Upload
others
View
1
Download
0
Embed Size (px)
Citation preview
Collaboration for Active Target ProjectShinsuke OTA
Center for Nuclear Study, University of Tokyo
1
Two types of Active Targets are constructedLeft (CAT): 106-7 Hz, A~50-80, ~100-300 MeV/u Right (GEM-MSTPC): 10~5Hz, A~10-30, ~10MeV/uBoth can be irradiated with high intensity beam
(CAT) GEM-MSTPC
(α,p) astrophysical(a,a’) EOS, (d,2p) GT
GEM
Akimoto (Master Thesis), Hamagaki, Gunji
High Energy Group
DAQTRIGGER
Ota, Tokieda, Dozono,Kikuchi, Matsubara,Mitimasa, Uesaka
Spin / SHARAQGroup
TGEMElectronics
Hashimoto, Daid,Yamaguchi, Kubono
AstrophysicsGroup
RIKENKyoto Univ.
Univ. of TsukubaMiyazaki Univ.
Collaboration for active targetIntergroup collaboration
to maximize the effective human resourceto exchange the technique and knowledge
CNS, Univ. of Tokyo
physicsmotivation
Active Target
MSUAT-TPCelectron
multiplier
History2009
Collaboration begins inside CNS
Construction of prototype CAT was done in November
Test experiment in December, 2009 (CAT) and January and February in 2010 (GEM-MSTPC)
Start collaboration for a part of TPC development between CNS and MSU under EFES grant
1st meeting
2010
Two (a,p) exp. in 2009 (GEM-MSTPC)
Operation of Gem with D2(+CO2) gas : how to make a system safety?
Explosion affair !! (CAT) => Fixed in 3 months
Test exp. w/ 250MeV/u 56Fe @HIMAC (CAT)
2nd collaboration meeting (this meeting)
2011...
Today’s Program
CAT
CNS Active Target for (d,2He) (S. Ota)
Results from test experiment with 4He gas
Test Experiment with D2 gas and 250-MeV/u 56Fe beam
Operation of GEM with D2 gas and safety (H. Tokieda)
GEM-MSTPC (T. Hashimoto)
Development of the GEM-MSTPC for (a,p) reactions of astrophysical interest
AT-TPC (D. Suzuki)
....
CNS Active Target for (d,2He)
GT+ transitionGT+(, SDR...)
Supernova Nucleosynthesis
medium mass heavy nuclei (56Ni, 64Ge...)
(K. Langanke (2010) and refs therein)
GT+ transition in neutron-rich exotic nuclei
How much blocked? Quench?
The change of overlap of proton and neutron radial wave functions (ex. along Ni isotope chain) in neutron rich exotic nuclei far from stability line
Effect by the difference of fermi-surface
Potential depth and shape
(n,p) type reaction
normal kinematics
RCNP: (n,p), KVI: (d,2He), RIBF: (t,3He)
inverse kinematics
(7Li,7Be*)
bound states w/ gamma-ray spectroscopy
ΔL is hardly to be extracted
(angular resolution for scattering heavy nuclei is not enough)
(d,2He)
bound and unbound states w/ missing mass spectroscopy
ΔL can be extracted from the angular dist. of recoils
low energy recoils
(d,2He) on neutron rich exotic nuclei in inverse kinematics
w/ active target at RIBF
nesses of 1.5 mg/cm2 and 4.3 mg/cm2, respectively, wereused as targets. Reference measurements were taken with a12C target !98.9%". The two outgoing protons were momen-tum analyzed by the BBS and detected in coincidence withthe ESN detector. The ESN detector is a focal-plane detec-tion system consisting of two vertical drift chambers and aset of four multiwire proportional chambers as an additionaltracking detector. Reference #21$ describes in detail how(d ,2He) experiments with the BBS-ESN setup are being per-formed and analyzed.The excitation energy spectrum, which was measured
with the 1.5 mg/cm2 target covering scattering angles %c .m .&2°, was used to determine the peak positions with highprecision. The fitting of the peaks was done manually sup-ported by a '2 minimizing algorithm of the program FIT #22$.For this particular target an energy resolution (E of 110 keVfull width at half maximum !FWHM" was obtained for allpeaks. Excitation energies up to the neutron separation en-ergy of about 6.4 MeV were considered. The analysis wasrepeated with the excitation energy spectrum from the4.3 mg/cm2 target. Because of the high statistics acquiredwith the thick target, the scattering angle could now be con-fined to %c .m .&1°. The corresponding excitation energyspectrum is shown in Fig. 1. (E was about 125 keV!FWHM".For the (n ,p) and (p ,n) elementary charge-exchange re-
actions at intermediate energies the measured cross section isconnected to the GT strength #7,8$:
d)!q!0 "
d*!! +
,-2" 2 k fki NDJ).
2 B!GT". !1"
Here, J). denotes the volume integral of the spin-dependentisovector central part of the effective nucleon-nucleon inter-
action at q!0 and can be obtained from Refs. #14,23$. Thedistortion factor ND is usually determined by calculating theratio between the distorted-wave and plane-wave cross sec-tions. The proportionality of cross sections at q!0 andB(GT) for composite probes over a wide mass range is, e.g.,shown in Ref. #9$. For the (d ,2He) reaction an additionalfactor must be included, because here the cross section de-pends on the range of integration over the 2He internal en-ergy #18$, which is taken between 0 and 1 MeV.The cross section d)(q!0)/d* is obtained by extrapo-
lating the measured cross section to q!0 using a DWBA!distorted-wave Born approximation" model calculation:
d)!q!0 "
d*!
)calc!q!0 "
)calc!% ,q "
d)exp!% ,q "
d*. !2"
The DWBA calculations for the present analysis were car-ried out with the code ACCBA #24$. The one-body transitiondensities were calculated in the formalism of normal modesassuming collective wave functions. These are regarded assuperpositions of /-particle–,-hole states. Optical modelparameters from d"58Ni at 170 MeV #25$ were used for theentrance channel, and for the exit channel a global opticalpotential for p"51V was extrapolated to 82 MeV #26$. Therange of scattering angles and excitation energies consideredhere corresponds to momentum transfers q up to 0.14 fm#1.For the extrapolation to q!0 in Eq. !2", the calculationsyield a factor that varies between 1.11 at Ex!0 MeV and1.49 at Ex!6.3 MeV.Although charge-exchange reactions near %!0° selec-
tively proceed via GT transitions, there are usually smallcontributions from higher multipoles #see, e.g., the groundstate !g.s." transition in Fig. 1$. To identify transitions with(L01, we employed the following procedure: The peak
FIG. 1. Excitation energy spectrum for 51V(d ,2He)51Ti. The spectrum has been corrected for the acceptance of the spectrometer asdetailed in Ref. #21$. The data were taken at a spectrometer setting of 0° covering a center of mass !c.m." angular range between 0° and 1°.The energy resolution is 125 keV !FWHM". Sp and Sn denote the proton and neutron separation energies, respectively.
RAPID COMMUNICATIONS
C. BAUMER et al. PHYSICAL REVIEW C 68, 031303!R" !2003"
031303-2
C. Bäumel et al. (2003)
(d,2p) reaction as (d,2He)d(~3S1 T=0) => 2p in 1S0 state (T=1)
sensitive to ΔS = 1, ΔT=1 transition
small relative energy in final state of 2p indicates the existence of 2He (2p in 1S0) at the charge exchange reaction
56Ni
d
56Co
2Hep
p small relative energy
D. Ishikawa, Master Thesis (2009)
d(12C,12B)2He@RCNP
Kinematics of (d,2p) reaction
2H(56Ni,56Co)2He
most recoils emitted around 80-90 deg. (sideway)
δθ~14 mrad for dEx =1MeV @Elab = 1 MeV
δEx~0.5 MeV
angular resolution is important
multiple scattering (deg.)He2
lab!78 80 82 84 86 88 90 92 94
(MeV
)H
e2 la
bE
0123456789
10
He @250MeV/u2Co)56Ni,56H(2Kinematics
= 5 MeVxE = 6 MeVxE = 10 MeVxE
CNS Active Target(d,2He) measurement
106-7 Hz medium heavy RI beam
inactive are along beam path to avoid the effect of the spacial charge
D2 (+CO2) as target and detector gas
~ 1atm
GEM as multiplier
quick response
3-dim. position from charge division and drift time
combined w/ external tracking detectors for beam particles
Geometrical Design of CAT
Beam
Figure 2.3: The layout of the field cages, the GEMs and the readout pad boards.
The properties of the gases were evaluated using the simulation program Magboltz, whichsolves the Boltzmann transport equations for electrons in gas mixtures under the influence ofelectric and magnetic fields [5]. The evaluated gases were He (90%)/CO2 (10%), He (95%)/CO2 (5 %),He (99 %)/CO2 (1 %), d2 (90 %)/CO2 (10 %). Furthermore, Ar (70 %)/CO2 (30 %), which iswidely used for gas detectors was simulated for comparison.The evaluated items were drift velocity, transverse and longitudinal diffusion, and Townsend co-efficient. The results are shown in Fig.2.4 and Fig.2.5. Left and right of Fig. 2.4 show electrondrift velocity and that about Townsend coefficient as a function of the electric field, respectively.Left and right of Fig.2.5 show transverse and longitudinal diffusion coefficients as a functionof the electric field, respectively. In He/CO2 gas mixtures, drift velocity has a shoulder at theelectric field E ∼ 1 kV/cm. At E = 1 kV/cm, the drift velocity for these gases are increases asthe abundance of CO2 becomes small. The Townsend coefficients for He/CO2 gas mixtures arelarger than those of d2/CO2 (10 %) and Ar/CO2 (30 %) at the electric field around 10 kV/cm. 1
2.3 Field Cage
2.3.1 Overview of the Field CageFigure 2.6 shows the field cages. The size of the field cage is 125× 230mm2 × 250 mm (drift
direction) and the distance between two field cages is 40 mm as shown in Fig.2.7. The lengthfor the drift direction and the distance between the two field cages are determined by the beamsize at the focal plane of SHARAQ beam line in RIBF in dispersion matching mode, which isabout 10 × 50 (drift direction) mm2 in RMS. The distance between the two field cages can beadjusted for each experiment to suppress the incidence of the beam-halo and the incidence ofδ-lay from the beam.
The requirements to the field cage are as follows;1As will be discussed later (Sec 2.4.3), the operational region of the voltage difference of GEM was
350 ∼ 500 V for 100 µm thickness, which results in the electric field inside the GEM holes is about35 kV/cm ∼ 50 kV/cm.
7
25cm
10cm
NaIArray
NaIArray
field = 1kV/cm
Electrons drift w/ ~1μs/cm, multiplied by GEMs and collected
by pads
Electronics and DAQ software
6 x 6 x 2 pads x 2 (144ch) readout (for now)
preamp
FADCs
Trigger
DAQ (babirl: Baba-san’s talk)
!"
! #"
!!
"
!""
!"
!"#$%&&
!"#$%&&
Preamp. (RPA-210) REPIC❖ RPA-210 (REPIC) (CXA3653Q chip)
❖ 24ch -1.0pC - 1.0pC
❖ 0.8 V/pC
❖ τ=80 ns
❖ GEM-Preamp: 80cm flat cable
Typical signal (He+CO2(5%))!
!"#$
very low noise (~1mV)
Output of preamplifier(RPA-210)
FADCs
FADCresolutionsampling
ratecost
zero suppressio
nthreshold architecture production dead time availability
COPPER II
12bit65MHz (max)
1.3MJPY/32ch
software software1cpu/32ch
KEK readout 144 ch*
SIS3301
14bit105 MHz
(max)
1MJPY / 8ch
hardware
each ch VME SIS no 40 ch
V174012bit
65MHz (max)
1MJPY / 64ch
hardware
every 8ch
VME CAEN no 64 ch
GET cheap? hardware?
each ch?
*most part is property of KEK
Top-α
α
field shaping wires
GEMa, d,
NaIArray
NaIArray
Event and Sampling Trigger
Event trigger
delayed Beam AND NaI
(for high momentum recoils)
GEM (itself not pad)
Sampling trigger (in SIS3301, V1740)
self-trigger is generated under or below threshold
clock synchronized => time-stamp track identification
Gate for COPPER II is open when the previous event was finished
Beam test w/ 4He+CO2(5%)
3.2. SETUP 21
Particle 4He2+Kinematic energy (before scatterer) 30 MeVKinematic energy (after scatterer) 29.8 MeV
B · ρ (after scatterer) 0.788 T mBeam current (before scatterer) 1 ∼ 10 enA
Table 3.1: Summary of the beam condition.
3.2.2 BeamThe condition of the beam which was used at the performance test is summarized at Tab.3.1.An Aluminium foil was used to degrade the energy of the beam. It was put at the end of the beam
line.
3.2.3 TPCThe setup of the TPC are as follows;
• Gas: He (95 %) + CO2 (5 %) was used to keep flowing. The flow rate was 200 sccm.
• Electric field of the drift region: The electric field of the drift region was 700 V/cm.
• Readout pad: The 36 pads were used for the performance test. Figure 3.3 shows the numberingof the readout pad. When the beam passed at the center of the active area of the field cage, atthe third line, the pads from the second line to the fourth line were used. When the beam passedat the end of the active area of the field cage, the first line, the pads from the first line to thethird line were used.
The drift velocity of electrons, the transverse diffusion coefficient and the longitudinal diffusion co-efficient for the condition were about 2 cm/µs, 250 µm and 160 µm, respectively.
Figure 3.3: The numbering of the readout pad. The pads which colored yellow are in the second rowand in the forth line.
16.45 mm
22 CHAPTER 3. PERFORMANCE TEST AND DISCUSSION
3.2.4 Front End ElectronicsA charge sensitive preamplifier modules (RPA-211; REPIC Co., Ltd) were used to convert the
charge signal from the readout pads to the voltage signal. CXA3653Q ADS chip (Sony, Inc.) wasused for the preamplifier module. The gain, the time constant and the dynamic range for the inputsignal of the preamplifier module were 0.8 V/pC, 80 ns and -1.0 pC ∼ +1.0 pC, respectively.
A VME-FADCmodules (SIS3301; SIS GmbH) were used to digitize the signal from the TPC. Thedynamic range and the resolution of the FADC module were -2.5 V∼ +2.5 V and 14 bit, respectively.Although the maximum sampling rate of the FADC module was 105 MHz, 100 MHz external clockwas used as the clock for sampling the signals. The FADC modules were operated in the wrap modeduring the measurement. When the FADC module is operated in the wrap mode, it will acquire dataand write data to a ring buffer until it is stopped by the stop condition. The stop condition was madeby the stop trigger signal. Figure ?? shows the trigger logic for the stop trigger. Since the width ofa pulse was several hundreds nano-seconds, the number of the sampling for one event was set to be128, which corresponds to 1.28 µs.
3.3 ResultFigure 3.4 shows a typical event. Each histogram shows the pulse from each readout pad in the
second, third and the fourth line. A beam passed at the third line of the pads.
Figure 3.4: A typical event. Each histogram shows the pulse from each readout pad in the second,third and the fourth line.Charge from pads are integrated by preamplifier
and waveforms are digitized by flash ADCs
@Univ. of Tsukuba24MeV 4He beam
structure of readout pad
typical resolution700 um (charge-div.)
90 um (drift time)
x (z) ∝ (Q1-Q2)/(Q1+Q2)y ∝ drift time
2.4. READOUT CHAMBER 15
Therefore the positions of the tracks are derived by the charge proportion of the neighboring twopads. The way to derive the position is different by the type of pad pair and by the incident condi-tion (Tab.2.3.
x = Q1−Q2Q1+Q2
·Lx
2+ P0;x x = −
Q1−Q2Q1+Q2
·Lx
2+ P0;x
z = Q1−Q2Q1+Q2
·Lz
2+ P0;z z = −
Q1−Q2Q1+Q2
·Lz
2+ P0;z
x = P0;x x = P0;x
z = −Q1−Q2Q1+Q2
·Lz
2+ P0;z z = Q1−Q2
Q1+Q2·
Lz
2+ P0;z
x = −Q1−Q2Q1+Q2
·Lx
2+ P0;x x = Q1−Q2
Q1+Q2·
Lx
2+ P0;x
z = P0;z z = P0;z
Table 2.3: The position derived by the charge proportion of the neighboring two pad for differenttypes of pad pair and for different incident conditions. Q1 and Q2 are the charges for each pad. Lx
and Ly are the lengths of each side of the pad. P0;x and P0;y are the center positions of the two pads.
In order to decide the pad size, simulations for position resolution were performed.
2.4.3 Simulation for Position ResolutionMonte-Carlo simulations for position resolution were performed with different energy resolutions
and with different pad sizes since the position resolution is affected by the energy resolution and thesize of the readout pad. The simulations were performed for the ideal case where there was no angularstraggling which limits the minimum angular resolution and where the energy loss was constant. Aprocedure of the simulation is as follows;
Step 1. Generate a uniform random number which corresponds to the position where a ionization oc-curs.
Step 2. Drift the electron with transverse diffusion from the generated point to the GEM and calculate
R. Akimoto Master Thesis (2010)
Position resolutionsPerpendicular to drift direction Drift direction
less than 700um from charge division
less than 100 um from time projection
Incident position [mm]-8 -6 -4 -2 0 2 4 6 8
Posi
tion
reso
lutio
n [m
m]
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Center of field cageEdge of field cageCenter of field cageEdge of field cage
Incident position [mm]-8 -6 -4 -2 0 2 4 6 8
Posi
tion
reso
lutio
n [m
m]
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Center of field cageEdge of field cageCenter of field cageEdge of field cage
Figure 4.10: The resolution of the position of fourth row (j = 4) in y-direction as a functionof the incident position in y-direction. The left figure shows the resolution of y1(i, j) and theright figure shows the resolution of y2(i, j). The squares represent the results when the beampassed at the third column and the circles represent the results when the beam passed at the firstcolumn.
the recoil nuclei pass through the field cage with the incident angle of 0, 11 hit points can bemeasured. are injected inside the field cage with incident angle of 0,
4.3.2 Position Resolution in the Drift DirectionThe position in the drift direction (z-direction) was calculated from the arrival time and the
drift velocity of electrons, as shown in Eq.4.5. The position in z-direction at j-th row, whichwas named z(j) was calculated using the following equation.
z(j) = vd · t0(j), (4.5)
where z(j) is the position in the drift direction (z-direction) at j-th row, vd is the drift velocityof electrons, and t0(j) is the arrival time of j-th row. The position resolution in the z-directionwere derived with the same way in y-direction. Although the absolute position in z-direction cannot be calculated since the arrival time of the beam was not measured, the position resolution inz-direction can be evaluated from the relative position in the neighboring rows.Figure 4.11 shows the distribution of the residual when the beam passed at third column andFig.4.12 shows that when the beam passed at the first column. Figure 4.13 shows the result ofthe position resolution in z-direction. The squares represent the results when the beam passed atthe third column (center of the field) and the circles represent the results when the beam passedat the first column (near the edge of the field). The position resolution in z-direction was about80 µm for all incident positions and the similar results were obtained when the beam passed atthe third column and at the first column.
4.4 Gain Variation and Energy ResolutionGain variation and energy resolution were evaluated for various incident regions and pad
rows. The energy deposit of the beam was about 110 keV at a individual row and 680 keV
36
Incident position [mm]-8 -6 -4 -2 0 2 4 6 8
Posi
tion
reso
lutio
n [m
m]
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
0.2
Center of field cageEdge of field cageCenter of field cageEdge of field cage
Figure 4.13: The resolution of the position in the drift direction (z-direction) as a function ofthe incident position in the pad direction (y-direction). The squares represent the results whenthe beam passed at the third column and the circles represent the results when the beam passedat the first column.
summed over all rows. The energy resolution was defined by the ratio of the RMS and the meanof the Gaussian used to fit the distribution of the induced charge.Figure 4.14 shows the distribution of the induced charge for individual row (left) and summedover all rows (right). The red lines are the results of the fitting with Gaussians. Left of Fig.4.15shows the relative gain for each incident region (shown in Tab.4.1) and right shows that for eachpad row. The relative gain was defined by the mean of the Gaussian normalized so that themean of the relative gains will be 1. The red points represent the result when the beam passedat the third column (center of the field) and the blue represents the result when the beam passedat the first column (near the field edge). For the blue data, since some electrons go out of theactive area of the readout pads, relative gains at near side of the field edge decrease. No clearcorrelation can be seen between the red and the blue data. The variation is within 5% for thegain of each incident region when the beam passed at center and within 20% for the gain ofeach pad row.Left of Fig.4.16 shows the energy resolution of each incident region at the third column andright shows that of each pad row when the beam passed at the incident region #2. The resultsshows the energy resolution for the energy deposit for individual row was about 9.2± 0.2% andthat summed over all rows was about 3.9 ± 0.05%. They are almost scaled by the number ofelectron-ion pairs.
4.5 Dependence of Parameters
The gas gain and drift length dependence of the performance of the TPC were studied.
38
2.4. READOUT CHAMBER 15
Therefore the positions of the tracks are derived by the charge proportion of the neighboring twopads. The way to derive the position is different by the type of pad pair and by the incident condi-tion (Tab.2.3.
x = Q1−Q2Q1+Q2
·Lx
2+ P0;x x = −
Q1−Q2Q1+Q2
·Lx
2+ P0;x
z = Q1−Q2Q1+Q2
·Lz
2+ P0;z z = −
Q1−Q2Q1+Q2
·Lz
2+ P0;z
x = P0;x x = P0;x
z = −Q1−Q2Q1+Q2
·Lz
2+ P0;z z = Q1−Q2
Q1+Q2·
Lz
2+ P0;z
x = −Q1−Q2Q1+Q2
·Lx
2+ P0;x x = Q1−Q2
Q1+Q2·
Lx
2+ P0;x
z = P0;z z = P0;z
Table 2.3: The position derived by the charge proportion of the neighboring two pad for differenttypes of pad pair and for different incident conditions. Q1 and Q2 are the charges for each pad. Lx
and Ly are the lengths of each side of the pad. P0;x and P0;y are the center positions of the two pads.
In order to decide the pad size, simulations for position resolution were performed.
2.4.3 Simulation for Position ResolutionMonte-Carlo simulations for position resolution were performed with different energy resolutions
and with different pad sizes since the position resolution is affected by the energy resolution and thesize of the readout pad. The simulations were performed for the ideal case where there was no angularstraggling which limits the minimum angular resolution and where the energy loss was constant. Aprocedure of the simulation is as follows;
Step 1. Generate a uniform random number which corresponds to the position where a ionization oc-curs.
Step 2. Drift the electron with transverse diffusion from the generated point to the GEM and calculate
0mm
-8mm
+8mm
Effect of diffusion
Drift directionPerpendicular to drift direction
dependency on the drift length
almost constant (within 100um) resolution=> diffusion does not largely affect
Most probable pulse height of a pad [ADC ch.]0 20 40 60 80 100 120 140 160
m]
µPo
sitio
n re
solu
tion
[
0
100
200
300
400
500
600
700
800
900
1000
Gain: 1000Gain: 600
Gain: 300
Most probable pulse height of a pad [ADC ch.]0 20 40 60 80 100 120 140 160
m]
µPo
sitio
n re
solu
tion
[
0
50
100
150
200
250
300
Gain: 1000Gain: 600
Gain: 300
Figure 4.18: The left figure shows the position resolution in the pad direction (y-direction) as afunction of the most probable pulse height. The right figure shows the position resolution in thedrift direction (z-direction) as a function of the most probable pulse height.
depend on the drift length. The position resolution in the drift direction (z-direction) was im-proved as the drift length become shorter. Similar results were obtained when the beam passedat the third column and at the first column.
Incident position [mm]0 50 100 150 200 250
Posi
tion
reso
lutio
n [m
m]
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Center of field cageEdge of field cageCenter of field cageEdge of field cage
Incident position [mm]0 50 100 150 200 250
Posi
tion
reso
lutio
n [m
m]
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
0.2
Center of field cageEdge of field cageCenter of field cageEdge of field cage
Figure 4.19: The left figure shows the position resolution in the pad direction (y-direction) andthe right figure shows the position resolution in the drift direction (z-direction) for various driftlengths. The squares represent the results when the beam passed at the third column and thecircles represent the results when the beam passed at the first column.
41
Figure 2.3: The layout of the field cages, the GEMs and the readout pad boards.
The properties of the gases were evaluated using the simulation program Magboltz, whichsolves the Boltzmann transport equations for electrons in gas mixtures under the influence ofelectric and magnetic fields [5]. The evaluated gases were He (90%)/CO2 (10%), He (95%)/CO2 (5 %),He (99 %)/CO2 (1 %), d2 (90 %)/CO2 (10 %). Furthermore, Ar (70 %)/CO2 (30 %), which iswidely used for gas detectors was simulated for comparison.The evaluated items were drift velocity, transverse and longitudinal diffusion, and Townsend co-efficient. The results are shown in Fig.2.4 and Fig.2.5. Left and right of Fig. 2.4 show electrondrift velocity and that about Townsend coefficient as a function of the electric field, respectively.Left and right of Fig.2.5 show transverse and longitudinal diffusion coefficients as a functionof the electric field, respectively. In He/CO2 gas mixtures, drift velocity has a shoulder at theelectric field E ∼ 1 kV/cm. At E = 1 kV/cm, the drift velocity for these gases are increases asthe abundance of CO2 becomes small. The Townsend coefficients for He/CO2 gas mixtures arelarger than those of d2/CO2 (10 %) and Ar/CO2 (30 %) at the electric field around 10 kV/cm. 1
2.3 Field Cage
2.3.1 Overview of the Field CageFigure 2.6 shows the field cages. The size of the field cage is 125× 230mm2 × 250 mm (drift
direction) and the distance between two field cages is 40 mm as shown in Fig.2.7. The lengthfor the drift direction and the distance between the two field cages are determined by the beamsize at the focal plane of SHARAQ beam line in RIBF in dispersion matching mode, which isabout 10 × 50 (drift direction) mm2 in RMS. The distance between the two field cages can beadjusted for each experiment to suppress the incidence of the beam-halo and the incidence ofδ-lay from the beam.
The requirements to the field cage are as follows;1As will be discussed later (Sec 2.4.3), the operational region of the voltage difference of GEM was
350 ∼ 500 V for 100 µm thickness, which results in the electric field inside the GEM holes is about35 kV/cm ∼ 50 kV/cm.
7
125mm
250mm
0mm
Most probable pulse height of a pad [ADC ch.]0 20 40 60 80 100 120 140 160
m]
µPo
sitio
n re
solu
tion
[
0
100
200
300
400
500
600
700
800
900
1000
Gain: 1000Gain: 600
Gain: 300
Most probable pulse height of a pad [ADC ch.]0 20 40 60 80 100 120 140 160
m]
µPo
sitio
n re
solu
tion
[
0
50
100
150
200
250
300
Gain: 1000Gain: 600
Gain: 300
Figure 4.18: The left figure shows the position resolution in the pad direction (y-direction) as afunction of the most probable pulse height. The right figure shows the position resolution in thedrift direction (z-direction) as a function of the most probable pulse height.
depend on the drift length. The position resolution in the drift direction (z-direction) was im-proved as the drift length become shorter. Similar results were obtained when the beam passedat the third column and at the first column.
Incident position [mm]0 50 100 150 200 250
Posi
tion
reso
lutio
n [m
m]
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Center of field cageEdge of field cageCenter of field cageEdge of field cage
Incident position [mm]0 50 100 150 200 250
Posi
tion
reso
lutio
n [m
m]
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
0.2
Center of field cageEdge of field cageCenter of field cageEdge of field cage
Figure 4.19: The left figure shows the position resolution in the pad direction (y-direction) andthe right figure shows the position resolution in the drift direction (z-direction) for various driftlengths. The squares represent the results when the beam passed at the third column and thecircles represent the results when the beam passed at the first column.
41
Dependence on Gas Gain
Perpendicular todrift direction
Drift direction
gas gain was varied by changing high voltage supply to GEMAs expected, larger gain (up to 103), better resolution.
Energy resolution
energy resolution ~ 10% (σ) for one row3.3% (σ) for total (6) rows
Incident postion [mm]-8 -6 -4 -2 0 2 4 6 8
Ener
gy re
solu
tion
[%]
0
1
2
3
4
5
6
7
8
9
10
Pad row0 1 2 3 4 5 6 7
Ener
gy re
solu
tion
[%]
0
2
4
6
8
10
12
14
Figure 4.16: Left shows the energy resolution for various incident region and right shows thatfor various pad row. The beam passed at the third column (center of the field).
that with the gain of 1000.
Most probable pulse height of a pad [ADC ch.]0 20 40 60 80 100 120 140 160
Ener
gy re
solu
tion
[%]
0
1
2
3
4
5
6
7
8
9
10
Gain: 1000Gain: 600
Gain: 300
Figure 4.17: The energy resolution as a function of the most probable pulse height.
4.5.2 Drift LengthThe drift length dependence of the position resolution and energy resolution were studied
with various drift lengths of 25 mm, 75 mm, 125 mm, 175 mm and 225 mm, where the beampassed at the third column and of 25 mm, 125 mm and 225 mm, where the beam passed at thefirst column. The incident angle of the beam was about 0◦ and the voltage supplied to the GEMwas 450 V.Left of Fig.4.19 shows the position resolution in the pad direction (y-direction) and the rightfigure in Fig.4.19 shows the position resolution in the drift direction (z-direction) as a functionof the drift lengths. The squares represent the results, where the beam passed at the third columnand the circles represent the results, where the beam passed at the first column.The results show that the position resolution in the pad direction (y-direction) did not so much
40
Beam test w/ D2+CO2(5%)
Pad ID
90100
110120
130140
150
DriftTime (20ns)
020
4060
80100
120
Puls
e H
eigh
t (a.
u.)
0100200300400500600700
particle
56Fe 250 MeV/u
GEMs were operated online w/ D2 gas for the first time
TPC worked with ~106 Hz beam irradiated between the cages
Particle tracks were observed successfully (even though the particle may not be light particle)
Beam
Timeline for (d,2He) measurement
2009
An active target was designed and constructed.
Test w/ 4He gas was done at Univ. Tsukuba
2010
Property of D2 gas has been studied
Beam test w/ 106-Hz 56Fe beam
2011
Test of (d,2He) measurement w/ 56Fe in inverse kinematics
First physics RUN : 56Ni(d,2He) measurement in inverse kinematics (@HIMAC)
after 2011
(d,2He) measurement for N-rich nuclei at RIBF
Conclusion
(d,2p) reaction has been a good tool for the study of spin flip transition
the measurement of (d,2p) in inverse kinematics is possible with CNS Active Target
CNS Active Target
present: 700um (pad), 100um (drift)
future: would be improved by optimizing the size and shape of readout pad
applied for B(GT+) measurement in the next fiscal year
efficiency
0 0.5 1 1.5 2 2.50
1
2
3
4
5
6
7
8
9
10
hrefEntries 439Mean x 0.7504Mean y 5.051RMS x 0.6412RMS y 2.349
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
hrefEntries 439Mean x 0.7504Mean y 5.051RMS x 0.6412RMS y 2.349
Kin vs Rel
0 0.5 1 1.5 2 2.50
1
2
3
4
5
6
7
8
9
10
hrefEntries 371Mean x 0.6909Mean y 5.473RMS x 0.6163RMS y 2.235
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
hrefEntries 371Mean x 0.6909Mean y 5.473RMS x 0.6163RMS y 2.235
Kin vs Rel
present ~0.5 atm