by M Heine The calibration of the DSSSDs was done using a set of scripts written by V Panin Pan12 and modified by M Holl To select a certain reaction channel several conditions on list mode data 39 CHAPTER 4 IDENTIFYING THE REACTION CHANNEL x tea
155. pplicable Those deliver very similar results in case of a high statistics data samples But are not all usable in the case of a low statis tics experiment Different methods will be described here following the ideas described in BR69 and BC84 The used notation is ni number of events in the it bin experimental measurement N N total number of events N X ni i l yi number of events predicted by the model to be in the i bin N No total number of events predicted by the model No D gt yi i 1 For the commonly used x tests two choices are possible 1 errors based on the sample data Neyman s chisquare OG see figure 5 6 and formula 5 7 2 errors based on the parent distribution Pearson s chisquare x5 see figure 5 7 and formula 5 8 where the names are taken from BC84 This leads to the following two formulae for X m Bei eee 5 7 i l S ni y p a 5 8 ll mn 65 CHAPTER 5 ANALYSIS Number of Measurements Figure 5 6 Histogram drawn from a Gaussian parent distribution with mean p 5 0 and standard deviation o 1 corresponding to 100 total measurements The parent distribution y x N P x is illustrated by the large Gaussian curve The smaller dotted curves represent the Poisson distribution of events in each bin based on the sample data Figure is taken from BR69 Number of Measurements 0 0 2 0 4 0 6 0 8 0 10 0 Figure 5 7 The same his
156. rces in chiral EFT can explain the so called oxygen anomaly OSH 10 for the first time Single Particle Energy Effective Single Particle Energy Effective Field Theory N 280 Z 8 N 20 Neutrons v IEIET As 2 i 7 2 N f gt o 472 S Ca Z 20 N 20 aN 1 SQ pfshell Neutrons v 1s1 2 x gt ne te Od5 2 oo 1p3 2 3 i sy E iga E H l N l I II Op3 2 1 i Z Pl 1 Od3 2 i 1 ea 1si 2 7 20 081 2 ea g i 7 0d5 2 E ke a A e A 2 8 10 12 14 16 18 2 Op ya Z i Proton Number Z 0p3 2 051 2 Figure 1 2 Shown are the neutron ESPEs of N 20 isotones for Z 8 to 20 Magic numbers are indicated by numbers in yellow circles The middle part of the figure is taken from OUF 02 the idea of the drawing is adapted from Sch11 giving a microscopic explanation In figure 1 3 a the neutron ESPEs derived from nucleon nucleon NN interactions are shown as function of the neutron number N for oxygen isotopes Z 8 Following the x axis it is indicated by colored dots in which corresponding ESPE the new added neutron has to be filled to create the next more neutron rich oxygen isotope As long as ESPE lt O added neutrons are bound within the nucleus As seen in figure 1 3 a for that type of calculation this is possible even up to 780 To correct for this wrong prediction empirical forces can be used as shown in figure 1 3 b The there sho
157. re fully stripped In the general representation the charge number Z of the ion has to be replaced by the charge q The radius p is fixed by the given trajectory defined by the beam line The constant is given by const _ 31 Tm Bis the strength of the magnetic field 8 is the velocity of the ion and y is the associated Lorentz factor Formula 3 8 therefore means that for a certain magnetic field rigidity only ions with a specific A Z can follow the trajectory determined by the beam line The FRS momentum acceptance Ap p 2 GAB 92 leads to the fact that the beam contains several different nuclides The nominal magnetic rigidity Bp at the exit of the FRS was set to 9 88 Tm for the experiment described here The AE selection is accomplished by inserting a degrader wedge shaped material 28 3 2 RIB PRODUCTION GSI AND FRS from SIS to Cave C fragmentation target optical axis 2 AT yy dipole magnet I a in plastic scintillator Figure 3 2 A schematic layout of the FRS is shown The Bp AE Bp method is applied using dipoles to bend the beam Bp and a degrader to have a position and Z dependent energy loss AE Two 3 mm thick scintillator paddles were operated in the FRS beam line by the RB collaboration at S2 and S8 to do a ToF measurement of the incoming ions Figure is taken from Lan12 as shown in figure 3 2 Since according to the Bethe Bloch formula the energy loss is proportional
158. rmore in BC84 also a multinominal likelihood is given which is derived replacing the Poisson by the multinominal distribution N ni Pan NINN TJ 2 i i 1 ji m 5 15 BEvington s Poisson Likelihood BAker s Poisson Likelihood 67 CHAPTER 5 ANALYSIS this leads to a x which will here get the index BANL N XBANL 22 gt m In ni yi 5 16 i 1 X p formula 5 14 and x2 gpz formula 5 12 are the same despite a constant term and give therefore the same minimum This can well be seen in figure 6 2 comparing the two spectra in the middle row pad C and D Both spectra look exactly the same and deliver the same minimum but the scale of both histograms is different which reflects the constant which is neglected for x3 Equation 5 16 and 5 14 have the problem that they can not deal with empty bins For n 0 the logarithm is not defined this problem is solved by replacing the corresponding summand by zero The five methods x3 Nas ne XBAPL and XBANL described so far are all used in the following procedure see next chapter 1 Fold test input function to response matrix figure 5 2 or figure 5 3 2 Do a y projection of the obtained two dimensional matrix 3 Use this function e g shown in figure 5 5 and the experimental data to calculate a x with the corresponding formula As an alternative 6 method the following procedure can be used to construct a x The measured relati
159. ron and active detectors was investigated The detection principle then relies on neutrons inducing reactions in the iron converter Charged par ticles mainly protons from these reactions are detected subsequently in MRPC detec 12 2 1 MRPC BASED NEUTRON DETECTOR CONCEPT tors The usage of a converter to create charged particles from the neutron interactions is adapted from the LAND design Before detailing the MRPC design and prototype tests for NeuLAND a short review about the principle of an RPC is given The precursor of a RPC is a PPC This detector type consists of two electrodes limiting a gas volume A charged particle which crosses this volume creates electron ion pairs If the field is high enough an avalanche is created leading to a sizable voltage change The fast drifting electrons produce a prompt signal which is very well suited for timing purposes In this simple detector design the problem occurs that on one side a higher field is desired to get faster response but on the other side this leads to a longer recovery time deadtime due to the larger amount of charge released Gon06 To improve the detector design a highly resistive material glass is inserted in the plate chamber PPP71 SC81 This new detector type is called RPC and offers the possibility to work at very high fields while avoiding its breakdown as result of an avalanche The resistive plate is essentially transparent to the induced signal The signa
160. s namely the TacQuila readout The document is a mixture of a description of the individual components and a manual for the usage of the new readout electronics It is indented that the write up is also understandable for beginners that s why a lot of details are given The here described situation is the one of the TacQuila setup in May 2011 If you have corrections suggestions updates or comments to this document please write an email to c caesar gsi de B 1 Software TacQuila MBS The DAQ of the TacQuilas is based on MBS this TacQuila MBS DAQ will be introduced in this section The slow control is currently done using LabView and will be explained in Sec B 3 The implementation into EPICS is still ongoing Two TacQuila DAQ versions are available One version was modified by H Johansson to integrate the TacQuilas into the main LAND DAQ The original sources from the EE are written by N Kurz The different sources can be found in lynx Lynx land usr land TacQuila_ DAQ And the modified version can also be found in the experiment specific folders of the newest experiments e g u land lynx landexp aug2010 tacquila multi_land While starting MBS one should gets messages among others which will be similar to the ones shown here R3 30 read_meb start TACQUILA initialization add to sam 6 gtb O number O tac 1 Experiment Electronic Department 93 APPENDIX B NEW LAND ELECTRONICS TACQUILA add to sam 6 gtb 1 number
161. s mehreren Detektoren gemessen Dieses Detektor System erlaubt eine kinematisch vollst ndige Messung wel che auch das detektieren von Y Strahlen beinhaltet Beide neutronen reiche Sauerstoff isotope zerfallen mittles der Emission von Neutronen Diese wurden im LAND Detektor nachgewiesen Die vorgestellte Auswertung konzentriert sich auf die Bestimmung der Relativenergie zwischen Neutronen und Fragmenten Eine Struktur konnte im Ein Neutronen Kontinuum des O n Systems beobachtet werden Auch die O 2n Relativenergie Verteilung weist eine Struktur auf Die O 1n Relativenergie Verteilung wurde mit Hilfe einer Resonanz beschrieben diese spiegelt den Grundzustand des Systems wieder Zur Beschreibung der O 2n Relativenergie Ver teilung wurden zwei Resonanzen benutzt um sowohl den Grundzustand als auch einen angeregten Zustand zu charakterisieren F r die Grundzust nde von O und 7 O wurden jeweils die Lebensdauern bestimmt Diese ist f r O direkt mittels der Breite des Zu standes gemessen worden Die Breite wurde durch den Fit einer Breit Wigner Resonanz in der Ein Zustand N herung an die Daten bestimmt F r den 6O Grundzustand wur de eine obere Grenze f r die Lebensdauer unter Ber cksichtigung der Flugzeit eines Ions vom Target zum ALADIN Magneten bestimmt Eine Zusammenfassung der gewonnen Resultate wird im Folgenden gegeben 1 250 Grundzustand E Tore keV und a keV 2 260 Grundzustand E Doce keV 3 2 O An
162. s not possible to set voltages here The idea is that one can only change the voltage values at the devices themselves In principle the power supplies can also be controlled using a web interface landlv002 gsi de The user is admin no password is needed The web interface is not working reliable so it should normally not be used However as backup solution to check the voltages it could be useful in case that e g the EPICS server does not work To restart the EPICS server in this case follow the HowTo on the land02 webpage To be able to estimate typical values of the currents which one TacQuila system draws the consumption per system is given in table B 8 No voltage V comment current A 1 7 needed by FEE and external reference clock 0 04 2 5 analogue and digital needed by TacQuila and QDC 1 2 3 5 used by TRIPLEX 0 6 4 5 used by QDC 0 3 5 6 used by FEE pre amp and TRIPLEX 1 8 Table B 8 The first column shows the numbering of the lv modules In the second one the nominal voltages for a TacQuila readout system are given The voltages at the lv modules might differ e g due to losses The third column explains which parts use the different voltages The last column gives the current one full TacQuila system draws for each power supply On top small currents are needed by e g LVTTL to PECL converter The voltages which one TacQuila system needs are given in figure B 38 Out of the
163. s of interest a second test case was investigated using incoming 0O ions For the simulation the charge number was fixed Z 8 while the mass number was changed from A 16 to A 20 resulting in the five bands visible in figure 4 4 The other parameters have been varied according to values which are comparable to deviations found in the data for unreacted beam The position on target was changed by 0 35 cm one o which translates one to one in a horizontal shift on the GFIs The velocity 8 of the incoming ions deviates only very little from the mean value In the O case a gaussian with Bmean 0 7226 and co 0 0005 was found The value of 8 was changed in a range Bmean 20 which results in a small increase in the width of each band The most influential parameter is the angle in the x z plane under which an ion enters the magnetic field The angle was changed in a range 1 for the spectrum shown in figure 4 4 being responsible for most of the visible spread A quite good agreement between simulation and data is found in particular the width for unreacted beam can be explained for this the deviations have been determined from the data For the reaction fragments the position and slope can be reproduced The width of each band is as expected much wider in the experiment 43 CHAPTER 4 IDENTIFYING THE REACTION CHANNEL counts GFI2x GFIlx in cm zu Tree ee nn an a a an ee To E l E E E E 0 15 10 5 0
164. s the complexer the subject the better the teachers should be So the following paragraph is devoted to those who contributed in different ways to my education during the last ten years and helped me with my first steps in the complex world of nuclear physics of course in particular during the last time as a PhD student Either directly by teaching me physics and related things or indirectly by being supportive and encouraging and making me believe that it is always worth to continue learning The work presented in this document would not have been possible without the help guidance and support of many other people so first of all i would like to thank all those who contributed to the success of this work Particular i would like to thank my supervisor Prof Dr Thomas Aumann for giving me the opportunity to join his research group at GSI and TU Darmstadt for being always enthusiastic giving me confidence that nice physics results can be extracted and all the long discussions about the analysis Many thanks also to my second referee s Being first enrolled at University of Mainz the second referee was Prof Dr Frank Maas Thanks a lot for all the nice PhD Committee Meetings and for giving valuable input to this work Changing the University to TU Darmstadt Prof Dr Joachim Enders accepted to be second referee for my thesis thanks a lot for this Thanks for the interest in the thesis the good supervision and all the advises also on bureaucratic i
165. se experiments the following modifications have been done 1 software In the initial version only 10 of the 12 bits were read and the QDC was read at the time the trigger on GTB reached the TacQuilas Nowadays all 12 bit are read 8 clock cycles after the comparator triggered 2 hardware The resistor of the RC component was replaced by a 150 kQ component giving a smaller time constant Figure B 36 shows K Koch s measurements with the new resistor and software the values are shown in table B 7 The modifications of the QDC board improve the charge measurement in two aspects both the dynamic range and the rate capability increase voltage V mean QDC pulsel mean QDC pulse2 3 67 2617 490 1 93 2477 400 1 03 2329 355 0 475 2033 290 0 125 1279 188 0 055 809 114 Table B 7 The QDC characteristics have been investigated for two different pulses Pulse 1 has tfau 15 ns and trise 100 ns Pulse 2 has tfau 2 ns and trise 10 ns The amplitude has been modified to scan the QDC range See figure B 36 for a graphical representation New Time of Flight wall 126 B 5 QDC oo S x S a S wn peak position of QDC Ch N Ww gt S S oO Co a S p injected charge a u Figure B 33 The injected charge in a u vs the mean of the QDC peak is shown The measurements were done using a pul
166. sed to control the pulser The meaning of each of the 8 bits will be described now The bits are numbered from the left to the right 1 2 3 4 5 6 7 8 1 2 Pulse Branch here it is chosen to which branch the pulser is send both J5 and J7 bits 00 none bits 11 only left J5 bits 01 only right J7 bits 10 3 4 5 6 Channel here one can select a specific channel if bit 7 is set to 1 7 Select 0 fire all channels 1 fire single channel 8 Mode 0 thr 5V 1 thr 5V Examples for pulser settings are 00000001 looks in LabView like 1 pulser fires each channel in the tree 00000011 looks in LabView like 11 pulser fires only channel 1 on each card in the tree e ReadPortDirection Reg_i i 1 2 expects a 1 digit binary value Defines if the val ues should be written 0 to or read from 1 the CPLD of the TRIPLEX In principle the setting should here always be 0 Remark to LabView Each box in LabView indicates what type of value it expects d decimal x hex b binary Here one has to be careful LabView cuts the leading zeros and does also not align to the right So a shown 1 means for example in reality 00000001 Complex Programmable Logic Device 109 APPENDIX B NEW LAND ELECTRONICS TACQUILA 800 7 rue een ee en ee En a en 200 200 400 Voltage used as threshold at compartator mV 600 CARER E EEE PEER PEER P PEt 800 L 1
167. sellschaft f r Schwerlonenforschung IRIKEN Projectile Fragment Separator The Institute of Physical and Chemical Research japanese abbreviation National Superconducting Cyclotron Laboratory Superconducting Intense Source for Secondary Ions 88Grand Acc l rateur National d Ions Lourds Flerov Laboratory of Nuclear Reactions Rare Isotope Beam Factory CHAPTER 2 THE NEULAND TIME OF FLIGHT NEUTRON SPECTROMETER e FAIR in Darmstadt Germany e SPIRAL 2 in Caen France e FRIB at MSU3 in East Lansing USA The experiment and developments discussed within this thesis have been conducted at GSI FAIR can be understood as a major upgrade of the nowadays GSI accelerator and experiments Within the transition from GSI to FAIR the LANDS ALADIN ex perimental setup will be replaced by the R B experimental setup whose start version is schematically displayed in figure 2 1 Its main components are as highlighted in the figure the R B Si Tracker the y calorimeter CALIFA the superconducting magnet GLAD and NeuLAND The currently used setup is commonly called LAND R B setup indicating the transition phase from GSI to FAIR This name will also be used in this document The RB collaboration is part of NuSTAR the reader is referred to RCb and Aum07 for the full physics programs of R B and to NC for other physic topics covered by NuSTAR The different experimental equipment which is currently being developed for R B
168. ser the pulse was scaled using amplifiers and attenuators The original pulse injected charge 1 has a charge of 23 nC Hence the measurements are in a range from 2 3 nC to 230 nC For smaller values the threshold could not be adjusted in a way that the signals were still accepted One sees that only a quarter of the 12 bit QDC range is covered For higher charges the measurements become more difficult due to saturation effects RE A ee en a er u es 2 2 sigma of QDC peak Ch 1 6 ja oo LG Du Va U DB u a va m rat 1 4 I j AN oe TER j O Ca ae i jae HERR DEREN i CED IE RR j aes CORE CHE i a ee i Li i 1 2 i A RER NR 2 3 4 5 6 T 8 9 10 injected charge a u Figure B 34 Shown is the sigma of the QDC peak vs the charge in a u A decrease of resolution with higher charges can be seen 127 APPENDIX B NEW LAND ELECTRONICS TACQUILA counts 1000 a 3 o 5 1 3 900 kei Ss oO i 300 Q 3 amp 700 gt 10 po oO A 600 500 1 400 ea L L L j L L L i L L L j L L it I j L j L L j L L L L j L L L 30900 L 400 500 600 700 800 900 1000 Energy FaBu readout Ch Figure B 35 The correlation between the TacQuila and FaBu readout is shown for data taken during the s393 experiment Plotted is the Emean vEi Eo for the y paddle which was mainly hit by the beam measured with the TacQuilas and FaBu respectively The red dotted line represents the
169. so to Clemens Herlitzius for sharing the Cherry Lane apartment a office and a laboratory during the stay at NSCL thanks a lot for this nice time Many of the people mentioned so far became also friends besides the education work relationship I would like to thank all of you for making the time at University of Mainz MSU NSCL GSI and TU Darmstadt also enjoyable Thanks for the fun things we did together it was is always nice and a pleasure to spend time with you Last but not least i would like to thank my family for always being a moral sup port And finally yet importantly I would like to thank my wife Bianca For your understanding support endless patience and encouragement when it was most required throughout the duration of this PhD work Words are inadequate in offering my thanks for tolerating that i invested so much time in this project I simply want to thank you for being a part of my life ii BIBLIOGRAPHY Lebenslauf Name Adresse Geburtsdatum Geburtsort Staatsangeh rigkeit Familienstand Schule 1989 1993 1993 1999 1999 2002 06 2002 Zivildienst 2002 2003 Universit t 2003 2005 04 2005 2005 09 2006 01 2007 09 2007 01 2008 10 2008 12 2008 Seit 01 2009 Christoph Caesar Lortzingstra e 3 64546 M rfelden 05 Dezember 1982 Mainz Deutschland deutsch verheiratet Friedrich von Schiller Grundschule Wiesbaden Gerhart Hauptmann Gymnasium Wiesbaden Carl von Ossietzk
170. ss each Only the two outermost iron layers have a thickness of 2 5 mm adding up again to 5 mm while stacking two paddles LAND measures the ToF of neutrons as well as the position of interaction Using this information the momentum of the neutron is reconstructed As an additional quantity also the deposited energy of neutrons is reconstructed see chapter 4 2 3 4 Data Acquisition Using the detectors described in the last section the interaction of nuclei nucleons with matter is used to create electrical signals In the most common detector type a plastic scintillator coupled to a PMT the visible light created by the scintillation is converted to an electrical pulse Those electrical charge pulses are stored as digital data by the DAQ In most applications the quantities which one wants to derive are the energy deposited by the particle in the detector and the time of arrival The DAQ accomplishes this task via signal processing conversion from analog signals into digital numeric values and in the final step writing the data files Those files can later on be analyzed The DAQ is build from hardware as well as software The LAND DAQ software is based on MBS EK03 The hardware is composed of various different digitizers ADC s QDCs and TDCs having specific characteristics for the different detectors The route of a signal will be discussed briefly in the next paragraph to explain the individual steps one example for such a signal flo
171. ssues Many thanks go also to HGS HIRe for offering a scholarship lecture weeks and giving BIBLIOGRAPHY the PhD time a more structured organization Special thanks to Dr Dominic Rossi You were the one who had to and was willing to answer all the questions when i was a newcomer at GSI when we were sharing a office at that time It s actually hard to remember all the things you did for me within the three and a half years Naming helping through C ROOT 1land02 linux problems giving me a ride to GSI providing the calibration scripts used for LAND and being the LEG expert for all the simulations presented in this thesis is still only a part of the list Thank you so much for all you patience and help Thanks to the NeuLAND working group especially Dr Konstanze Boretzky Dr Michael Heil J rg Hehner Omar Nusair and the FZD team which i spend a lot of time with in my first year during all the prototype tests in Dresden and Uppsala Thanks to Dr Diego Gonzalez Diaz for contributing to my understanding of RPCs also it s still little for sharing a office with me and for being always ready for any discussion I m also gratefull to Dr Haik Simon Dr Karsten Koch and Dr Nikolaus Kurz for all the help connected on my work concerning the TacQuila readout electronics To Haik thanks a lot for having always a solution for any computer and electronics problem for always knowing whom to ask at GSI and being willing to help with anythi
172. ssuming a linear conversion 12 ps Hence the here achieved resolution is o 10 5 ps In general the relation 1 channel 12 ps can be used as a figure of merit for the TacQuilas used to monitor the movement of the right edge with temperature But in the LAND DAQ only the physics trigger is used This was suggested by N Kurz for the following reason the calibration trigger destroys the next physics event The statement in the EE is In general the calibration trigger works but the implementation in the 124 B 5 QDC B 5 QDC The working principle of the current QDC version will be explained in this section the electronic board itself is shown in figure B 32 it has for each channel three stages e The first stage is a pulse integrator it converts the voltage to a current Only positive pulses are accepted For the input on the FEE this means negative signals since the amplifier does also invert the signal The end of this stage is a RC component which R 680 kQ and C 100 pF in the first prototype version has a time constant of 68 us The QDC has a virtual gate This means it is software wise set such that it reads the charge value 8 clock cycles 200 ns after the corresponding comparator fired The discharging is only done via the RC component there is no reset after a readout The time constant defines directly the input frequency at which the measurement suffers from pileup this is at 15 kHz if one assumes that
173. t 661 Supplement 1 0 S145 S148 2012 X Workshop on Resistive Plate Chambers and Related Detectors RPC 2010 Cern ROOT http root cern ch CFA 12a G Christian N Frank S Ash T Baumann D Bazin J Brown P A Deyoung J E Finck A Gade G F Grinyer A Grovom J D Hinnefeld E M Lunderberg B Luther M Mosby S Mosby T Nagi G F Peaslee W F Rogers J K Smith J Snyder A Spyrou M J Strongman M Thoen nessen M Warren D Weisshaar and A Wersal Exploring the Low Z Shore of the Island of Inversion at N 19 Physical Review Letters 108 3 032501 January 2012 164 BIBLIOGRAPHY CFA 12b G Christian N Frank S Ash T Baumann P A Deyoung J E Finck A Gade G F Grinyer B Luther M Mosby S Mosby J K Smith J Sny der A Spyrou M J Strongman M Thoennessen M Warren D Weisshaar and A Wersal Spectroscopy of neutron unbound 2 8F Physical Review C 85 3 034327 March 2012 Chul2 L Chulkov R Matrix Theory private communication 2012 CSC 07 M Ciobanu A Schuttauf E Cordier N Herrmann K D Hildenbrand Y J Kim Y Leifels M Marquardt M Kis P Koczon X Lopez M Petrovici J Weinert and X J Zhang A Front End Electronics Card Comprising a High Gain High Bandwidth Amplifier and a Fast Discriminator for Time of Flight Measurements IEEE Transactions on Nuclear Science 54 1201 1206 August 2007 CSdL88 R Cardarelli R Sant
174. t The TacQuila Data AcQuisition t Ger TeBus engl Device Bus tor VMEbus Versa Module Europe Steuerungs und Auslese Modul engl Control and readout module TMulti Branch System GSI Data AcQuisition software 22 2 2 READOUT ELECTRONICS FOR NEULAND system is no multi hit device If one channel fires this event is kept till either the event is picked up by the trigger or the channel is reset because the reset time is exceeded The reset time can be set by the user to values between 75 ns lt reset time lt 6375 ns in 100 ns steps If a second event occurs in the particular channel which triggered already it will be lost In nuclear physics experiments in general it is usually required to control equipment via software settings Moreover it is crucial to also monitor environmental parameters of the experimental setup this includes for example low and high voltages or magnetic fields Since these environmental parameters change at relatively slow rates compared to the real physics data this control software is commonly referred to as slow con trol Kie81 For this purpose EPICS is used in the LAND R B setup The TRIPLEX board offers the possibility to control the TacQuilas via this general slow control soft ware This gives e g the possibility to set individual thresholds for each TacQuila channel The TRIPLEX cards of the individual TacQuila systems are inter connected in a tree like structure A Mull and OR sig
175. ta acquisition becomes an issue As a solution to this problem the following procedure was used to determine the efficiency The TDC window was set to 200 ns Within this time the multi hit TDC records data dead time free Four beam micropulses take place within 200 ns shown in figure 2 6 The first dominant peak represents events which caused a trigger For the trigger a coincidence between the Macro Pulse ON signal and the MRPC OR signal was used The second third and fourth peaks represent neutrons which arrive later at the detector and which are correlated with later beam micropulses In the following these events will be called detected neutrons while those corresponding to the first peak will be called triggers The ratio of the integral of one of the detected neutrons peaks Np to the integral of the triggers peak N represents the probability for detecting a neutron per beam micropulse Dividing this by the number of neutrons per beam micropulse Np leads to the efficiency e of the counter This is represented by the following formula Nn Nt eS 2 1 N 2 1 N and Nn are obtained from the timing spectra see figure 2 6 The latter one can be determined independently for the three detected neutron peaks reducing the statistical error The number of neutrons per beam micropulse is obtained using A Fh Nin 2 2 nam Np 2 2 with e F neutron flux at detector position in Hz cm e A irradiated
176. the event in the channel and the one in the 17t As a consequence the time measurement is now not limited But it can be restricted The restricting parameter called reset time can be set software wise see also Sec B 1 If the reset time is enabled in tacset txt it is the maximum up to which the clock counter can go If the 17 channel did not fire in the time window set by the reset time the event is discarded The 17 channel is not connected on the FEE but directly on the TacQuila board a signal in PECL standard has to be used here At the moment the 17 channels of all TacQuila boards are feed using the level converter shown in figure B 41 GTB IN connector for 17th channel LAND FEE connector for connectors for piggyback QDC GTB OUT external clock Figure B 25 Shown is the TacQuila board It is indicated where to connect the piggyback QDC the LAND FEE the GTB bus GTB In means the side which is closer to the SAM while Out corresponds to the side which is closer to the terminator of the chain the clock the 17 channel and the lv The TacQuila time measurement is realized in the following way see also figure B 26 Each channel starts its own time measurement The so called TAC value is measured till the next rising edge of the clock At this point the clock counter of this particular Positive Emitter Coupled Logic tthe official GSI name of this device is SIS CLOCK DISTRIBUTION 120
177. the exact names of the used parts to build cable the system will be tabulated B 8 1 Needed Cabling To show what cabling is needed for one TacQuila system a possible design for a next generation front panel is shown in figure B 45 De N ERDE REN If Poti s go not into the Slow Control OO000000 kaanse Eo a a E OO0000000 Figure B 45 Shown is how the front panel of a next generation TacQuila system could look like This displays nicely which connections are needed Serial Number 140 B 8 MISCELLANEOUS The individual connections shown in figure B 45 are e The TacQuilas are readout via GTB A cable has to be connected from the SAM to one TacQuila in our setup the most right one in each crate The other TacQuilas are daisy chained to this one The last module in the chain will be called number 1 in the data The bus has to be terminated at the end of the chain e Each TacQuila has to be connected to the clock see figure B 25 and figure B 47 e The 17 channel of each TacQuila has to be connected see figure B 25 and fig ure B 46 e Each TacQuila board has to be connected to the the lv power distribution This is the only connection which is not shown in figure B 45 since this would stay at the backside of the module e The TRIPLEX interface has to be connected to the I C to USB converter see figure B 16 e The I C to USB converter has to be connected to the USB ov
178. togram as shown in figure 5 6 with dotted curves representing the Poisson distribution of events in each bin based on the parent distribution Figure is taken from BR69 66 5 3 CHI SQUARE x AND LIKELIHOOD METHODS Besides x tests likelihood methods are commonly used to test how good a certain dis tribution describes a test sample Starting from the Poisson probability formula 5 9 and applying the method of maximum likelihood which is equivalent to maximize its natural logarithm one can convert this again to a x see formula 5 12 Were the general relation between a probability distribution and x formula 5 11 is used ni Pp y n il Go er 5 9 i 1 N In P gt ni In yi yi const 5 10 i 1 xX 2 In P 5 11 The derived x will here be called BEPL and can be expressed in the following way N XBEPL 22 5 yi ni In y 5 12 i 1 In BC84 a likelihood ratio is formed and then a x is calculated using again for mula 5 11 were X is used instead of the probability function P The likelihood ratio A is built using the probability distribution e g Poisson see equation 5 9 divided by the distribution which one would get using the true values m if there were no errors 5 13 Replacing m by n and taking the natural logarithm In one gets for x which will here get the index BAPL N XBap 22 X yi ni ni In ni yi 5 14 i 1 Furthe
179. tron Tracking The unbound O and O nuclei decay in flight and the neutrons fly undisturbed by the magnetic field straight into LAND The detection mechanism of LAND consists of a two step process 1 Neutrons are converted into charged particles and y rays in the iron sheets 2 Those secondary particles are then detected in the scintillator sheets A neutron induces therefore a shower of secondary particles in the detector The first analysis step dealing with LAND data is hence to reconstruct those showers and to distinguish primary from secondary hits Meaning e g to make sure if two hits are recorded inside LAND whether they belong to one neutron or two The term hit refers here to the fact that both PMTs of one paddle did fire in coincidence Figure 4 9 Shown is a particle shower in LAND induced by one neutron 1n event The picture was generated by D Kresan using a R B ROOT simulation The tracks are shown as black solid lines The red dots represent a hit meaning that both PMTs of one paddle fired in coincidence The described task of sorting hits into showers and therefore neutrons is done by a routine called shower algorithm in the LAND analysis framework A shower in LAND induced by one neutron 1n event is exemplary shown in figure 4 9 in which 48 4 2 NEUTRON TRACKING four hits belonging to a In event are depicted The first such algorithm used by the LAND collaboration was developed in the early
180. ull e Next version of TRIPLEX should be modified Register addressing and baseline modifications can be done better e Sensing for lv distribution has to be implemented 700 counts 601 Ss At 25ns 501 Ss 40 301 Ss 201 Ss 101 Ss S eee TA CF 460 480 500 520 540 time of one TacQuila Channel ns Figure B 48 Shown is the time of one LAND PMT on SYNC level for a cosmic run taken during s394 AsyEos One would expect one broad peak but two side peaks which are shifted by 25 ns one clock cycle are clearly visible 145 APPENDIX B NEW LAND ELECTRONICS TACQUILA 5 3000 nn EEE ls chad nai EET cee FERBERREEESEUSSETSSTT PETE ET a BEN ae be A F ei N 7 ee l J H les ee 9500 Fiss Be rer eeeerenne aeeennnnnnnnnni DO00 I ES es EPLAN EAEE LI BEATA EEA SATE PELATA NETA E se idbatoiexanetyachedetened emery 1500 i a 1000 er Benneeneeenetennnnnd Beet een benennen C 1 1 1 1 1 1 j 1 1 1 1 1 1 1 L 1 1 1000 1500 2000 2500 3000 TAC Ch Figure B 49 Shown is a TAC vs TAC spectrum for a TRIPLEX pulser measurement The channels are on purpose chosen such that the delay is approximately half a clock cycle The non linearity at the edges are clearly visible This bug is called hook The amount of events in the hook itself is smaller 1 however the
181. ve energy is used event by event A x projection of the response matrix see e g figure 5 3 for a small y region serves as probability distribution The y region is chosen such that it reflects the measured relative energy The obtained probability distribution gives the probability to which the measured relative energy corresponds to a certain real energy Such a probability function is selected for each event The individual probabilities are then multiplied to an overall probability This overall probability can again be converted to a x using formula 5 11 And is called Keo in the context of this thesis The methods Xo Gir ees ene and ae and Xon op Will be referred to as method A to F in the following chapter The various x methods have been investigated to gain confidence that the results do not depend on the analysis procedure This can be underlined by the following citation For finite sample size small N general results are lacking one must carefully study the problem at hand in order to choose and interpret a test statistic BC84 Method A can not handle empty bins correctly since the assigned error is zero Method B assumes that the real distribution is known already in advance which is in most cases not true While methods A and B both assume that the error is described by VN methods C and D use a Poisson distribution to describe the error This is for low statistics the better choice As conclusion one has to say
182. visualizing the cut at low energies into the 2n channel Analyzing the data of this reaction channel differs from the analysis procedure applied to the O in the following sense As discussed in section 5 1 the low energy part of the 2n channel of this kind of reaction is shifted to the In channel making it necessary to analyze both channels simultaneously Looking at the real In channel of the O analysis figure 6 1 one sees that no events are observed within the first 200 keV of the spectrum Comparing this to the fake In channel found in the 6O analysis see figure 6 3 lower panel one sees a substantial amount of events in the first 100 keV of the spectrum which can be therefore assigned to originate from the 2n channel Due 75 CHAPTER 6 RESULTS to the low statistic which makes it impossible to extract the width of the state a flat distribution with a width of 50 keV was used as a test function To fit the In and 2n channels simultaneously the following procedure has been applied 1 Select test function E width is fixed to 50 keV 2 Scale 2n test function such that the integral is the same then experimentally ob served 3 Scale In test function by the same factor 4 x methods are applied to In and 2n simultaneously In total the 2n channel of the 2 O data is described with a low and a high energy part ground and excited state see blue and purple contributions in figure 6 3 upper panel The low energy part
183. w can also be seen in figure B 3 Traditionally the electronic readout systems consists of two chains The timing chain made out of a discriminator a TDC and or a scaler for the timing counting acquisition and an energy chain which holds a delay line and a QDC The discriminator generates if its threshold is exceeded an logical output from the analog input pulse A TDC can be pictured as the combination of a TACT and an ADC TDC TAC ADC A TAC needs as inputs a start and a stop pulse The start pulse initiates the linear ramping of a voltage while the stop pulse stops this process As a consequence the generated amplitude is proportional to the time interval between the start and the stop pulses This generated voltage can be given to an ADC which converts it to a digital number In combination the TAC and the ADC create a digital value which is proportional to the time interval between two pulses meaning they act asa TDC In the energy branch cable delay is commonly used to delay the signal The QDC can be pictured as being built from a capacitor plus an ADC An ADC converts the input voltage to a digital value meaning it is sensitive to the amplitude of an input Analog to Digital Converter tTime to Amplitude Converter 35 CHAPTER 3 EXPERIMENTAL METHOD AND SETUP pulse Adding a capacitor an integrator is build meaning a QDC measures the charge the area below a pulse This quantity represents the energy information of the i
184. with transformator to readout HV last glass plate Europlex 2mm thick conductive on both sides R 64 k results 154 Due to problems with the HV the prototype never worked prototype name GSI 6 tested during TSL Uppsala 03 11 09 13 11 09 HZDR 25 01 10 28 01 10 number of strips structured anode 8 anode strip width 25 mm strip length 40 cm active area 800 cm number of gaps 2 4 gap size 0 3 mm glass plate thickness 1 0 mm HV Lycron Spray anode iron steel plate thickness 4 mm anode strip distance 1 0 mm GFK material glued outer electrode thickness 2 mm comments all 8 strips the same simplest solution strip ends rectangular no transformators pictures results Not analyzed yet 155 APPENDIX C DATA SHEETS OF MRPC PROTOTYPES prototype name GSI 7 tested during TSL Uppsala 03 11 09 13 11 09 HZDR 25 01 10 28 01 10 number of strips structured anode 8 anode strip width 25 mm strip length 40 cm active area 800 cm number of gaps 2 4 gap size 0 3 mm glass plate thickness 1 0 mm HV Lycron Spray anode iron steel plate thickness 4mm anode strip distance 0 6 mm GFK material glued outer electrode thickness 2mm comments all 8 strips the same triangular strip ends rectangular no tra
185. wn energy levels reflect very well two experimental findings 1 The opening between the v0ds 2 and v1sj 2 levels which corresponds to the exper imentally found doubly magic 2O BR05 and references therein 2 The ESPE of the v0d3 2 level is positive which means that 240 is the heaviest bound oxygen isotope However the shown results are not obtained using ab initio calculations they use em pirical forces which are derived from a fit to experimental data measured for neighboring nuclei Results which are obtained from a theory which gives as the first one a micro scopic picture and predicts that O is the last bound Oxygen isotope OSH 10 are CHAPTER 1 INTRODUCTION AF a Viowk forces derived from NN theory b Phenomenological forces se 0 7 e a ee Single Particle Energy MeV 8 14 16 20 4 16 20 Neutron Number N d Schematic picture of two valence neutron interaction induced from 3N force Vigw k NN 3N A fotces ESPE gt 0 Single Particle Energy MeV NN 3N A NN 8 14 16 20 Neutron Number N Figure 1 3 In a b and c neutron ESPEs are shown as a function of the neutron number N for oxygen isotopes Z 8 d shows a schematic picture of two valence neutron interaction induced by 3N forces with a nucleon in the 160 core Details are discussed in the text Figures are adapted from OSH 10 shown in figure 1 3 c Th
186. written to the data to identify the SAM 2 GTB id is 0 or 1 and identifies the 2 GTB cables of the above men tioned SAM 3 TacQuila_id on each GTB one can have up to 10 TacQuilas here one can define the numbering of those 4 reset on 1 off 0 enable or disable the reset time 5 threshold NOT used anymore was threshold setting of comparator on FEE value 0 3F in the pre TRIPLEX phase 6 reset time value 0 3F specifies how long an event is kept If the event is not read out within the reset time it is discarded 94 B 1 SOFTWARE TACQUILA MBS Remark The tacset HEX value 0 3F is converted into a 6 bit binary value This value is then extended by two bits The two added least significant bits are fixed to 11 Hence the real value in decimal goes from 3 to 255 in steps of four Therefore one can use the following formula to get the reset time in ns reset time ns tacset HEX value 4 3 clock cycle As a consequence the reset time can be set within the following window using a 40 MHz clock 75 ns lt reset time lt 6375 ns in 100 ns steps 7 generator on 0 off 1 On the FOPI FEE there is a pulser which can be turned on with this parameter For the LAND FEE this is of no use at all 8 TacQuila in 0 out 1 TacQuila board will not be read out also it is connected to the GTB if the setting is 1 The parameter of most importance here is the reset time During tests this w
187. y from a 3 2 state to the 0 ground state of 740 Al 2 which leads to 5 p P E _ 5 2 E 9438p 4 pi 5 2 18 3p Sy E _ _ 5 3 2 E TEETE 5 3 with V2nE Besen 5 4 Re Where k is the relative momentum and u is the reduced mass R is commonly called channel radius in BM69 e g the better description range of the potential is used This parameter divides the configuration space in the R matrix formalism into an internal and an external region and has to be chosen such that R is larger than the radius of the BM69 contains a typo in formula 3F 38 for A here a factor k R is missing Chul2 62 5 2 BREIT WIGNER LINE SHAPE potential This condition ensures that in the external region r gt R only the asymptotic part of the wave function has to be accounted for Since every choice which fulfills that R is larger than the size of the nuclear potential is suited the channel radius has no direct physical meaning It is only needed for computational purposes As a consequence one can use the independence of the obtained physics results resonance position and width on the channel radius as a validity test DB10 The Breit Wigner line shape as given in the one level approximation can also be used to determine the single particle width of the state using 2 c 21 1 Ts P E which is for l 2 und 2 2 3 Tsp eae 5 6 uR 9 3 p
188. y Oberstufengymnasium Wies baden Abitur Gesamtnote 2 1 MSHD Deutsches Rotes Kreuz Kreisverband Wies baden E V Grundstudium Dipl Physik Johannes Gutenberg Universitat Mainz Vordiplom Note gut Begin des Hauptstudiums Dipl Physik Praktikum am National Superconducting Cyclotron Laboratory NSCL der Michigan State University MSU Michigan USA m ndliche Diplom Pr fung Note sehr gut Aufenthalt am NSCL zur Anfertigung der Diplomar beit Diplom Physik Gesamt Note sehr gut Promotion in der AG Aumann TU Darmstadt GSI iv Eidesstattliche Erkl rung Hiermit erkl re ich dass ich die vorliegende Dissertation selbst ndig verfasst keine an deren als die angegebenen Hilfsmittel verwendet und bisher noch keinen Promotionsver such unternommen habe Christoph Caesar Darmstadt im Juli 2012

Beyond the Neutron Drip-Line: Superheavy Oxygen Isotopes

Contents

Download Pdf Manuals

Related Search

Related Contents