Active targets for the study of nuclei far from stability

Progress in Particle and Nuclear Physics 84 (2015) 124–165

Contents lists available at ScienceDirect

Progress in Particle and Nuclear Physics journal homepage: www.elsevier.com/locate/ppnp

Review

Active targets for the study of nuclei far from stability S. Beceiro-Novo a,∗ , T. Ahn b , D. Bazin a,c , W. Mittig a,c a

National Superconducting Cyclotron Laboratory, Michigan State University, 640 South Shaw Lane, East Lansing, MI 48824, USA

b

Department of Physics, University of Notre Dame, 225 Nieuwland Science Hall, Notre Dame, IN 46556, USA

c

Department of Physics and Astronomy, Michigan State University, East Lansing, MI 48824, USA

article

info

Keywords: Active targets Gas detectors Inverse kinematics

abstract Weakly bound nuclear systems can be considered to represent a good testing-ground of our understanding of non-perturbative quantum systems. Reactions leading to bound and unbound states in systems with very unbalanced neutron-to-proton ratios are used to understand the properties of these systems. Radioactive beams with energies from below the Coulomb barrier up to several hundreds MeV/nucleon are now available, and with these beams, a broad variety of studies of nuclei near the drip-line can be performed. To compensate for the low intensity of secondary beams as compared to primary beams, thick targets and high efficiency detection is necessary. In this context, a new generation of detectors was developed, called active target detectors: the detector gas is used as target, and the determination of the reaction vertex in three dimensions allows for good resolution even with thick targets. The reaction products can be measured over essentially 4π . The physics explored with these detectors together with the technology developed will be described. © 2015 Elsevier B.V. All rights reserved.

Contents 1. 2.

3.

∗

Introduction.......................................................................................................................................................................................... Examples of active targets ................................................................................................................................................................... 2.1. Ikar ............................................................................................................................................................................................ 2.2. Maya ......................................................................................................................................................................................... 2.3. Active target (ACTAR) .............................................................................................................................................................. 2.4. Multi-Sampling and Tracking Proportional Counter (MSTPC).............................................................................................. 2.5. Center for Nuclear Study Active Target (CAT)........................................................................................................................ 2.6. Micro-PIC based Active target for experiments in Inverse Kinematics at o degrees (MAIKo)............................................ 2.7. Active Target Time Projection Chamber (AT-TPC) ................................................................................................................. 2.8. The TRIUMF Annular Chamber for the Tracking and Identification of Charged particles (TACTIC) ................................... 2.9. The Array for Nuclear Astrophysics and Structure with Exotic Nuclei (ANASEN)............................................................... 2.10. MagIc Numbers Off Stability (MINOS).................................................................................................................................... 2.11. Optical Time Projection Chambers (O-TPC) ........................................................................................................................... Electronics ............................................................................................................................................................................................ 3.1. Amplifiers and digitizers ......................................................................................................................................................... 3.2. Trigger ......................................................................................................................................................................................

Corresponding author. E-mail address: [email protected] (S. Beceiro-Novo).

http://dx.doi.org/10.1016/j.ppnp.2015.06.003 0146-6410/© 2015 Elsevier B.V. All rights reserved.

125 127 127 129 132 136 141 142 143 148 149 150 152 152 152 155

S. Beceiro-Novo et al. / Progress in Particle and Nuclear Physics 84 (2015) 124–165

4.

5. 6.

Data acquisition.................................................................................................................................................................................... 4.1. Front end modules ................................................................................................................................................................... 4.2. Data concentration .................................................................................................................................................................. 4.3. Data storage ............................................................................................................................................................................. Tracking algorithms and simulations ................................................................................................................................................. Conclusion ............................................................................................................................................................................................ References.............................................................................................................................................................................................

125

156 156 157 158 158 160 161

1. Introduction Our understanding of non-perturbative quantum systems can be tested with great detail in weakly bound nuclear systems. Non-perturbative systems are those in which small changes, such as adding one nucleon, or a small change in excitation energy or spin, may lead to a complete rearrangement of the system, and can therefore not be treated as perturbation. These systems can be studied by reactions using rare isotope beams. Great progress in this domain has been reached from an increase of many orders of magnitude in rare isotope beam intensities, and by the development of new highly efficient detectors such as active targets. This combination enables the exploration of completely unknown regions of the nuclear chart. It is now possible to study reactions leading to bound and unbound states in systems with very unbalanced neutron-to-proton ratios. A recent review on the progress in this domain can be found in [1]. In active target detectors the counter gas acts as both a target and detector, enabling the investigation of fusion, isobaric analog states, cluster structure of light nuclei and transfer reactions, performed without significant loss in resolution due to the thickness of the target. Active target detectors also allow to investigate fission barriers and giant resonances with rare isotope beams produced via fast fragmentation. Low-energy resonances of astrophysical interest can be explored with active target detectors. Reaction cross sections have been measured to study big-bang nucleosynthesis, seeds of the r process, rp process, and nucleosynthesis in X-ray bursts. Applications of active target detectors to these domains of physics will be illustrated by experiments performed with existing detectors. Historically one could say that active target detectors go back to the beginning of nuclear physics. Bubble chambers for instance can be considered as the first example of an active target detector: the liquid hydrogen served as target and at the same time to reveal the tracks as visible bubbles. With modern high density electronics, computerized read-out of tracks has become possible, with very complex high-energy detectors such as ATLAS [2,3]. A historic review of the evolution in detector technology can be found in the review of F. Sauli [4]. Many of the technologies that have been developed for high-energy physics have applications in low-energy nuclear physics. However, there are important differences in the requirements between low and high energy. The complexity of nuclear reactions in low to medium energy is much reduced as compared to high energy physics where often several thousands of charged particles emerge from the reaction. This makes analysis at low energy comparatively easier and the number of electronics channels does not need to be as high as in high energy physics, of the order of 10,000 compared to several millions. In addition, the energy loss of charged particles in a low energy detector varies by several orders of magnitude, as a function of the atomic number of the reaction partners and their energy, whereas in the high energy case essentially all particles are at minimum energy loss. This feature implies that the electronics used for the active target must have a high dynamic range in order to accommodate all possible energy losses. Due to the low energy of the reaction products in many experimental situations, the trigger cannot rely on an external ancillary detector, but must be provided by the detector itself. The development of high-energy detector technology has benefited low-energy nuclear physics, but adaptation to its specific features is necessary. The development of active target detectors for the study of nuclei far from stability is driven by the combination of these three main features: 1. Inverse kinematics: the use of secondary beams implies the change from direct kinematics to inverse kinematics. 2. Low recoil energies: in inverse kinematics the recoil particles have very low energies in quasi-elastic reactions such as (d, p) and (α , α ′ ). 3. Thick targets and high solid angle: to compensate the limited intensity of secondary beam, thick targets and high detection efficiency without loss of resolution are needed. In the following we discuss and illustrate these statements with more details. 1. Reactions with stable beams and stable targets mostly use light beams such as p, d, 3 He, 4 He, because they have a simple structure, to bombard targets such as 12 C, 48 Ca, 208 Pb, that are the subject of the study. Let us consider the case of 32 Mg as example. The lifetime is too short to prepare this nucleus as a target. However 32 Mg can be provided by recent facilities as a beam. Then the equivalent studies imply that the light particles p, d, 3 He, 4 He and so on have to be the targets. This is what is meant by inverse kinematics. The change from direct to inverse kinematics does not, of course, change the physics of the process to be studied: in the center of mass system, the two are completely equivalent. However, in the laboratory system, several important changes take place. To illustrate this, in Fig. 1 the relative energy variation dE /E for a change of excitation energy of 100 keV, for a reaction 32 Mg(d, p) is shown for the most important forward angles in the center of mass system. In normal kinematics, the energy of the beam-like outcoming proton is changed by about 0.6%. This change

126

S. Beceiro-Novo et al. / Progress in Particle and Nuclear Physics 84 (2015) 124–165

Fig. 1. (Color online) Example of a transfer reaction at 15 MeV/n in normal 32 Mg(d, p)33 Mg and inverse kinematics d(32 Mg, 33 Mg)p. The relative energy change, i.e. the change of energy for a difference in excitation energy of 100 keV divided by the energy, is shown for forward angles in the center of mass.

Fig. 2. (Color online) Simulated energy versus angle plot of the hypothetical reaction 78 Ni(d, d′ ) at 100 MeV/n populating a giant monopole resonance. The simulation includes angle and energy loss straggling in the detector gas and electronic resolution of electronics. The center of mass angles of the inelastic scattering are indicated, too.

of energy can be resolved by standard Silicon detectors or by a medium resolution magnetic spectrometer. In inverse kinematics, the relative energy change for the beam-like 33 Mg is only about 10−4 , and the center-of-mass angular range of Fig. 1 is covered by one degree only in the laboratory frame. This, together with eventual Doppler broadening if γ -rays are emitted, would not allow to achieve the necessary resolution even with the best high resolution spectrometers. On the other hand, the relative energy change for the recoiling proton is actually higher than in normal kinematics, more than 1%. As a result, the detection and characterization of the light recoil particle is mandatory in inverse kinematics in order to obtain the necessary resolution for the excitation energy of the heavy partner. 2. In direct reactions at energies well above the Coulomb barrier, the most forward center of mass angles are important to extract information about the transition, such as angular momentum and transition strength. In this case the momentum transfer is small, therefore the recoil energy of the light particle is also small. This is illustrated in Fig. 2, for a hypothetical inelastic scattering leading to a giant monopole resonance in 78 Ni. The recoil energy of the deuterons is very low, only 200 keV. In a standard experiment with a solid target, the target would have to be extremely thin, incompatible with the low intensity of such an exotic secondary beam. Here the simulation is shown for 0.25 atm of D2 gas and for a magnetic field of 2 T, in an active target device such as the AT-TPC (see Section 2.7). The figure shows that a very good

S. Beceiro-Novo et al. / Progress in Particle and Nuclear Physics 84 (2015) 124–165

127

Table 1 Active targets in operation or being constructed. Name

Lab

Gas ampl.

Ikar Maya ACTAR MSTPCb CAT MAIKo pAT-TPC AT-TPC TACTIC ANASEN MINOS O-TPC

GSI GANIL GANIL CNS CNS RNCP MSU FRIB TRIUMF FSU/LSU IRFU TUNL

NA Wire µmegas Wires GEM µ-PIC µmegas µmegas GEM Wires µmegas Grid

a b c

(cm3 )

Volume

Pressure (atm)

Energy (MeV/n)

Electronics

Number of chan.

Statusa

Ref.

60 · 202 π 30 · 28.32 203 70 · 15 · 20c 10 · 10 · 25 143 50 · 12.52 π 100 · 252 π 24 · 102 π 43 · 102 π 6000 21 · 302

10 0.02–2 0.01–3 <0.3 0.2–1 0.4–1 0.01–1 0.01–1 0.25–1 0.1–1 1 0.1

&700 2–60 2–60 0.5–5 100–200 10–100 1–10 1–100 1–10 1–10 >120 ∼10

FADC Gassiplex GET FADC FADC FADC GET GET FADC ASIC Feminos Optical CCD

6*3 1024 16,000 128 400 2 × 256 256 10,240 48 512 5000 2048 · 2048 pixels

O O C, P O T T T, O O T O O O

[6] [7] [8] [9,10] [11] [12] [13] [14] [15] [16] [17] [18]

O: operational, C: under construction, P: Project, T: test device. Two GEM versions: GEM-MSTPC (CNS) [19,20] GEM-MSTPC (KEK) [21,22]. GEM-MSTPC (CNS): 23.5 · 29.5 · 10.0, GEM-MSTPC (KEK): 10.0 · 10.0 · 10.0.

resolution can be obtained, about 200 keV (FWHM) in the invariant mass system. This is about the same as is obtained in direct kinematics using the best magnetic high resolution spectrometers. 3. The use of thick targets is mandatory with the most exotic secondary beams in order to achieve a luminosity high enough to measure the reactions of interest. Due to the very low recoil energies, most experiments using solid targets need to use very thin targets, of the order of 100 µg/cm2 or less, in order to preserve the resolution. For an active target device, the reaction vertex can be determined, therefore the reaction kinematics can be fully corrected for the energy loss of the beam particles in the target gas prior to the reaction. At the extreme, the target gas pressure may be adjusted to stop the beam particles in the detector, and the full energy range from incoming energy to zero is covered in the detector, providing a direct way to measure the excitation functions of reactions in a single setting. This is of particular interest for resonant reactions, as for example in the low energy domain for reactions of astrophysical interest and in the study of resonances in light nuclei related to quasi-molecular structures. In addition, the solid angle for detection in an active target detector is essentially 4π , the best coverage possible. In the following, we will describe existing active target detectors listed in Table 1, examples of physics studied with them, detectors under study or being constructed, and specific technical issues. As shown in the table, various gas amplification systems are used. A recent review of micropattern gas amplifiers can be found in Ref. [5]. There are other projects in an early stage of definition, such as projects at KUL (Belgium), at Texas A&M, in South Korea and China, and maybe others we are not aware of, that are still in a too preliminary stage to be included. TPC’s not used as active targets and other closely related detectors used for decay studies, are out of the scope of this review. The detectors described were designed and optimized with specific scientific programs in mind. Consequently, we will discuss the specific physics examples together with the detector characteristics. Then, more technical aspects will be presented, in particular the electronics and data acquisition systems used to readout these detectors. Finally, a conclusion will outline the present and future works in progress in this exciting new domain of experimental nuclear physics. 2. Examples of active targets 2.1. Ikar The Ikar detector was first developed to study high energy (1 GeV) p–p collisions [23]. This detector uses H2 gas at 10 atm as target and detector material. It could perform measurements with intensities of 104 –105 particles/s. Later it was adopted for the elastic scattering of light exotic beams at GSI [6]. It is integrated in the general beam line structure with the standard beam diagnostics that determine the secondary beam particle characteristics, such as incident angle and position. In some of the experiments it was followed by the large gap magnetic analyzer Aladin (A Large Acceptance DIpole magNet) to identify the outcoming reaction products for better background suppression. The scheme of the detector is shown in Fig. 3. It has a cylindrical symmetry around the beam axis. In the version shown it contains six 10 cm long ionization chambers. This subdivision was chosen in order to limit drift times in the high pressure H2 gas, at a maximum cathode voltage of 15 kV. The drift time for 10 cm is 23 µs. The scheme of one of these ionization chambers is shown in Fig. 4. Elastic nucleon scattering is and has been used since a long time to determine matter distributions in stable nuclei. A recent discussion can be found in [24] and references cited. A large number of precision elastic scattering experiments were performed at GSI exploring the halo structure in light nuclei such as 6 He, 8 He, 11 Li, 12,14 Be. As an example we show here the

128

S. Beceiro-Novo et al. / Progress in Particle and Nuclear Physics 84 (2015) 124–165

Fig. 3. (Color online) Schematic view of the ionization chamber IKAR consisting of six identical modules. The thicknesses of the anodes and cathodes are shown enlarged. The positions of the calibration α -sources are indicated. Source: From [6].

Fig. 4. (Color online) Schematic view (not in scale) of one of the six chamber modules of IKAR, along with a typical event of scattering of a Be projectile on a target proton in the space between grid and cathode. TR denotes the proton recoil energy, z is the distance of the vertex point Z from the grid, and ΘR (ΘS ) is the recoil proton (projectile) scattering angle. Source: From [6].

results for 12 Be of a recent publication [25]. The experimental cross sections are divided by an exponential function, in order to enhance the visibility, as shown in Fig. 5. To increase the precision of the matter distributions extracted from absolute differential cross-sections, elastic (p, 6,8 He) scattering data were obtained at angles leading to the first diffraction minimum and spanning more than two orders of magnitude. They complement the earlier measurement at smaller angles performed at the same energy with Ikar. Both data sets agree well in the overlap region and enable a refined and comparative study of the radial distribution of the nuclear matter density [26]. A recent analysis of the combined data for 6,8 He can be found in [27]. An example of the analysis is shown in Fig. 6. The proton–nucleus elastic scattering at intermediate energies is a well-established method for the investigation of the nuclear-matter density distribution in stable nuclei. With the Ikar detector, this method was applied in inverse kinematics for the investigation of radioactive nuclei. Mainly two methods of analysis are used. One consists in using theoretical microscopic calculations as input for Glauber multiple scattering formalism. The other method minimizes different parametrizations for the matter distributions, such as single Gaussians, or sum of Gaussians, to achieve the best fit to the data, see for example [25]. The resulting error for the root mean square radius is typically of the order of some percent, an extremely valuable quantitative information to characterize radioactive nuclei.

S. Beceiro-Novo et al. / Progress in Particle and Nuclear Physics 84 (2015) 124–165

129

100 10–1 1.2 p(r), fm–3

dσ/dt/C0exp(B0t)

1.4

1

10–2 10–3

0.8 10–4

0.6 0

0.01

0.02 0.03 0.04 –t, (GeV/c)2

0.05

10–5

0

2

4

6 r, fm

8

10

12

Fig. 5. (Color online) Nuclear-matter density distribution deduced with the Sum Of Gaussian parametrization [25]. The error band in the density distribution on the right hand figure is determined by a χ 2 analysis of the contribution of the different Gaussian components.

104 103 102 101 100 10–1 10–2 10–3

0

10

20

30

40

50

Fig. 6. (Color online) Experimental results [28,29] for p + 6 He elastic scattering angular distributions at 717 MeV/nucleon compared to theoretical model [27]. For the detailed description of the different curves, see [27].

2.2. Maya The Maya detector is operating at GANIL since about 2003. A technical description can be found in [7]. We may mention that to our knowledge this may be one of the last projects in our domain of this importance that was realized without being controlled and approved by technical and financial committees. This was possible partially because of gradual upgrades, implying gradual increase of costs. It has a rectangular geometry, with the beam entering the detector parallel to the cathode, and in the center between the cathode and the imaging anode. The following design criteria were used for the Maya detector:

• The cost and security problems are limiting factors for the size and the operational pressure of gaseous detectors. Therefore the pressure was limited to 3 atm to achieve a reasonable low drift time without a too high voltage applied to the cathode. It is important to realize that safety issues for explosive gases are an integral part of the detector conception and its running. In particular special procedures must be defined for hydrogen and isobutane filling of the detector, running and pumping down. A vulnerable part of the detector is always the entrance window, that is chosen as thin as allowed by safety concerns. The height of the active area was limited to 20 cm and the width was chosen to be 25 cm to limit the area to be covered by expensive ancillary detectors. These dimensions allow to stop most of the GANIL–SPIRAL [30] beams and heavy reaction products. However light reaction products will in many cases leave the active area. • Therefore a wall of 20 (5 × 5 cm2 ) Si–CsI telescopes was added to stop the particles emitted to forward angles. The telescopes consist of 700 µm Si PIN diodes and the CsI crystals that are 1 cm thick. • For light charged particles and typical ranges of 10 cm, the TRIM code [31] calculates a range straggling around 1%. Hence a resolution for the range and position measurements of the same order of magnitude, or better, should be achieved. For position-sensitive gaseous strip detectors, a resolution of 1/10 of the size of the pads is easily achievable [32,33]. To obtain a resolution of 1 mm in the position or range measurements, the size of the pads needs therefore to be about

130

S. Beceiro-Novo et al. / Progress in Particle and Nuclear Physics 84 (2015) 124–165

Fig. 7. Schematic diagram of the internal structure of MAYA. The beam particles enter from the left. At the forward angles a wall of Si–CsI telescopes can detect particles that are not stopped in the gas.

1 cm (for square pads). An anode of 30 × 25 cm2 , requires about 1024 electronic channels. Instead of rectangular pads, hexagonal ones were adopted to minimize the effect of preferential directions. As the energy loss of light recoil particles can be very low, the electrons from the primary ionization need to be amplified. Proportional wires for amplification were chosen, with the wires being parallel to the beam. The detector chamber is a nearly cubic stainless-steel container, about 40 cm wide, with three large openings, at the front, rear and bottom, and flanges on the sides and top used for signal output, gas flow and high voltage. A frame holding the 20 CsI crystals and the 20 silicon detectors can be inserted between the container and the panel, see Fig. 7. The front panel has 4 openings for square flanges, used for signal output. The preamplifier cards for the proportional wires are directly plugged on one of these flanges via airtight connectors. The back panel is a machined piece of epoxy, with a 1.3 cm diameter Mylar entrance window. The thickness of the entrance window is chosen according to the gas pressure, for example a 6 µm thick window is used for a gas pressure of 2 atm. The vessel has been tested and safety approved for the use at gas pressures up to 3 atmospheres, including explosive gases. The materials used have been chosen to reduce out-gassing, another common problem of gas detectors. The pollution by even small amounts of impurities can drastically alter the amplitude of the signals, due to electron–ion recombination [34]. Hence it is particularly important to prevent outgassing or pollution by atmospheric gases. As an example, the detector was filled with D2 gas and then sealed. During this experiment the change in signal amplitude was less than 20% after several days of running. Inside the vessel the active volume is 283 mm long (X ), 258 mm wide (Y ) and 200 mm high (Z ). The internal structure is fixed to the front panel by 4 machined bars, giving an easy access to the structure by removing the front panel. Printed circuit boards (PCB) with copper field strips are mounted on the front panel and on each side of the active area to obtain a homogeneous electric field between the cathode and the Frisch grid at the bottom. The forward board of field strips is replaced by a frame holding field wires to allow the detection of the escaping particles at forward angles in the solid state telescopes. On top there is a stainless steel cathode. The active area is separated from the detection area by a Frisch grid. The amplification volume contains the plane of proportional wires taut along the beam axis, and the detection plane of hexagonal-shaped pads. The amplified signal from the wires induces a signal on the pads, thus allowing to determine the center of gravity of the charge. The distance between the Frisch grid and the proportional wires plane is 15 mm and the distance between the proportional wires and the pads is 10 mm. This geometry is optimized for the best resolution given the size of the pads. The diameter and spacing of the proportional wires can be chosen depending on the experiment. Frames with 5, 10 and 20 µm diameter gold-plated tungsten wires were used. The 5 and 10 µm wires were quite difficult to use and turned out to be very fragile, therefore mostly the 20 µm wires were used. Two different spacings between the wires with 2 or 3 wires per row of pads were used. GASSIPLEX [35] daughter boards for the pads are plugged underneath the detection board. The hexagonal pads have a side length of 5 mm, aligned along the beam axis. There are 37 rows of 35 pads, this is 1295 pads, but only 32 rows of 32 pads are used. The pad board is a multi-layer printed circuit. The tracks from the pads to the GASSIPLEX are printed on the inner layer of the board. This board is the most important part of the detector: it contains the connections, grounding, and powering tracks. It is composed of 8 layers to fulfill these needs. The GASSIPLEX electronics need an external track and hold signal that freezes the pulse amplitude in the chip memory. The pulses are induced by the wires and therefore the wire signals are used to produce this signal. A multiplexed readout transfers the amplitudes to a synchronized flash ADC. Different algorithms for track analysis were tested [36]. The different methods for tracking reconstruction for twodimensional projected tracks determine projected angles, as well as the stopping and starting points of the measured tracks.

S. Beceiro-Novo et al. / Progress in Particle and Nuclear Physics 84 (2015) 124–165

131

300

200

100

0

1

1.5

2

2.5 3 E c.m. (MeV)

1

1.5

2

2.5 3 E c.m. (MeV)

300

200

100

0

Fig. 8. 12 C + p elastic scattering cross-section as function of the center-of-mass energy (energy in 13 N above the proton breakup threshold), measured at two center-of-mass angles as indicated in the panels. The red line is a R-matrix fit. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.) Source: From [37].

The resolution possible in these measurements was extracted from simulations of realistic experimental setups, covering the different experiments performed with MAYA. These reconstruction techniques may be used in similar detectors where the tracking is obtained from segmented images of the trajectories. The use of an active target for the measurement of high quality resonance studies is illustrated by the results obtained with a stable 12 C test beam with the Maya detector at Cern-Isolde [37] as shown in Fig. 8. The detector contribution to the resolution of the excitation function was 20 keV (σ ). This demonstrates that in the future many resonance studies with low intensity beams will become possible, and that Silicon detectors may be used in combination with the active target. In the thick target method of measurement of resonances by the detection of light recoil particles near zero degrees the energy of the scattering can be deduced from the energy of the detected particle. However, there are two unknowns, the energy at which the scattering occurred and the excitation energy of the outgoing heavy partner. This may lead to confusion because two excitation functions overlap, elastic and inelastic scattering [38]. This ambiguity is lifted by the determination of the reaction vertex with an active target. A large variety of experiments was realized with this detector. A description of early results can be found in Ref. [39]. The study of the halo nucleus 11 Li was one of the first experiments [40]. The angular distributions for the 11 Li(p, t)9 Li was recently re-analyzed in Ref. [41]. The comparison of experiment with theoretical calculations is shown in Fig. 9. Potel et al. [41] conclude that through a unified structure and reaction nuclear field theory with a Feynman diagrams analysis

S. Beceiro-Novo et al. / Progress in Particle and Nuclear Physics 84 (2015) 124–165

dσ/dΩ (mb)

132

101

100

10–1

10–2

0

20

40

60

80

100

120

140

160

180 θCM

Fig. 9. Experimental [40] and theoretical differential cross sections (including multistep transfer as well as breakup and inelastic channels) of the 1 H(11 Li, 9 Li)3 H reaction populating the ground state 3/2− and the first excited state 1/2− at 2.69 MeV of 9 Li. Source: For details see Ref. [41].

of the experiment of Tanihata et al. [40] they could show that the virtual quadrupole vibration of 9 Li plays a central role in polarization effects for nuclear Cooper pair stabilization in 11 Li. This example shows that loosely bound systems open a window on new phenomena, and unexpected behaviors near the dripline. Missing mass measurements are an important tool to determine states in the unobserved reaction partner. This type of reaction allows to observe and characterize even states that are particle unbound. An example is the reaction 8 He(12 C, 13 N)7 H studied with the Maya detector [42]. Detection of the very low energy of the 13 N recoil nuclei was possible operating at very low pressure, 20 mbar. This experiment illustrates the unique feature of active target-TPCs to detect very low energy particles by adjusting the pressure of the detector gas. A discussion of this result can be found, too, in ‘‘Probing nuclear forces at the extreme of isospin: the 7 H resonance’’ [43]. This resonance has been observed and characterized, even if statistics were low, only 7 events. Non-perturbative description of the decay of such systems is without any doubt a major challenge for present theories. The multi-particle decay of nuclei near the drip-line was described in Ref. [44] and references cited. Binding energies of dripline nuclei can be measured by missing or invariant mass spectroscopy. This is illustrated by the study of the reaction p(11 Li, 9 Li)t already cited above. By measuring the reaction Q -value leading to 9 Li that has a known mass, the mass of 11 Li could be determined with a precision of ±22 keV [45] by energy–energy and angle–angle kinematics reconstruction. The derived 11 Li two-neutron separation energy is S2n = 363(22) keV. This value is in good agreement with values from the MISTRAL transmission spectrometer [46] and the TITAN penning trap measurement [47] of S2n = 378(5) keV and S2n = 369.15(65) keV respectively. These measurements were significantly different from the Atomic mass evaluation at this time [48] of 299.6(19.4) keV. The corrected value was important to evaluate the binding energy effects and understand the nature of the 2 neutron halo of 11 Li. The study of giant resonances by inelastic scattering is an example of the use of Maya at high energy. Due to the negative Q -value, and in order to have predominantly a direct reaction mechanism, energies of 50 MeV/n or higher are necessary. Two Ni isotopes have been studied with the Maya detector at GANIL: 56 Ni and 68 Ni [49,50]. We will illustrate the results for 68 Ni, a neutron rich isotope. The inelastic scattering spectra were decomposed in contributions as shown in Fig. 10. For the first time, an indication of a soft monopole mode around 13–14 MeV and of isoscalar dipole strength in the same region is obtained. The multipole decomposition of the angular distributions for different excitation energies is shown in Fig. 11. The soft monopole resonance can be understood as a neutron compressional mode, without participation of the protons, as shown in Fig. 12(a) from [50,51]. The 21 MeV resonance is described by an oscillation of protons and neutrons in phase, this is the extensively studied monopole compressional mode, important for its relation to the nuclear incompressibility. The Maya detector showed clearly the large possibilities of such a detector: with secondary beams of down to ∼1000 particles/s, a full range of experiments, with beam energies from some MeV/nucleon to more than 50 MeV/nucleon, resonances, transfer reactions and inelastic scattering were successfully studied. 2.3. Active target (ACTAR) The design of the ACTAR detector is based on the experience and results obtained with the original MAYA detector (see previous Section 2.2). The geometry of the design is similar, but several improvements are implemented in order to

S. Beceiro-Novo et al. / Progress in Particle and Nuclear Physics 84 (2015) 124–165

133

Fig. 10. (a) Excitation energy spectrum of Ni obtained with reaction 68 Ni(α, α ′ ), for all angles deduced from the alpha kinematics and corrected for efficiency. (b) Same with a zoom on the ISGR region. The subtracted background is represented in green dot-short dashed line. The resonances, resulting from the fit, which are expected to be monopole are represented in red solid line, and the one to be quadrupole is represented in blue large dashed line. The solid black line corresponds to the sum of all contributions. (c)–(g) the same at center-of-mass angles indicated. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.) Source: From [50] and private communication.

alleviate some of the limitations, as well as expand the domain of reactions for which this detector can be used. These include in particular resonant scattering, inelastic scattering, and transfer reactions. The improvements required are the following: (i) detect particles with large differences in charge, (ii) increase spacial resolution to resolve multiple tracks and (iii) improve rate capability. They can be realized by a combination of design changes in detector hardware, electronics and data acquisition. The projected detector hardware is shown schematically in Fig. 13. It is composed of an active volume of cubic geometry surrounded by a field cage where the electric field drifts the primary electrons towards a pad plane. This pad plane is made of 2 × 2 mm2 independent pads. In the case of Maya the timing was giving per wire, meaning that a whole row of pads has only one timing information, thus if several particles arrive at the same time they cannot be detected. To overcome this problem, all pads in ACTAR have an individual timing. The electron amplification device uses the Micromegas technology [52], providing gain factors ranging from 103 to 106 depending on the gas and its pressure. The beam entrance side of the cube is equipped with a thin window, whereas other sides can be tiled with ancillary detectors aimed at detecting and identifying the light particles that do not stop inside the gas volume. The total number of square pads covering the active area is about 16,000. The geometry of the detector can be adjusted depending on the particular needs of the experiment. For instance when the detection of beam particles is of no interest, the pad plane can be oriented perpendicular to the beam direction in order to minimize the number of pads that are blind due to the projection of the beam tracks. The maximum

dσ/dΩ [mb/sr]

S. Beceiro-Novo et al. / Progress in Particle and Nuclear Physics 84 (2015) 124–165

dσ/dΩ [arb. units]

134

dσ/dΩ [mb/sr]

θCM [deg]

θCM [deg] θCM [deg] Fig. 11. (a) Experimental angular distribution extracted for the resonance at 12.9 MeV and fitted with theoretical prediction assuming a ℓ = 0 multipolarity (a1), ℓ = 1 multipolarity (a2), or ℓ = 2 multipolarity (a3). (b) The same for the resonance at 15.9 MeV assuming a ℓ = 2 multipolarity. (c) The same for the resonance at 21.1 MeV. Source: From [50] and private communication.

Fig. 12. (a) Proton (solid line) and neutron (dashed lines) (b) Same at 21 MeV. Source: From [50].

68

Ni monopole transition densities at 13 MeV, calculated with the RPA and SkI2 interaction.

rate of incoming beam particles is set to 106 particles/s, which is rather ambitious given the typical drift velocities of about 5 cm/µs that can be achieved in Time Projection Chambers. For a drift length of about 20 cm, the average multiplicity of beam tracks over this drift distance is about 4 at this rate. Therefore resolving the track that produced the reaction from pileup will depend solely on the performance of the sampling electronics and the shaping time of the pulses. The electronics chosen to read the pads comes from the GET project (General Electronics for TPC) [54]. It is based on a high density analog sampling ASIC (Application Specific Integrated Circuit) that buffers the signals continuously until a good trigger condition is reached (see Section 3.1). The configuration shown in Fig. 13 was simulated for a (d, p) transfer reaction on 132 Sn at 5 MeV/u (see Fig. 14). Note that other reaction channels such as (d, d′ ), (d, t) or (d, 3 He) can be measured simultaneously, provided the kinematic region where they occur is covered by the geometry of the Si–CsI telescopes. At atmospheric pressure of D2 gas, the total thickness

S. Beceiro-Novo et al. / Progress in Particle and Nuclear Physics 84 (2015) 124–165

135

Rangeproton [mm]

Fig. 13. Schematic drawing of the ACTAR detector, shown in the configuration for the measurement of (d, p) transfer reaction in inverse kinematics. Two examples of the recoiling proton tracks are shown, one with low energy stopping in the gas volume, the other escaping to one of the Si–CsI telescopes, here placed on the sides of the gas volume. Source: From [8], see also [53].

200

150

100

50

0 80

100

120

140

160 θlab [°]

Fig. 14. Particle identification simulated for protons for the 132 Sn(d, p) reaction at 5 MeV/u feeding excited states at 854, 1363, 1561 and 2005 keV. The estimated resolution is 150 keV. The coordinates are the trajectory length on the vertical axis, and laboratory scattering angle on the horizontal axis. Source: From [55].

corresponding to a 20 cm length is about 1.1 × 1021 atoms/cm2 , which is 30 times more than a typical CD2 solid target of 0.5 mg/cm2 thickness. As a result, good statistics can be obtained with beam intensities down to 100–1000 particles/s, as opposed to the 104 /s typically necessary for this type of measurement. In addition, the determination of the vertex location means the influence of the energy loss in the target material on the final energy resolution is reduced by more than an order of magnitude with respect to a solid target. The protons emitted at backwards angle from the (d, p) reaction are clearly visible down to about 110°, until the range of the protons is long enough to hit the Si–CsI telescopes. At these higher energies, the particle identification is relayed by the classic E–∆E technique. The GET electronics have a feature specific for decay experiments such as two-proton decay of very neutron-deficient isotopes. In this case the radioactive nuclei are implanted (stopped) inside the gas volume from the typical fragmentation energies where they are produced (between 50 and 300 MeV/u). The challenge is to observe their subsequent decay by detecting protons of energies in the MeV range. This is solved in the GET electronics by the possibility to use half of the analog memory during implantation and the other half for the decay. Although not an active target application per se, this technique is clearly the best as it provides nearly 100% efficiency, as well as the possibility to directly measure the angular and energy correlations between the protons.

136

S. Beceiro-Novo et al. / Progress in Particle and Nuclear Physics 84 (2015) 124–165

Fig. 15. Drawing of MSTPC showing its geometry and dimensions. Source: Figure taken from Ref. [9].

2.4. Multi-Sampling and Tracking Proportional Counter (MSTPC) The Multi-Sampling and Tracking Proportional Chamber (MSTPC) is a gas-filled detector with a rectangular geometry that was constructed for use with low-energy radioactive beams [9]. It has gone through a series of upgrades and modifications that have enhanced the capabilities of the detector that will be discussed in this section. The MSTPC is motivated by the study of nuclear structure and reactions especially as it relates to element production in astrophysical scenarios. The MSTPC is also used to measure cross sections for light-ion reactions that are relevant to nuclear astrophysics such as (α , p) and (α , n) at low-energy. At low-energy the cross sections are typically very small and thus techniques that can improve statistics can play a large role. Several experiments were performed to study the possible seed nuclei for the r-process. Cross sections have been measured using the MSTPC that are needed for understanding reaction rates relevant to understanding the rp-process, break-out from the hot CNO cycle, element production in X-ray bursts. Predecessors to the MSTPC were the Multi-Sampling Ionization Chamber (MUSIC) [56] and Multi-Sampling Proportional Counter (MSPC) [57], which have been successful in studying reactions with radioactive beams [58,59]. Recently, a newlyconstructed MUSIC detector was used in conjunction with radioactive beams to access reactions for unstable carbon isotopes that shed light in nucleosynthesis of neutron-rich isotopes on the surface of neutron stars [60]. It was realized that if the MSPC was used as a TPC, one would be able to distinguish reaction channels that involve multiple charged-particles in the final state, a capability that was not possible in the MUSIC or MSPC detectors. To record vertical positions in the MSTPC, fast sampling of the electron signals were needed. To meet this need, Flash Analog-to-Digital Converters (FADC’s) were used and a custom data acquisition system was used to handle the large amount of data from the FADC’s. The MSTPC is an active-target detector that is able to handle up to a few hundred Torr of gas. It consists of a rectangular field cage housed in an aluminum vacuum chamber. The field cage consists of a high-voltage cathode plate held with four insulating G10 pillars that are wrapped with Cu–Be conducting wires that provide a uniform electric field down to the grid and anode wires. A schematic diagram of the detector is shown in Fig. 15. In the original version of the MSTPC, the primary electrons that were created by the ionization of charged particles in the detector’s gas volume were drifted to the grid and multi-wire proportional-counter anode wires by the uniform electric field provided by the field cage. Once the primary electrons reached the grid and anode wire cell, they were amplified by the large electric field near the anode wire and collected on the wire. The induced signal was detected on the surrounding cathode pads in the detection cell. The cell was three sided in order to measure more of the induced charge. The geometry of the cell can be seen in Fig. 15. The original MSTPC pads had a strip design which was long in the perpendicular beam direction and short in the beam direction. It also had a triangular pad design called the backgammon pattern, which is shown in Fig. 16. This pattern could be used for determining the position in the direction perpendicular to the beam by taking the ratio of charge collected on adjacent pads [61]. The vertical position is determined by the drift time of the electron tracks. Measurements have shown that position resolution was better than 1 mm. The pad signal was sent to a charge-sensitive preamplifier, shaping amplifier, and then digitized by FADC’s. The FADC’s had up to a 70 MHz sampling rate with 8-bit resolution. The original data acquisition could accommodate up to 20 events/s with 20 kB/event.

S. Beceiro-Novo et al. / Progress in Particle and Nuclear Physics 84 (2015) 124–165

137

Fig. 16. Schematic diagram of MSTPC pad plane. The backgammon geometry of the pads is shown. Source: Figure taken from Ref. [9].

Fig. 17. Schematic diagram of GMSTPC showing the implementation of the gating grid. The gating grid is opened by the detection of neutrons in an auxiliary detector. Source: Figure taken from Ref. [10].

The MSTPC as a stand-alone device can be triggered in two ways. The discriminator for all channels can be set with common level to detect an abrupt change in the beam-track charge density such as from those produced by heavier nuclei produced in a transfer or fusion reaction similar to the MUSIC detectors. In the original detector, the total anode wire signals . The MSTPC can also be triggered by using can also be used to trigger the system when there is an abrupt change in dE dx auxiliary detectors such as multi-channel plate (MCP) detectors, parallel-plate avalanche counters (PPAC), neutron barrel counters. The original MSPTC was limited to beam rates of 104 particles/s due to non-negligible space-charge effects [62]. To deal with these space-charge effects at higher beam intensities, a gating grid was implemented. This allows for a total beam rate higher than the 104 particles/s range. A schematic of the MSTPC with the gating-grid is shown in Fig. 17. Auxiliary neutron detectors are used to trigger the opening of the gating-grid to allow for the amplification and recording of ionization tracks when a neutron is detected from a reaction such as 8 Li(α, n)11 B.

138

S. Beceiro-Novo et al. / Progress in Particle and Nuclear Physics 84 (2015) 124–165

Fig. 18. Schematic diagram of GEM-MSTPC. Double GEM configuration as well as the pad plane is shown. Source: Figure taken from Ref. [22].

The gating grid’s performance was tested in a number of experiments under various beam conditions at the University of Tsukuba and the JAERI Tandem facility. Space charge effects were found to be less than 7% for beam rates up to 105 particles/s, which show the effectiveness of the gating grid [10]. To further increase the beam injection rate into the MSTPC, the use of gas electron multiplier (GEM) foil [63] for high rates was implemented to replace the MSTPC’s multi-wire proportional counter. Using a multi-wire proportional counter, the MSTPC’s rates were limited to about 104 particles/s. Two detectors based on the original MSTPC design of Mizoi et al. [9] were constructed at different laboratories. Both are called GEM-MSTPC in the literature. One GEM-MSTPC detector was constructed at the CNS Tokyo has an active-volume dimension of 23.5 cm × 29.5 cm × 10.0 cm. It has two type of anode pads with a backgammon style geometry, low-gain pads in the central beam region and high-gain pads in the outer regions for detection of light charged-particles [19,20]. The other GEM-MSTPC was built at KEK and has an active-volume dimension of 10 cm × 10 cm × 10cm [21,22]. Various GEM foils of different thickness (50 µm, 400 µm) in single-, double-, and triple-foil configurations were tested for use in this GEM-MSTPC. An example of the double-GEM configuration is shown in Fig. 18. The use of GEM foils can increase the MSTPC’s ability to use beam injections rates by an order of magnitude to 105 particles/s [21,64]. A triple-foil 400-µm-thick GEM configuration was also used to study and optimize the suppression of positive ion feedback, the phenomenon of positive ions moving back into the drift region and distorting the electric field. Using this configuration, the measured positive ion feedback ratio was reduced to <2% [21]. With the ability to increase the beam injection rates in the MSTPC combined with a large gain in the THGEM, the effects of positive-ion feedback becomes important, even for low ratios <2%, as it can result in enough positive ions to distort the local electric field of the drift region [65]. A model of electric field distortion due to positive ion feedback has been made as presented Ishiyama et al. [22], which is based on Orthen et al. [65]. The distortions in the electric field due to the backflow of positive ions were modeled as a static pillar of positive charge, which represents the average amount of positive charge for a given beam rate, amplification gain, and backflow percentage. The total effect on the drift time of a primary electron coming from ionization due to the beam can be obtained by integrating the additional distorting electric field over the path of the drift electron and this can be compared with data. The calculated field distortions are for a 12 C beam with an intensity of 105 particles/s. Calculated field distortions as well as measurement of backflow, electron transmission, and drift velocity are shown in Fig. 19. Measurement of electron transmission was deduced from the measured pulse height and positive ion backflow by measuring the signal on the drift cathode with a sufficiently long integration time, 100 µs in this particular case. The measurement of the variation in the electron drift times was measured as a function of the drift electric field. Larger fields suppressed the distortion and the

S. Beceiro-Novo et al. / Progress in Particle and Nuclear Physics 84 (2015) 124–165

139

Fig. 19. Field distortion calculated from positive-ion feedback model for GEM-MSTPC. Source: Figure taken from Ref. [22].

measured value of the distortion 2.9% was close to the value calculated from the ion pillar model, 2.4%. Position distortion due to ion feedback was reduced to 1.5 mm because of the high drift field value of 1 kV/cm atm and the use of double GEM configuration [22]. There have been recent developments in electron-amplification devices that completely eliminate ion backflow [66]. Development in electron amplification technology such as the zero ion backflow electron multiplier would allow for very large range of beam rates that can be accommodated in active-target detectors. Last, the extraction of relevant observables requires a robust reconstruction of the ionization tracks recorded in the MSTPC and similar detectors. Progress in this direction is being made by developing the tracking algorithms that are necessary for MSTPC experiments [67]. A more general discussion of tracking for active-target detectors can be found in Section 5. The MSTPC has been used in a number of experiments to measure excitation functions of reactions with light radioactive beams. Many of these measurements were motivated by the study of nucleosynthesis in astrophysical environments. These include Big Bang nucleosynthesis, X-ray bursts, and supernovae. One of the most well measured reactions is 8 Li(α, n)11 B, which has a role in Big Bang nucleosynthesis and the study of r-process seed nuclei. Another motivation for studying this reaction is to investigate the theory of a ‘‘hot bubble’’ in supernova and the formation of r-process seeds. In an initial experiment shown by Mizoi et al. [68], a radioactive beam of 8 Li was produced by projectile fragmentation at RIKEN degraded to 2–4 MeV/u with a rate of 2000 particles/s. The MSTPC was filled with a He:isobutane mixture in a ratio of 90:10 with a pressure of 400 Torr. The electron drift velocity was 1 cm/µs. The vacuum chamber 100 µm mylar-foil windows were so that neutrons emitted in the reaction could be measured with auxiliary neutron detectors that surrounded the MSTPC. An excitation function of the 8 Li(α, n)11 B reaction is shown in Fig. 20. The astrophysical S factor derived from the measured 8 Li(α, n)11 B reaction cross section is shown in Fig. 21. [68]. Several additional measurements of the 8 Li(α, n)11 B cross section were performed with higher statistics at the JAERI Tandem facility that include lower energy [69,70]. The gating-grid MSTPC was used and the measured excitation function is shown in Fig. 22. An energy range of Ecm = 0.45–1.75 MeV was measured [70,71]. There is a significant discrepancy between the inclusive and exclusive measurements of the 8 Li(α, n)11 B reaction. La Cognata et al. state thresholds for neutron detection in the exclusive measurements may have excluded the detection of states in 11 B, but more detailed measurements in the low-energy region of the cross section are important and needed [72]. A series of other measurements have been performed at the JAERI Tandem facility and the Center for Nuclear Study Radioactive-Ion Beam (CRIB) facility located at RIKEN to measure other reactions that are of relevance for astrophysics. Although some of the results have been presented as preliminary, the scope of the reactions shows what some of the applications are for studying low-energy reactions with radioactive beams. At the JAERI Tandem facility, the 16 N(α, n)19 F reaction was studied with a 32-MeV 16 N beam. A preliminary excitation function was measured for the energy range Ecm = 1.5–4.5 MeV [73]. Another reaction important for nucleosynthesis,

S. Beceiro-Novo et al. / Progress in Particle and Nuclear Physics 84 (2015) 124–165

Cross section (mb)

140

1000 900 800 700 600 500 400 300 200 100 0

0

1

2

3

4

5

6 7 Ec.m. (MeV)

S factor (MeV barn)

Fig. 20. Measured excitation function of 8 Li(α, n)11 B reaction in the MSTPC is shown. The exclusive measurement of the reaction cross section is compared with other inclusive measurements [68].

104

103

102

10

1

0

1

2

3

4

5

6

7

Ec.m. (MeV) Fig. 21. The astrophysical S factor deduced from the measured 8 Li(α, n)11 B reaction [68].

B(α, n)15 N was also studied at the JAERI facility [74]. At the CRIB facility, the 22 Mg(α, p) reaction, which is important for understanding the α p-process in Type I X-ray bursts, was studied with the GEM-MSTPC. Cross sections in the energy range of 1.0–4.2 MeV in the center-of-mass frame where measured, which corresponds to a Gamow energy window of 1–3 GK. MSTPC was used with auxiliary Si detectors to measure scattered α particles and protons [75]. A measured excitation function is shown in Fig. 23. The 18 Ne(α, p)21 Na reaction was also studied at CRIB due its possible role in the breakout from the hot-CNO cycle to the rp process. Cross sections were measured in the 0.5–3.4 MeV center-of-mass energy range [19]. In addition to the 18 Ne(α, p) and 22 Mg(α, p) reaction, the 30 S(α, p) reaction was also pursued. These three reactions are predicted to affect the total energy generation in X-ray bursts by more than 5% [20]. 12

Cross section (mb)

S. Beceiro-Novo et al. / Progress in Particle and Nuclear Physics 84 (2015) 124–165

141

1000

800

600

400

200

0 0.0

0.5

1.0

1.5

2.0

2.5

3.0 Ecm (MeV)

Fig. 22. An excitation function of 8 Li reaction with higher statistics and measured to lower energy with the MSTPC is shown. This is compared to the existing measurements from inclusive measurements [70].

200

150

100

50

0 1

1.5

2

2.5

3

Fig. 23. Measured excitation function of 22 Mg(α, p) reaction. Peaks corresponding to excited states can clearly be seen. Source: Figure reproduced from Ref. [20].

2.5. Center for Nuclear Study Active Target (CAT) The Center of Nuclear Study Active Target (CAT) is a detector that has been developed to study reactions with fast beams that require a low-energy threshold for scattered or emitted light-ion particles. Such situations arise when looking at small center-of-mass scattering angles in experiments using fast radioactive-ion beams produced by fragmentation. Small centerof-mass scattering angles result in scattered particles with low energy. It can overcome the difficulty of using solid targets for detecting low-energy protons and α particles that would normally be stopped in the target material. Using a gas target allows for a very low-energy threshold for light-ion charged particle detection [11]. The following description of the CAT detector is largely taken from the Center for Nuclear Study (CNS) Annual Reports. The CAT detector is focused on studying the structure of unstable nuclei using a gas target for reactions. Inelastic scattering using unstable medium and heavy nuclei can be used as a probe of the nuclear equation of state. Isoscalar reactions such as (α , α ′ ) and (d, d′ ) can be used to probe monopole giant resonances. The properties of giant resonances are related to nuclear incompressibility and nuclear symmetry energy, which have implications for understanding the properties of neutron stars. Gamow-Teller (GT) transition strengths are important for determining electron capture for understanding nucleosynthesis of nuclei from iron to uranium. GT transitions strengths can be measured with the (d, 2p) charge-exchange reaction. CAT’s ability to track particles would make it very well suited to measure the two outgoing protons from this reaction. These examples highlight that the choice of reactions allows for spin–isospin selectivity, which can be used to probe specific levels or proton–neutron dependencies in nuclear structure [11].

142

S. Beceiro-Novo et al. / Progress in Particle and Nuclear Physics 84 (2015) 124–165

Fig. 24. Drawing and photograph of the CAT detector is shown. The configuration of the field cage, GEM detectors, and auxiliary Si detectors is shown. Source: Figure is taken from Ref. [81].

The measurement of light-ion recoils at forward scattering angles are important for the study of various aspects of nuclear structure. For example, reactions that occur near the nuclear surface have large cross sections in the forward scattering angle for L = 0 transfers, which are important for extracting the strength of giant monopole resonances through multipole decomposition [76,77]. The CAT detector can detect these low-energy forward scattered recoils due to its use of a gas target. Two versions of the Prototype CAT detector were started in collaboration with the SHARAQ, CRIB, and quark-physics group [78,79]. Both detectors have a rectangular geometry and has a similar layout to the MSTPC. One of the Prototype CAT detectors has a double field cage configuration, which defines two different drift regions. The sensitive detector region excludes the beam to allow for high beam intensities greater than 106 particles/s to be used. The second Prototype CAT detector has a single drift region, which is used to track both the beam and scattered or emitted charged particles [80,78]. Later a modified design added a better optimized readout pad geometry and added two arrays of NaI detectors to opposite sides of the active-target region to measure the high-energy recoils that are not stopped in the gas [11]. The NaI array was later replaced with an array of Si detectors [81]. A drawing and photograph of the modified CAT detector is shown in Fig. 24. Forward or low-angle scattering in center-of-mass frame yields very low energy recoils of protons, deuterons, or α particles such that a gas target is needed to detect these recoils at small scattering angles. For example for the (d, d′ ) reaction with a 100 MeV/u beam of 132 Sn, the scattered d energy is 500 keV at a center-of-mass scattering angle of 1°. The CAT detector was developed to have an angular resolution of 1° (7.5 mrad in lab) and a total kinetic energy resolution of 10% (1 MeV in center of mass) for use with high-energy radioactive beams with light-ion targets [82,80]. For amplifying the electron ionization tracks, the CAT detector uses a 400 µm-thick GEM. Tests with pure deuterium gas at low pressure of 0.2 atm was investigated and found to give a gas gain of up to 103 . The use of pure gas is advantageous due to not having competing background reactions [83,84,82]. The GEMs were used with a triangular readout pad geometry with a higher granularity in the central beam region for higher sensitivity to scattering angles. Signals are digitized with Flash-ADC’s and read out to a data acquisition system [11]. The CAT detector can be surrounded by two NaI or Si arrays to detect scattered particles that are not stopped in the gas volume. The auxiliary array can be used to generate a trigger for the detector. Several test experiments were performed to develop and commission the prototype detectors at various accelerator facilities. At the 12UD Pelletron Tandem accelerator complex at the University of Tsukuba, the double-field cage Prototype CAT was used to measure position and angular resolution with an 4 He beam of 30 MeV. A He:CO2 (95:5) gas mixture with a pressure of 1 atm was used in the detector. With a triangular backgammon pad design, the angular and energy resolutions were found to be within the design requirements [80]. Experiments were performed the Heavy-Ion Medical Accelerator in Chiba (HIMAC) [85] using a 250-MeV/u beam of 56 Fe. A D2 : CO2 (95:5) gas mixture was used as the target gas at a pressure of 1 atm. The whole CAT system was tested including the double-GEM configuration, array of NaI detectors, trigger, and electronics [11]. Commissioning experiments were performed for the modified CAT active-target at HIMAC using two different reactions. A d(56 Fe, d′ ) reaction was used to test the 120 element NaI array [86] while a d(16 O, d) reaction was used to measure the gains of three layers of 100 µm-thick GEM foils with 1 atm of pure deuterium gas. The gain was measured to be 2000, which is sufficient for the measurement of ionization tracks in inelastic scattering reactions with the advantage of not having contaminant reactions that result from using a quench gas [11,83]. An in-flight radioactive beam of 14 O at 100 MeV/u was used for a d(14 O, d′ ) reaction performed [87]. Last, a measurement with 132 Xe at 100 MeV/u was performed with a triple thick-GEM configuration. Beam intensities of 103 –106 particles/s were used and a gain of over 5000 was measured [81]. 2.6. Micro-PIC based Active target for experiments in Inverse Kinematics at o degrees (MAIKo) The Research Center for Nuclear Physics (RCNP) of Osaka University in collaboration with Kyoto University developed an active target with a micro-pixel chamber (µ-PIC) [12] for the detection of low-energy recoil particles in inverse kinematics.

S. Beceiro-Novo et al. / Progress in Particle and Nuclear Physics 84 (2015) 124–165

143

Fig. 25. The left image shows a schematic view of MAIKo and the right one a detailed view of the structure of the micro-PICs. Source: Both from [12].

MAIKo was conceived to study the role that excess neutrons might play in α clustering phenomena and clustering correlations in nuclear physics. The ideal study case for the collaboration is 12 Be, which, according to generalized two-center cluster model, should show a monopole excitation of the ground state of 12 Be with 2α cores oscillate against each other [88]. The monopole strength of these collective modes should be larger than that of the single particle modes. Therefore, to clarify the scenario a systematic study of monopole strengths is necessary. The typical experiment to measure the strength would be to use inverse kinematics and the invariant mass method. However, due to the high multiplicity of reaction products in the case where the cluster states are located around particle decay thresholds, this is not a suitable method. Instead, one can use missing mass spectroscopy which measures the excitation energies by detecting only one recoil particle. This method is especially suited when using an active target detector such as MAIKo where one can easily detect the low-energy particles involved (see Section 1). MAIKo is based on a time projection chamber (TPC) providing a three-dimensional reconstruction of charged particle tracks and as many other active target detectors, uses the detection gas of the TPC as reaction target. One of the main reactions of interest for MAIKo is inelastic scattering with α particles so the typical gas they use is He with a small fraction of quench gas. The pressure of the gas can be chosen according to the energy loss desired conditions for each experiment and can range between 300 Torr to atmospheric pressure. For the amplification of the drifted electrons the (µ-PIC) was chosen [89]. The µ-PIC is a micro-pattern gaseous detector with anode pixels surrounded by cathode strips fabricated on a polyimide substrate with a high position resolution. The anode is normally biased to around 400 V and has an active area of 10 × 10 cm2 . The structure of the µ-PIC is drawn on the right panel of Fig. 25. The anode pixels are 50 µm thick spaced 400 µm away from each other. The anode signal provides one spacial cordinate for the particle track, and the cathode signal provides the second one. The drift time provides a measurement of the vertical position of the particle trajectory. The angular resolution achieved is better than 6 mrad. Si and CsI(Tl) detectors are installed at left, right and downstream sides of the TPC and are used to produce a trigger signal for the scattering event. The CsI crystals are also used to stop high-energy particles. The total volume of the TPC cage is 14 × 14 × 14 cm3 . The homogeneous electric field of about 200 V/cm is produced with an Al plate and a Ni grid connected to a high voltage. The voltage of the grid is adjusted to be transparent for the electrons drifted from the active volume but opaque for the positive ions produced in the avalanche on the surface of the µ-PIC. The µ-PIC are read out via 512 electronic channels: 256 anode strips and 256 cathode strips. The electronic signals collected by these strips are amplified and discriminated using EF2009bal chips originally developed for the compton camera experiment at Kyoto university [90]. The information is afterwards registered using VME memory modules. The first test MAIKo test experiment was performed at RCNP in 2013. The main goal was to study the dependence of the TPC properties on the beam rate and to acquire scattering events to develop a tracking algorithm. A 4 He beam at 12.5 MeV/u with an intensity of 1000 particles/s was used and the detector filled with isobutane. Fig. 26 shows the experimental setup for this test. Fig. 27 shows an example of 12 C(4 He, 4 He′ )12 C∗ inelastic scattering. In this event, 12 C was excited above the three α decay threshold (7.27 MeV) and decayed into three α particles. The incident and scattered α particles have a smaller energy loss within the detector than the α particles coming from the decay, and thus they leave a weaker signal (thinner line in Fig. 27). In these conditions, they achieved an angular resolution better than 2.8 mrad. 2.7. Active Target Time Projection Chamber (AT-TPC) The AT-TPC and its prototype were conceived to perform experiments with mainly low energy rare isotope beams. The physics themes covered by this detector include studies of resonant states that can exhibit exotic clustering properties,

144

S. Beceiro-Novo et al. / Progress in Particle and Nuclear Physics 84 (2015) 124–165

Y(mm)

Y(mm)

Fig. 26. Schematic view of the setup used for the first test experiment with MAIKo. Source: From [12].

140 120

140 120

100

100

80

80

60

60

40

40

20

20

0

0

20

40

60

80

100

0

0

Z(mm) (a) Anode.

20

40

60

80

100

X(mm) (b) Cathode.

Fig. 27. Example triple α event coming from 12 C(4 He, 4 He′ )12 C∗ inelastic scattering from carbon. Source: From [12].

transfer reactions with light particles to probe the single-particle structure of nuclei away from the valley of stability, low energy reactions of astrophysical interest, fusion and breakup studies, and at higher beam energy the study of giant resonances and heavy ion reactions to explore the nuclear Equation Of State. In this detector, the concept of gaseous active target is taken to the extreme, as there are no ancillary detectors added to the Time Projection Chamber. All reaction products emitted in the gas are solely detected from their ionization of the gas, and subsequent recording of their tracks. This means the tracks from beam particles that enter the gas volume but do not result in a reaction are also potentially recorded. Since they are of no interest, a trigger condition needs to be implemented that rejects such events. This is achieved by using dedicated electronics from the GET collaboration (General Electronics for TPC) [54], in which a specially designed Application Specific Integrated Circuit (ASIC) implements not only the amplification and storage of the charge collected on the pads, but also individual discriminators signals that are summed to form a multiplicity (see Section 3.1). By means of this electronics, an internal trigger can be generated that fulfills particular conditions, such as the exclusion of a pad region covered by unreacted beam tracks only (see Fig. 32 in the following). A half-scale prototype (see Table 1 for dimensions) of the detector has been constructed prior to the full-scale, in order to assess the feasibility and look for performance issues of the AT-TPC [91,13]. Several experiments have been performed with this prototype, including resonant scattering to study α -cluster states in 10 Be [92] and 14 C [93], as well as measurements of fusion cross sections around the Coulomb barrier. The first experiment performed with the prototype AT-TPC is the study of 6 He + 4 He resonant scattering where differential cross sections could be measured over a wide angular and energy range in a single experiment [92]. The results are shown in Fig. 28 where angular distributions are displayed for various energy bins from 2.7 to 5.8 MeV in the center-of-mass system. The data were obtained after 4 days of experiment with a beam rate of only about 1500 particles per second. Comparison with Coupled Reaction Channel (CRC) calculations show a clear ℓ = 4 signature for this 10 Be resonance located at 9.98(15) MeV. The partial α decay width can be extracted from the angleintegrated cross section to characterize the nature of this state. The large value obtained indicate a well-developed α cluster structure that can be interpreted as a molecular structure where the two valence neutrons are delocalized over the two α cores [94,95]. In addition, the energy of this observed 4+ state nicely fits the J (J + 1) dependence of a rotational band with + + that of the known 0+ 2 state and 2 state at 7.542(1) MeV. However, no significant 4 resonance strength was found in this energy domain of band crossing, illustrated in Fig. 29. This experiment could set an upper limit of decay strength of other 4+ state in this energy domain. The result supports the limits of clustering in 10 Be due to the spin degree-of-freedom, and calls for more detailed spectroscopy of individual cluster states in 10 Be and related microscopic theoretical studies. Another particularly well suited type of experiment that can easily be performed with this detector is proton inelastic scattering below the Coulomb barrier, where resonant states such as the Isobaric Analog State (IAS) can be excited. Thanks

S. Beceiro-Novo et al. / Progress in Particle and Nuclear Physics 84 (2015) 124–165

102

145

a

b

c

d

e

f

g

h

10

1

102

10

1

102

10

1

102

10 50 1 0 60 80 100120 40

60

80

100

120

140

40

60

80

100

120

140

Fig. 28. Differential cross section of the 6 He + 4 He elastic scattering for different energy bins, as measured by the prototype AT-TPC (from [92]). The blue area correspond to the systematic errors from ambiguities in the beam angle. The results of CRC calculations (solid lines) are compared to the data. Only at 2.7 MeV a strong difference is observed between CRC calculation and the experiment. The inset displays the data at 2.7 MeV (full symbols) and 3.3 MeV (open symbols) on a linear scale. The angles where the Legendre polynomials PL (cos θc .m. ) for L = 4 and 6 become zero are denoted by the solid and dashed lines, respectively. This result could be interpreted as a signature of a 4+ resonance at this energy of 2.7 MeV. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

to the slowing down of the beam inside the gas volume, the excitation functions can be measured at once without the need to tune the incoming beam energy. The widths and cross sections of these states can be analyzed to extract spectroscopic information on the ground and excited states of the parent nucleus, i.e. of the nucleus A+1 Z, or the equivalent of a (d, p) reaction but with much larger cross sections. Like transfer reactions using the lightest nuclei, this type of measurement has been used extensively in the past on stable targets in direct kinematics [102]. The location of the IAS provides a direct measurement of the Coulomb displacement energy, and the cross sections are used to deduce spectroscopic factors in the conjugate neutron-rich nucleus. In inverse kinematics in the AT-TPC, the scattered protons are detected around 90° and the vertex of the reaction can be reconstructed for each event. The method is very efficient, because the excitation function can be obtained at once as the beam slows down in the gas volume. A test was conducted using the prototype AT-TPC with a 11.8 MeV/u 124 Sn beam at the ATLAS facility [103], in preparation for a future experiment using a radioactive 132 Sn beam. The preliminary results are shown in Fig. 30, where the previous direct kinematics measurement is superimposed for comparison. This demonstrates the feasibility of this technique using radioactive nuclei where making targets of the nuclei of interest is not possible. Note that in this experiment a large number of δ -electrons are produced and can be a major problem to separate the beam tracks from the detection of the recoil protons. This problem can be solved by introducing the active target in a magnetic field that confines the δ -electrons.

146

S. Beceiro-Novo et al. / Progress in Particle and Nuclear Physics 84 (2015) 124–165

Ex in 10Be [MeV]

15

10

5

0

20

10

0

J(J+1)

dσ/dΩ(mb/sr)

Fig. 29. Plot of Ex versus J (J + 1) for 10 Be. The band members of the ground and the second 0+ states are shown by the circles and squares, respectively. The linear extrapolation using the 0+ and 2+ states is shown for each band. The horizontal lines at J = 4 denote predicted level energies of the 4+ member of the ground state band from the β –γ constraint AMD method [94] (solid line), the variational AMD method [96] (dashed line), the four-body cluster model [97] (dotted line), the molecular orbital model [98] (dot-dashed line), the semi-microscopic algebraic cluster model [99], (double-dot-dashed line), and the multi-cluster generator coordinate method [100] (triple-dot-dashed line). The data of Refs. [98,97] were obtained from the calculated values with respect to the threshold energy of 2α + 2n at 8.386 MeV. The shaded area denotes the energy domain covered in the study of [92]. No significant 4+ resonance strength was found in this energy domain.

60 50 40 30 20 10 0

10

10.2

10.4

10.6

10.8

11

11.2

11.4

energy(MeV) Fig. 30. A preliminary excitation function is shown of the reaction 124 Sn(p, p) in inverse kinematics measured with the prototype AT-TPC. The 11.8 MeV/u Sn beam was produced at the ATLAS facility. The range of energies covered by the slowing 124 Sn projectiles was from 9 to 12 MeV/u in the center-of-mass frame. The preliminary cross section data was scaled to match the data from a direct kinematics measurement [101]. The ℓ = 3 resonance peak can be observed at 10.6 MeV, which corresponds to an excitation energy of 2.705 MeV in 125 Sn. 124

A schematic view of the AT-TPC is shown in Fig. 31. The detector is cylindrical in shape, the active volume being 55 cm diameter by 1 m long, for a total volume of about 250 l. The chamber has been designed to fit inside a large bore solenoid magnet that can apply a longitudinal field up to 2 T. The purpose of the magnetic field is multifold: (i) the curvature of the tracks induced by the field are a direct measurement of the magnetic rigidity of the charged particles, from which their momentum can be deduced, (ii) the bending of the trajectories inside the active gas volume means longer tracks can be detected, extending the range in which particles can be fully stopped in the gas, and (iii) the primary electrons emitted by ionization of the gas are focused during their drift to the electron amplification device, giving a better localization of the tracks in the Time Projection Chamber. The electric field needed to drift the primary electrons to the pad plane is shaped uniformly by a succession of 50 stainless rings held at fixed potentials by a resistor chain. The electron amplification device used on the pad plane is of the Micromegas type [52]. The geometry of the pad plane is optimized to give a larger granularity at the center, where small angle scattering events are tracked. The pads are equilateral triangles in shape with sides close to 1 cm, and those in the hexagonal central region are half in size to ensure consistent tiling of the pad plane surface. The total number of pads is 10,240 (see Fig. 32). In addition, a tilting angle of 7° can be applied to the detector (as shown in Fig. 31) in order to spread the image of small angle scattering tracks over a large number of pads. This feature has the added advantage to spread the images of incoming beam tracks over many pads rather than concentrating them in the center pads, and make

S. Beceiro-Novo et al. / Progress in Particle and Nuclear Physics 84 (2015) 124–165

147

Fig. 31. Conceptual view of the AT-TPC. The detector chamber houses the cylindrical active volume Time Projection Chamber where beam particles interact with the gas target and tracks are recorded. The chamber is placed inside a large bore solenoid magnet which applies a longitudinal magnetic field up to 2 T. The detector is shown in its tilted configuration of 7°, to allow better detection of small angle scattering as well as better pile-up rejection.

Amplitude [channels]

4000 3500 3000 2500 2000 1500 1000 500 0

0

100

200

300

400

500

600

Time buckets

Fig. 32. Typical event recorded by the AT-TPC. This event corresponds to elastic scattering of an α particle on 4 He gas. The inset (a) shows the time projection of the signals integrated over all pads, while the inset (b) shows a 3-dimensional rendition of the event, where the vertex location is clearly visible as well as the tracks before and after the reaction. Inset (c) shows the hit pattern on top of the triangular pad geometry.

pile-up rejection of unreacted beam particles a lot easier than in the non-tilted configuration. Preliminary tests indicate a gain of roughly a factor of 500 in the pile-up rejection.

148

S. Beceiro-Novo et al. / Progress in Particle and Nuclear Physics 84 (2015) 124–165

Fig. 33. Schematic view of TACTIC with detail on the GEM field lines. Source: From [105].

Commissioning of the full-scale AT-TPC was performed recently using a 4 He beam at 3 MeV/u provided by the ReA3 linac in operation at the NSCL [104]. The detector was set in its tilted position and the magnetic field inside the solenoid was set to 1 T. The target gas was a mixture of 90% 4 He and 10% CO2 with a pressure of 100 Torr. Several types of events were observed, with a clear distinction between scattering on 4 He nuclei and 12 C or 16 O nuclei. Fig. 32 shows a typical event of 4 He + 4 He scattering, where the two tracks of the 4 He are mirroring each other and the vertex of the reaction is clearly visible. Although this data is still under analysis, the curvature of the tracks induced by the magnetic field will greatly improve the energy resolution compared to a simple measurement of the ranges, as was used in the prototype. The magnetic field allows to fully characterize energetic particles that leave the gas volume.

2.8. The TRIUMF Annular Chamber for the Tracking and Identification of Charged particles (TACTIC) TACTIC [105] is an active target time-projection chamber that was designed to study reactions of astrophysical interest, in particular reactions important in the nucleosynthesis r-process. Network calculations, show that, for some models, two reactions chains: α(α n, γ )9 Be(n, γ )10 Be(α, γ )14 C and α(t, γ )7 Li(n, γ )8 Li(α, n)11 B are of importance to calculate the final abundances of r-process nuclei [106]. The TACTIC collaboration focused on the 8 Li(α, n)11 B reaction, which is experimentally challenging because of the very low energy of the 11 B recoil. The reaction had been measured before by the MSTPC collaboration [107] with beam intensity of around 2000 particles/s. Consequently, the TACTIC detector was designed to bypass this limitation and take full advantage of the beam intensities available at the ISAC radioactive beam facility at TRIUMF. TACTIC was optimized for the detection of charged particles with energies varying from a few MeV down to 100 keV. The beam region was designed to be excluded from the active region, allowing for a high beam intensity without saturating the electronics. TACTIC has a high solid angle coverage and the ability to run at different gas pressures and incident beam energies. In the original design, TACTIC was to be surrounded by a gamma array, hence the attenuation gammas rays minimized. Fig. 33 shows a schematic view of the final design of TACTIC. A cylindrical shape with 10 cm diameter was chosen. The beam pipe is separated from the gas volume with a 2 cm diameter thin window. The beam enters the detector’s active volume towards a central cathode region aligned with the optical axis of the beam which is surrounded by cathode wires running parallel to the beam axis and held at a negative voltage with respect to the rest of the detector. This is the target region which does not have active tracking. The region outside the wire tube is the active one and is also cylindrical with a diameter of about 10 cm. The electron amplification system chosen was (GEM) [63,108]. The GEM is located parallel to the beam and is read out via a cylinder of anode strips. It is segmented azimuthally to improve track reconstruction resolution and to be able to accept high beam intensities without saturating the electronic channels. The reaction products enter the active gas area, and the ionized electrons coming from the energy loss in the detector are drifted out by the radial field to the GEM, where it is amplified and the charge is transferred to the anodes. The detector allows for the collection of charge and timing information that can be used to calculate the angle of the track, the particle type and its energy. With tracking reconstruction the vertex of the reaction and center of mass energy can be determined. The data acquisition system is based on VME flash ADCs. The first reactions to be measured with TACTIC were 8 Li(α, n)11 B and 7 Be + p elastic scattering both using radioactive beams in inverse kinematics. This version of TACTIC was considered as prototype, and the experience gained with it will be used for a future realization [109].

S. Beceiro-Novo et al. / Progress in Particle and Nuclear Physics 84 (2015) 124–165

149

Fig. 34. Photo of the ANASEN detector with the external chamber slid out showing the internal components. Source: From [111].

Fig. 35. Solidworks model of the detector and photograph of the inner components of ANASEN. Source: From [16].

2.9. The Array for Nuclear Astrophysics and Structure with Exotic Nuclei (ANASEN) ANASEN [110,111] is a multi component detector designed to study charged particle reactions such as (p, p), (α , α ) and (α , p). The main goal of the collaboration between Florida State University and Louisiana State University was to get a better understanding of the structure of light isotopes and astrophysically-relevant reactions. The main features of ANASEN are similar to other active targets: large solid angle coverage, ability to measure a whole energy range with a single beam energy, and ability to measure the reaction vertex. A photograph of ANASEN with its internal components is shown in Fig. 34. The main design goal of ANASEN is the use as active gas target, but it can also be operated with thin, solid targets. As in all other active target detectors, the gas used for target is also used to read the signals, in this case with position sensitive proportional counters (PSPC). The beam line is separated from the gas chamber with a Kapton foil of 7.5 µm. This PSPC (see Fig. 35) consists of 19 carbon fiber anode wires evenly spaced around the beam axis and connected to preamplifiers at both ends. The charge division signal on the wires is used to measure the position. The sum of the total collected charge gives information about the charge that the particle produced in the gas. The trigger signal is originated in the Si-detectors, and the wire signals energy loss information is used for particle identification. ANASEN also contains a silicon (Si) detector array with two different configurations: two rings with a total of 24 double-sided position-sensitive Si detectors (DSPSSD) surrounding the beam axis with a total of 12 output channels; and a second array installed in the forward direction segmented along the beam axis into four strips with 8 electronic channels. The front silicons contain a resistive layer to measure relative position within the detector. They are segmented into 4 double sided quadrants, where the front side is segmented into rings and the back side into radial slices allowing for the reconstruction of scattering angles. The back of the detector is divided perpendicularly to the beam axis into four segments with one output channel each. There are 24 detectors in the Si barrel array and 16 in the forward direction. The purpose of these detectors is to determine the

150

S. Beceiro-Novo et al. / Progress in Particle and Nuclear Physics 84 (2015) 124–165

500

400

300

200

100

0

5

5.5

6

6.5

7

7.5

8

8.5

9

9.5

10

18F Excitation energy (MeV) Fig. 36. Excitation function for the system 17 O (p, α ) 14 N measured with ANASEN at the RESOLUT facility. Source: From [16].

energy of any particle that punches through the Si detectors. The final detector system will also be surrounded by cesium iodide (CsI) detectors. ANASEN uses Application Specific Integrated Circuit (ASIC) chips [112] designed and developed by Southern Illinois University Edwardsville and Washington University and preamplifiers built by Louisiana State University to be able to handle the hundreds of electronic channels that they need. The ASIC chips provide shaped energy signals and timing output for up to 16 channels. The chips are mounted in pairs on 16 boards (total of 32 chips and 512 channels) that can be simultaneously used. ANASEN has been used in the reaccelerator facility ReA3 [104] at the National Superconducting Cyclotron Laboratory and the RESOLUT facility at FSU [113]. The commissioning, and first experiments with ANASEN were performed at FSU. The first tests were performed using a solid target configuration. A solid polyethylene (CH2 )n target was installed to measure the 17 O(p, α)14 N reaction [16] with a 17 O beam produced at the tandem accelerator of FSU at four different beam energies between Ecm = 1.8–3.0 MeV. All the different reaction channels were clearly separated by kinematics with a charge resolution better than 5%. Even though the elastic scattering component was 10 times stronger than the inelastic, both components were resolved. Fig. 36 shows the excitation function for the 14 N(α, p)17 O. With a 17 F beam and a solid target (polypropylene) the 1 H(17 F, p)17 F and 1 H(17 F, α)14 O reactions were studied, being able to calculate the center of mass energy for each event from the measured angle, the energy of the light particle in the silicon telescope, and to reconstruct an excitation function. An experiment was performed using ANASEN in active target mode, with a 6 He beam with energies ranging from 7 to 29 MeV impinging the detector filled up with a mixture of He:CO2 (95:5) with pressures ranging from 160 to 700 Torr depending on the energy. With this excitation functions were able to be reconstructed for the 6 He–4 He and 6 He–4 He scattering systems [111]. The results are in agreement with those previously published by the AT-TPC collaboration [92]. In ReA3 ANASEN used the first reaccelerated beam to extract the excitation function for 37 K + p elastic scattering in solid target mode. Given the success of the first experiments with ANASEN, a second upgrade of the system is being performed. The upgraded detector will have 5 additional wires in the proportional counter for a total of 24 to improve the angular resolution. Another important upgrade in the electronic system will be the use of 500 MHz flash ADCs to improve the time resolution and broaden the dynamic range [114]. 2.10. MagIc Numbers Off Stability (MINOS) The MINOS detector [17] uses the concept of active target to mitigate a very specific difficulty linked to the technique of in-beam γ -ray spectroscopy. In this technique radioactive beams are used at velocities a large fraction of the speed of light (β ranging from 0.2 to 0.8) which induce a strong Lorentz boost on the γ -rays emitted by the reaction products in flight [115]. In order to recover a workable γ -ray resolution, this Doppler effect has to be corrected. The correction depends strongly on the emission angle of the γ -ray, but also on the velocity at which the reaction occurred. This last dependency strongly constrains the target thickness that can be used in this type of experiment, because the energy at which the reaction occurs – and the γ -rays are emitted – cannot be measured when using standard solid targets. The concept of the MINOS vertex tracker resolves this issue via two components: a liquid hydrogen target with a large thickness (15 cm), and an annular Time Projection Chamber surrounding the target vessel used to measure the vertex of each individual reaction, and therefore the energy at which it took place. The choice of liquid hydrogen as target material is primary driven by the need to have a sizable length that is compatible with the spatial resolution of the TPC, as well as a reaction that produces a recoil (in this case a proton) that can easily escape the target material and leave a track in the TPC. Besides this, the choice of H2 provides the higher possible number of target nuclei for a given energy loss. The principle of operation is pictured in Fig. 37. The incoming

S. Beceiro-Novo et al. / Progress in Particle and Nuclear Physics 84 (2015) 124–165

151

Fig. 37. Principle scheme of the MINOS device (from [17]). The recoil protons that escape the target can be detected in the TPC and used to track the vertex location and hence the velocity at the time of the reaction.

80

30 25

Counts / 3 KeV

counts / 20 KeV

60 20 15 10

40

20 5 0

0 200

400

600

800 1000 1200

E (keV)

400

450

500

550

E (keV)

Fig. 38. Simulations of the gains in luminosity and resolution (from [17]). Both example compare a (p, 2p) reaction on the H2 target and a proton knockout reaction on a 9 Be target. The left panel shows a clear luminosity gain when coupled to the NaI array DALI2 [118], but not much improvement in the resolution. The right panel shows the gains in both luminosity and resolution when coupled to a high resolution Ge array such as AGATA [119].

beam particles react via a (p, 2p) or (p, pn) reaction in the LH2 target at a certain depth corresponding to a certain velocity β . If at least one proton escapes the target and leaves a track in the surrounding TPC, the vertex of the reaction can be reconstructed. Simulations indicate a resolution between 2.6 and 5.8 mm depending on the type of reaction and kinematics model used. The efficiency is estimated at around 80%–90%. One of the main physics themes that is addressed using in-beam γ -ray spectroscopy is shell evolution away from the valley of stability. The principal reason stems from the high luminosity of this technique which extends its reach to the most exotic isotopes. The overarching question is the evolution – if any – of magic numbers. It is becoming clear that the magic number sequence observed in the valley of stability does not hold in particular on the neutron-rich side. The effects behind this evolution are related to changes in the proton–neutron balance of the nucleus, that modify the energies of single-particle levels. This evolution can close the gaps that give rise to the magic numbers, but also create gaps that induce new magic numbers. The components of the nuclear force at play such as the tensor force [116] for instance, are usually adjusted in shell model calculations [117] to reproduce the observables measured in these experiments. The promise of the MINOS device is to help reach isotopes that are the most difficult to produce, by increasing the target thickness significantly while retaining a good resolution after the Doppler correction of the γ -ray energies. The simulation presented in Fig. 38 illustrates the gains in luminosity and resolution that will be achieved. The most beneficial configuration is when coupled to a high resolution γ -ray Ge array such as AGATA [119] or GRETA [120] for instance. Besides in-beam gamma spectroscopy, MINOS has been successfully used to study (p, 2p) and (p, pn) missing mass and invariant mass measurements. A schematic view of the target and TPC is shown in Fig. 39. The ensemble is designed to fit inside a γ -ray detector array, here the DALI2 NaI array of RIBF [118]. Due to the relatively poor intrinsic resolution (about 7%) or the NaI γ -ray detectors, the resolution increase gained from the measurement of the reaction velocity is not dramatic, however MINOS enables the use of a much thicker target (150 mm of LH2 as compared to 10 mm of 9 Be), which boosts the luminosity of the experiment by roughly an order of magnitude. When used with higher resolution γ -ray detector array where tracking is possible, the gain in resolution will add even more to the sensitivity.

152

S. Beceiro-Novo et al. / Progress in Particle and Nuclear Physics 84 (2015) 124–165

Fig. 39. Schematic drawing of the Minos target and TPC. The cryogenic target is surrounded by the TPC, which extends to forward angles to maximize the solid angle for detecting the scattered protons. The drift field to collect the primary electrons produced in the gas is parallel to the beam axis, and the electron amplification device used is of the Micromegas type [52].

Fig. 40. Beta-delayed multi-particle decays recorded with the optical time projection chamber described in [121]. The left panel shows β -delayed threeproton emission from 45 Fe [122] recorded so the incoming track is not visible, the right panel the track of an 8 He ion entering from the right that after β decay breaks up into a triton (long weak track), an α particle and an invisible neutron [121].

2.11. Optical Time Projection Chambers (O-TPC) A quite different technology has been adopted in what is called optical TPC’s. In nuclear physics they were first used to observe rare decays, such as the 2 and 3 proton decay of 45 Fe illustrated in Fig. 40. This type of use is not strictly the subject of the present paper. We include it here because it is a quite evident extension to use Active Target TPC’s for their imaging property of an implantation decay process. Recently an optical TPC in the active target mode was used for the observation of γ induced reactions [123]. The device is described in [18]. The layout of the detector is shown in Fig. 41. The 2-dimensional optical image with a high granularity (2048 × 2048) commercial CCD camera is correlated to a drift time recording by photomultipliers to give the third dimension. The event number is limited to the image acquisition of the CCD camera, typically of the order of 10–50 frames/s. This speed is sufficient for many cases if the trigger is carefully tuned to limit to the events of interest. The γ excitation to the second 2+ in 12 C that is interpreted as a rotational state built on the Hoyle [125] 0+ at 7.654 MeV was measured. A typical image recorded by the CCD camera of three alpha particles from the reaction 12 C(γ , α 0 )8 Be (→ α + α ) is shown in Fig. 42. The contribution from the reaction 16 O(γ , α 0 )12 C could be clearly separated form the reaction of interest. The angular distributions obtained over 180° in the center-of-mass system allowed to separate E2 and E1 contributions, as shown in Fig. 43. This experiment resulted in the unambiguous identification of the second 2+ state. Transition probabilities B(E2) were obtained and are a strong constraint for theoretical models. 3. Electronics 3.1. Amplifiers and digitizers Active target detectors equipped with TPCs have most of the time a large number of channels. To provide a precise image of the tracks in the detector, a good granularity is needed. The precision of track determination should be limited only or mainly by straggling phenomena. The ionization image of the tracks is broadened by lateral and longitudinal straggling of the primary electrons and their statistics. The tracks themselves are blurred by the angular straggling of the particles in the gas. The specific energy loss dE /dx is modulated by energy loss straggling. If the particles are stopped in the detector, the sum of the energy loss straggling results in a range straggling. The overall process is quite complex, depending on the gas,

S. Beceiro-Novo et al. / Progress in Particle and Nuclear Physics 84 (2015) 124–165

153

Fig. 41. A schematic diagram of the Optical Readout Time Projection Chamber (O-TPC) at TUNL with an 16 O target nucleus dissociated by the γ -ray beam from the HIγ S (High Intensity Gamma Source, [124]) facility. The gas mixture and pressure as well as the operating voltages are indicated. The drift electrons that provide the third (time projection) upward dimension are shown schematically [18].

Fig. 42. A typical image recorded by the CCD camera of three alpha particles from the reaction 12 C(γ , α 0)8 Be (→ α + α ) [123].

the electric and magnetic fields, the particle, eventual contaminations, temperature, etc. Recent overviews can be found in Refs. [126–129]. A comprehensive compilation of detector gas properties can be found in [130]. Some simple guidelines may help in these considerations: light gases such as H2 and He have lower straggling as compared to heavier gases such as Ar or Kr. The straggling is a statistical process, hence proportional to the square root of underlying processes: at normalized field/pressure values the lateral and longitudinal straggling for the drift of electrons is proportional to the square root of the drift distance. Typical values of the coefficient are of√ the order of 0.2–0.5 mm. The precision of the final value will depend on the number of primary electrons, and improve as N where N is the number of primary electrons. These considerations show that for active targets in the context of nuclear structure studies there are some favorable characteristics as compared to high energy TPCs. This is closely related to the reactions of interest and their kinematics: principally light target/detector gases are of interest, and the recoil energies of the light particles are low, implying high specific energy losses and therefore a high numbers of primary electrons. For example, 1 MeV protons or α particles in He gas at 1 atm produce ∼20 and 150 primary electrons per mm respectively, about 2 orders of magnitude more than at relativistic energies. This provides high electron statistics that should allow one to determine local characteristics with much better precision as compared to high energy. However, the range may be much shorter, and therefore one should have anode patterns with higher density than at high energy. Therefore the density of electronics per unit area will be high in active target detectors in this domain of nuclear physics. In the AT-TPC (see Section 2.7), the inner part of the pad plane is composed of equilateral triangles with side 5 mm, 11 mm2 area, or ∼100,000/m2 . This high density is possible thanks to the high density GET [General Electronics for TPCs] electronics [54], developed by an international collaboration. In a TPC, each electronic channel must keep information in memory over a time span of at least the length of the maximum drift-time of the electrons of the ionization tracks. This is achieved by using circular memories as illustrated in Fig. 44. In this context, the sampling bin with a given clock cycle is often called a time-bucket. The total time to be ‘‘remembered’’ and the time binning, the time interval per time bucket, determine the depth of the circular memory. To adapt to drift times that may vary largely and to different detector sizes, the GET electronics have a writing clock frequency that can be chosen between 1 and 100 MHz, and a memory depth of 512 time buckets [54]. Two techniques are currently in use: one fast ADC per channel (see for example [131], or a SCA (Switch Capacitor Array) (see for example [54]) with multiplexed ADC readout. Present multiplexed ADCs have typically 12 bits for full scale amplitude, corresponding to 4096 channels, but are available up to 14 bits. In the SCA, the sampled amplitude of a time

154

S. Beceiro-Novo et al. / Progress in Particle and Nuclear Physics 84 (2015) 124–165

8

6 4

2 0

120

80

40

0

8.0

9.0

10.0

11.0

12.0

13.0

14.0

Fig. 43. (a) The measured E1 and E2 cross sections of the 12 C(γ , α 0)8 Be reaction as a function of the γ energy. (b) The measured E1–E2 relative phase angle Φ12 together with the phase angle calculated from a two-resonance model [123].

Fig. 44. Illustration of the TPC basic electronics: as shown on panel (a) a circular memory is filled with a given writing clock frequency; panel (b) illustrates the read-writing sequence: the circular memory is filled continuously until a validated trigger initiates a write stop and a read. (c) a typical pulse as registered in an electronic channel. Source: Panels a, b courtesy of P. Baron, IRFU.

S. Beceiro-Novo et al. / Progress in Particle and Nuclear Physics 84 (2015) 124–165

155

Fig. 45. Example of baseline noise measurement with the micromegas detector plane and the GET electronics of the AT-TPC. The root mean square noise is plotted as a function of the connection length between the front end electronics and the pads, in units of equivalent electron charges for the highest gain of the preamplifiers, 120 fC. The pads have triangular shape with base 0.5 cm the small ones and 1 cm the big ones.

bucket is stored on a capacitor. Storage times without deterioration can be of the order of ms, giving enough time for a multiplexed readout. Due to price and space considerations, the SCA solution becomes the most preferred as the number of channels increases. Present high density ASIC technologies (Application Specified Integrated Circuit, note that recent ASIC’s such as the one developed for the GET electronics contains on the order of 1 million components) can produce preamplifiers with a noise level of 200–300 electron equivalent rms (root mean square) intrinsically, or ∼700–1000 electrons taking in to account the connections and pad capacitance. In order to have signals that are ∼5 standard deviations above noise, a minimum signal of ∼4000 electrons will be necessary. Compared to the typical energy loss of low energy protons, this means the number of primary electrons must be multiplied by at least a factor 200. This gain is low as compared to high energy devices, where a gain a factor ∼100 higher is usually needed, corresponding to an overall gain of 104 or greater. This lower gain requirement helps to overcome the limitations due to the gas composition imposed by the active target function. An example of noise measurement is shown in Fig. 45 for the AT-TPC. The noise observed for electronic channels that are purposely not connected, called fixed pattern noise, is only about half of the values seen on this figure. The noise is shown for small pads (11 mm2 ) and large pads (44 mm2 ) as a function of the trace length carrying the signal from the pads to the input stages of the electronics. The mild correlation indicates that the contribution of the trace length to the noise level is only marginal. The relative noise level decreases of course with lower preamplifier gains, that can be as low as 1 pC full scale. The energy loss in typical active target devices can vary by very large factors, depending on the energy and the type of particles that are at the origin of the track. An extreme situation was encountered with the reaction p + 124 Sn that was studied as test case for a future experiment with 132 Sn with the prototype AT-TPC (see Section 2.7). The specific energy loss is about a factor 100 larger for the 124 Sn than for the protons. This difficulty can be mitigated by reducing the electron amplification gains of the pads collecting the electrons from the 124 Sn tracks, by polarizing them individually [91], and/or by using a lower gain for the preamplifiers. These considerations were taken into account in the design of the GET electronics, where the preamplifier gains can be adjusted for each channel individually between 120 fC and 1 pC. This feature does not exist in the electronics used at high energy. A different method was used in the Maya detector to decrease the gain in the beam region: an electrostatic mask around the beam was polarized negatively to reduce the transmission of primary electrons in this region [132]. 3.2. Trigger A difficult trigger problem arises in active target detectors used for nuclear structure studies in inverse kinematics. The beam particles are very often more ionizing than the reaction products, and the latter may stop inside the active volume of the detector. Therefore in many or most cases it is not possible to rely on external detector devices to provide a trigger signal. The trigger signal, that determines whether or not the event will be transferred to the data acquisition, should be produced when, and only when, a reaction of interest occurs, without losing part of the phase space of the reaction. How to best produce this trigger signal depends greatly on the detector, its geometry as well as the specific experiment. In the Maya detector for instance (see Section 2.2), the trigger was in many cases taken from wires that were out of the beam region: these wires are parallel to the beam, and at ∼1 cm from the beam, only reaction products will produce a significant signal on them. In the prototype AT-TPC [13] the signal from the mesh of the micromegas was used in some experiments using the difference of time structure between beam events (no reaction) and reaction events. The time

156

S. Beceiro-Novo et al. / Progress in Particle and Nuclear Physics 84 (2015) 124–165

between the begin and the end of the signal, determined by standard electronics, can be used as an anti-coincidence for times corresponding to beam events. In other experiments pads near but outside the beam region were connected to a decision unit outside the specific TPC electronics. In binary reactions, the coincidence between opposite sectors of a circular pad geometry can be used. In the GET electronics, specifically designed for TPCs with this type of application, two levels of trigger generation were devised to produce a trigger signal internally. (1) All electronics channels include a leading edge discriminator, that can be enabled and adjusted for each channel individually. The outputs of these discriminators are summed within a programmable time window sliding with the frequency of the digitizing clock during the acquiring (writing) phase of the analog memories. This sum signal forms a real time multiplicity that can be used to generate a level 1 trigger whenever it exceeds a given threshold. (2) Once a level 1 trigger is generated, the writing of the analog memories is interrupted and the hit pattern information generated by the discriminators is transmitted to the main logic unit of the data acquisition. This unit can then compare the hit pattern to a truth table to generate a level 2 trigger. This operation can be easily executed within 1 µs by the on-board FPGA, in order to limit the dead time. More complex trigger generation algorithms can also be programmed to a processing unit, at the expense of a longer dead time. The main purpose of this scheme is to effectively reject triggers that would be generated by beam particles that do not undergo a nuclear reaction with the active medium. In addition, it can be used in more complex ways to also reject uninteresting reaction channels (such as elastic scattering for instance) that have large cross sections and would otherwise overwhelm the data acquisition. Rejection of the tracks left by unreacted beam particles can also be realized in a more direct way. Tubes around the beam path (see for instance [131]) or grids [133] may be used to render the detector essentially blind to the beam. The main drawbacks are that short tracks will not be detected, and precision is lost in the determination of the reaction vertex. However, this technique enables the measurement of reactions with low cross-sections, as for example reactions of astrophysical interest, with beam intensities that are of the order of 105 –107 particles/s. 4. Data acquisition Data acquisition systems for Time Projection Chambers are notoriously complex. This complexity arises from the large number of channels, combined with the necessity to digitize the signals of each individual pad. A quick estimate illustrates the issue: for a 10,000 channel detector recording 500 samples (or time buckets) at a rate of 1000 triggers per second, the amount of data throughput is 60 Gigabits per second, assuming each datum is coded on 12 bits. Clearly some kind of data preprocessing and reduction at the front-end level is necessary to attain a more reasonable throughput that can be processed and stored by the back end. Another issue is related to the cost-effectiveness of the electronics used with large number of channels. Rather than using costly digitizers and digital memories for each channel, TPC electronics typically use the analog memory scheme provided by Switched Capacitor Arrays (SCA) followed by a limited number of sampling ADCs where high channel density can be achieved at a much reduced cost. This solution limits the data throughput because at the time of readout the charge stored in the SCA must be converted serially through a limited number of digitizers. Solutions where only the channels and/or time buckets containing valid data are digitized can greatly improve the performance of the data acquisition. These type of solutions are the only viable ones for large TPCs such as those found in particle physics detectors like the STAR experiment at RHIC [134] for instance, where the number of channels is in excess of 100,000. Although the types of nuclear physics experiments discussed in this article are not of the same scale, the number of channels they involve is only an order of magnitude or so less. To follow an example already cited in this article, the architecture of the GET electronics [54] used for acquiring data from the AT-TPC, ACTAR and several other types of detectors is shown in Fig. 46. Although this type of architecture is common among TPC data acquisition systems, such as the T2K system [135] or STAR TPC system [136], the particularity of the GET system lies in the possibility to generate a trigger internally without the help of an ancillary detector. As mentioned in Section 3.2, this is achieved through a running multiplicity signal generated from the discriminators embedded in the AGET ASIC chips that readout the pads of the TPC. This multiplicity signal is ultimately routed to the trigger module together with hit pattern information, from which the logic defined by the user can generate a trigger. Upon trigger generation, the circular buffers are stopped and read. 4.1. Front end modules The large number of channels necessary to readout TPC pad planes puts severe constrains on the design of the frontend modules that contain the charge preamplifier and shaping amplifier stages. It is of course highly desirable to locate at least the charge preamplifier stage as close as possible to the pad plane, so as to reduce the capacitive noise induced by the connection (wire of trace) between each pad and the input of its preamplifier. This on the other hand implies a large number of analog connections between the detector site and the digitizers. To mitigate this issue, some processing is therefore necessary at the level of the front-end electronics to reduce the amount of physical connections to the back-end

S. Beceiro-Novo et al. / Progress in Particle and Nuclear Physics 84 (2015) 124–165

157

Fig. 46. Schematic view of the GET electronics architecture [54]. The various components are the front-end boards (AsAd) equipped with the AGET ASIC chips, followed by the concentration boards (CoBo) that perform the data reduction and possibly additional signal processing, and route the resulting data through a network switch to back-end computers and storage units. The trigger is generated by the MuTanT module, that also serves to synchronize the clocks between all components of the system.

processing and storage units. The problem is exacerbated in TPC detectors by the need to sample and digitize the signal on each pad for a time more or less equal to the maximum drift time of the primary electrons in the active volume. It is in fact the equivalent of taking a snapshot of a large number of pixels each time a valid trigger is generated. For the AT-TPC for instance (see Section 2.7), the number of pixels in the 3D volume is equal to 5.24 million. Another important method for reducing the connectivity between the front-end modules and back-end data sinks is data serialization. By using sampling ADCs as transmitters and First-In First-Out (FIFO) stacks as receivers over differential signal pairs, it is possible to transmit large amounts of data through a relatively small number of wires. As an example, the AsAd front-end boards used in the GET electronics transmit the data of 256 channels with a depth of 512 time buckets over differential signal pairs, at a frequency of 25 MHz. Clocks and synchronization signals are of course required, but these methods are not far fetched from standard networking and telecommunication techniques. 4.2. Data concentration The very large amount of data generated by the front-end part of a digital system cannot in most practical cases be absorbed by data sinks of the receiver (or back-end) stages. Even using nowadays fast network technology does not solve the issue because the bottle-neck is then simply relegated to the data storage stage where the limitations arise from the writing speed of the hardware. A simple example can serve to illustrate the issue: considering a 10,000 channel detector with traces recorded over 512 time buckets at a rate of 1000 triggers per second, and assuming each datum consists of 2 bytes, the total throughput of data is 10 Gbytes/s. Parallelism techniques can of course be employed to reduce this amount by about an order of magnitude, but the data rate is still too high for regular storage media and the cost of storage hardware quickly becomes impractical. In addition, it is clear that the level of signal occupancy in the detector volume can be quite low, in particular for nuclear physics experiments where the particle multiplicities are also low (compared to particle physics where particle pair creation is a major source due to the large energy release). Data reduction can therefore be applied to considerably reduce the amount of data without compromising the quality and detail of the recorded traces. The difficulty lies in the speed at which this reduction needs to operate, which is basically the ‘‘raw’’ speed of incoming unfiltered data. This function can only be performed efficiently with specific wired logic that can be implemented in Field Programmable Gate Arrays (FPGA). Processors can be used to interface the communication protocols once the data has been reduced in size, but are not well adapted to process the raw flux of data. Such an architecture is generally adopted with variations depending on the systems or the needs of the detectors. In the GET electronics [54] developed for several active target detectors among others, the data

158

S. Beceiro-Novo et al. / Progress in Particle and Nuclear Physics 84 (2015) 124–165

reduction is performed by the Concentration Board (CoBo) module that drives and receives data from 4 front-end modules, instrumenting a total of 1024 channels. The main circuitry is embedded into a Virtex 5 FPGA produced by Xilinx, that also contains a PowerPC processor running the communication software via an Ethernet connection. The simplest data reduction algorithm consists of removing channels that have not recorded any signal. As noise is always present in all channels, this filtering involves the setting of a threshold that can be set above the noise level. Different levels of rejection can be applied, for instance the traces of channels that see a signal above threshold could be stored in their integrity, allowing a measure of the baseline for each event, or the algorithm could be set to only store the time buckets that have data above threshold, for a greater reduction. These choices are usually driven by compromises between the maximum trigger rate the detector is going to generate and the maximum data throughput sustainable by the data acquisition system. Too much dead time automatically leads to a reduction of the effective trigger rate, and the amount of statistics collected during the experiment. 4.3. Data storage Data storage is often the slowest component of the data acquisition chain, but at the same time the most important. While it was originally performed on magnetic tapes because of their large capacity, nowadays hard drive technology has improved the density of information beyond the tape capability, and storage is now using this type of media. Typical maximum writing speeds hover around 100 Mbytes/s, while maximum capacity has grown steadily and is up to 4 Tbytes per hard drive at the time of this writing. In terms of data acquisition performance, a good match of data throughput rate is desirable between the various components, in this case the data-receiving modules, back-end computer, network, and storing devices. Very often, some level of parallel architecture is required to boost the performance of the data collecting and storage stages. For instance, in the architecture shown in Fig. 46, each CoBo module can be uniquely linked to a receiving computer equipped with its local storage disk behind the network switch. The total traffic still has to be routed through the network switch, but network technology is typically faster than storage technology. In the case of the AT-TPC (see Section 2.7), a 10 Gbit/s network switch is used to route the data from its 10 CoBo modules to 10 individual computers, each with a network interface capable of 1 Gbit/s and matching storage speed of about 120 Mbytes/s. Note that even though this maximum performance can be achieved, running an entire experiment at this type of speed quickly becomes unpractical. The amount of data written at this rate would be about 100 Tbytes per day, requiring constant swapping and backing up of disks. 5. Tracking algorithms and simulations Already during the conception phase of a Active Target TPC, simulation of tracks are necessary in order to optimize the geometry of the detector for achieving the bets resolution with a necessarily limited number of pads. One of the major challenges in most experiments with active targets is to get a precise tracking algorithm that allows for the reconstruction of the kinematics of the measured tracks. In this sense there are big efforts historically done in the context of highenergy physics experiments and some of them are being applied in low-energy nuclear physics. However, the difficulties encountered between high-energy and low-energy tracking are quite different. In high-energy experiments the main issue comes from the large multiplicity of particles that leave tracks (easily in the thousands), with frequent occurrence of several hits in a single pad belonging to different tracks. At the same time, all particles are in the minimum ionization domain and cross the active area of the TPC from one end to the other. On the other hand, low-energy tracks can have very different lengths and ionization signals depending on their atomic number. The multiplicities however are usually quite low. Another important difference is the absence of a vertex detector in low-energy experiments. In active target detectors the determination of the vertex of the reaction is embedded in the same data that is used to determine the parameters of the particles emitted after the reaction. For these reasons, the techniques and algorithms typically used in high-energy experiments cannot be used ‘‘as is’’, and have to be tailored to the particular characteristics of the detector or type of experiment. Simulations will start in general by a more or less detailed description of the pulseshape expected at the end of the electronic chain as they will be registered. A comparison of experiment and simulation for the somewhat involved case of signals induced by alpha particles in pads in a resistive micromegas is shown in Fig. 47 with a resistive value of 4MΩ /square. The analytical formulas for the induced pulses with a resistive micromegas were taken from Ref. [137]. The analytical formulas were taken from Ref. [137]. As can be seen a very good agreement can be obtained. The extraction of the range of alpha particles in this case was done by a chi2 as shown in Fig. 48 for α particles of 5.6 MeV in a gas of He + CO2 (10%). A range resolution in range of 4 mm FWHM, or about 3.2%, corresponding to an energy resolution of 2%, is obtained. As can be seen, for each event the chi2 has much better localized minimum than the overall distribution, showing that the main limitation is due to range straggling. In some cases, full reconstruction can be achieved by a good energy loss calibration in the detector (see for instance [13]). As is shown in Fig. 49, by careful analysis of the range and energy loss it was possible to separate the 6 He from the 4 H. Other more complex cases need more elaborate algorithms, like for instance when complex pad plane geometries are used, or when the tracks are curved due to the magnetic field in the detector. An example of a common tracking algorithm is the Kalman filter [138] which is an algorithm that uses a series of measurements observed over time, containing noise (random variations) and other inaccuracies, and produces estimates of unknown variables that tend to be more precise than those based on a single measurement alone. More formally, the Kalman

S. Beceiro-Novo et al. / Progress in Particle and Nuclear Physics 84 (2015) 124–165

159

Fig. 47. Pulse shape as observed and simulated for a resistive micromegas. The integration time was 0.5 µs and the time is 60 ns/timebucket.

Fig. 48. χ 2 comparison of the maximum amplitude for pads after the Bragg peak for alpha particles, for 5 events, as a function of the simulated endpoint.

filter operates recursively on streams of noisy input data to produce a statistically optimal estimate of the underlying system state. In high energy physics very often the Kalman filter starts the iteration from the end of the track to find the energy at the vertex. In many of the active targets discussed in this review article, the tracks are fully stopped within the detector, so this approach has to be modified. An example of a Kalman filter applied to TPC’s can be found in [139]. Another common algorithm is the Hough transformation [140]. This particular algorithm is designed to identify the clusters of data that belong to the same track. This is a particularly arduous task in high-energy experiments where thousands of tracks need to be disentangled and identified. An example of this algorithm applied to a high-energy experiment is found in [141]. In low-energy experiments such as those conducted in active target detectors however, the issue is more to identify the clusters that belong to the particles before and after the reaction. Again, this implies a customization of the algorithm: for instance the tracks corresponding to the beam particles before the reaction are usually straight (even in a magnetic field parallel to the beam axis), whereas those corresponding to the emitted particles after the reaction are curved (see for instance Fig. 32). The consequence is that a linear Hough transform is adapted to identify the track prior to the reaction, while a circle Hough transform should be used to identify those after the reaction. The methods outlined above can be considered as good starting points for the determination of the parameters of the particles. However, because the dependence of the signals recorded on each pad with the trajectory of the tracks is highly non-linear, the best method to determine the parameters of the particles precisely and accurately is by comparing the data to simulations of the events, by minimizing a well-defined objective function such as a χ 2 . Due to the large non-linearities, this

160

S. Beceiro-Novo et al. / Progress in Particle and Nuclear Physics 84 (2015) 124–165

2500

2000

1500

1000

500

0

–80 –60 –40 –20

0

20

40

60

80 100

Fig. 49. Experimental profile of the energy deposit of reaction products in an event of elastic scattering of 6 He on 4 He. The best-fit simulations assuming He (solid line) or 4 He (dashed line) are shown together. Therefore the particle on the left and right clearly correspond to 4 He and 6 He respectively. Source: From [13]. 6

minimization cannot proceed via standard methods that rely on the determination of second derivatives, but rather using more randomized methods such as simulated annealing or Monte-Carlo. An example of a χ 2 minimization method based on Monte-Carlo is shown in Fig. 50, simulating the response of the AT-TPC detector presented in Section 2.7. The simulations used in this algorithm include various effects that influence the signals recorded on the pad plane, such as the longitudinal and transversal diffusion of primary electrons in the gas as they drift to the pad plane, fluctuations in the gain provided by the avalanche in the electron multiplication device (here a Micromegas [52]), and electronic noise of the pre-amplifier and amplifier stages of the electronics. In the example in Fig. 50(d) the best fit converges to the nominal value within 0.5%. 6. Conclusion The active target developments we described here stemmed from the increased use of secondary beams, that have a very limited intensity, and that imply for most studies of nuclear structure the use of inverse kinematics. There is already quite a large family of active targets detectors that are beginning to explore many reactions induced by radioactive beams. Different technical approaches were chosen in order to optimize with respect to the physics program, and this has made the family of active targets very rich technologically. These developments provide a high potential to pursue elastic and inelastic scattering, transfer reactions, excitation of giant resonances. These experiments surely will be pursued to gather new information about nuclear matter and nuclear structure away from equilibrium with respect to stable nuclei. The excess of neutrons or protons introduces a new degree of freedom. At low energy reactions of astrophysical interest can provide experimental cross section values, that can be introduced in astrophysical scenarios and network calculations, as well on the proton rich side for the r–p process, as well on the neutron rich side for the r-process. We have presented a number of active target detectors, many of them using or combined with Time Projection Chambers, where electrons produced from gas ionization are drifted to an electron amplification device, and their drift time is measured as one of the 3 dimensional position determinations. Quencher gases such as CO2 are often required to boost the electron amplification. The ability to use pure gases without quencher would be ideal to avoid having to disentangle contributions of the different components of the gas mixture. The test of different gas electron multiplier types and their eventual combination is presently an active domain of development, which should provide a way to good performance with pure gases such as H2 , D2 , 4 He and 3 He. The future increase of computer power for data analysis and data storage will enable to get complete physics results in short times for more complex experiments as compared to present. This also depends strongly on the development of precise track analysis, that will likely need a concerted effort to find and tune the appropriate algorithms. Intrinsic limitations from straggling phenomena are not presently the limiting factor of final achievable resolution. The resolution limitations are more due to eventual short track length, or quickly changing energy loss as a function of range in the detector, and require a precise simulation. As compared to high energy tracking the difficulties are quite different: the number of particles is much more restricted, and the dynamics of energy loss is much larger and more difficult to handle. A better combined resolution of the gas-amplifiers and the electronics will be necessary for best results for medium to heavy nuclei, and specific developments will be necessary mainly of the gas amplifiers. The combination of active target detectors with high performance gamma detectors is another domain of development. Two approaches are currently under way. The first is the construction of an active target detector with a relatively small volume, that makes it more or less straightforward to insert it inside a large γ -ray detector array. The second is the

S. Beceiro-Novo et al. / Progress in Particle and Nuclear Physics 84 (2015) 124–165

161

Fig. 50. Track analysis by Monte-Carlo test particles. The testparticles are simulated with the full fluctuations expected (angular straggling, range straggling, gain fluctuations, etc.). In panels (a)–(c) only the energy of the test particles is changed. The panel (a) shows the iterative restriction of the phase space of energy: every 450 test particles the domain of search is reduced, with as new center the value of the lowest χ 2 of the previous series. In panel (b) the χ 2 for triangular pads is plotted. The vertical bar indicates the energy value used to create the track to be analyzed. Panel (c) shows the same for rectangular pads. Panel (d) shows the same plot as panel (b), but all the parameters in the six dimensional space that describes the particle (origin x, y, z; angles of emission θ , φ , energy) are varied.

combination of a large volume device in a magnetic field, surrounded by special detectors such as those based on LaBr3 crystals for instance, that can operate in a strong magnetic field. These progresses go hand-in-hand with the future increase of radioactive beam intensities at facilities under construction such as FRIB [142] and FAIR [143], as well as constantly improved performance at existing facilities. The increase in luminosity provided by this type of detector will open the door to future studies within the next decade approaching drip lines of stability for heavier nuclei than presently accessible. The active target detectors described in this article will without doubt be at the forefront of exploring the properties of the most exotic nuclear systems. The keywords that distinguish them from other types of detectors are a unique combination of luminosity, thanks to the thick targets they can employ and the large efficiency they provide combined with good resolution, by the precise determination of the parameters of the particle they detect. References [1] [2] [3] [4] [5] [6]

I. Tanihata, H. Savajols, R. Kanungo, Prog. Part. Nucl. Phys. 68 (2013) 215–313. A. Collaboration, G. Aad, et al., J. Instrum. 3 (2008) S08003. F. Hartmann, A. Sharma, Nucl. Instrum. Methods Phys. Res. A 666 (2012) 1–9. F. Sauli, Phys. Rep. 403 (2004) 471–504. T. Francke, Innovative Applications and Developments of Micro-Pattern Gaseous Detectors, IGI Global, 2014. S. Neumaier, G. Alkhazov, M. Andronenko, A. Dobrovolsky, P. Egelhof, G. Gavrilov, H. Geissel, H. Irnich, A. Khanzadeev, G. Korolev, A. Lobodenko, G. Münzenberg, M. Mutterer, W. Schwab, D. Seliverstov, T. Suzuki, N. Timofeev, A. Vorobyov, V. Yatsoura, Nucl. Phys. A 712 (3–4) (2002) 247–268. http://dx.doi.org/10.1016/S0375-9474(02)01274-5. URL http://www.sciencedirect.com/science/article/pii/S0375947402012745. [7] C. Demonchy, W. Mittig, H. Savajols, P. Roussel-Chomaz, M. Chartier, B. Jurado, L. Giot, D. Cortina-Gil, M. Caamano, G. Ter-Arkopian, et al., Nucl. Instrum. Methods Phys. Res. A 573 (1) (2007) 145–148. [8] Actar: An active target detector for the study of extremely exotic nuclei, GANIL proposal, 2009.

162

S. Beceiro-Novo et al. / Progress in Particle and Nuclear Physics 84 (2015) 124–165

[9] Y. Mizoi, T. Fukuda, Y. Matsuyama, T. Miyachi, J. Nakano, N. Fukuda, M. Hirai, H. Kobinata, Y. Watanabe, H. Sakurai, Y. Watanabe, A. Yoshida, Nucl. Instrum. Methods Phys. Res. A 431 (1–2) (1999) 112–122. http://dx.doi.org/10.1016/S0168-9002(99)00253-3. URL http://www.sciencedirect.com/ science/article/pii/S0168900299002533. [10] T. Hashimoto, H. Ishiyama, T. Ishikawa, T. Kawamura, K. Nakai, Y. Watanabe, H. Miyatake, M. Tanaka, Y. Fuchi, N. Yoshikawa, S. Jeong, I. Katayama, T. Nomura, T. Furukawa, S. Mitsuoka, K. Nishio, M. Matsuda, H. Ikezoe, T. Fukuda, S. Das, P. Saha, Y. Mizoi, T. Komatsubara, M. Yamaguchi, Y. Tagishi, Nucl. Instrum. Methods Phys. Res. A 556 (1) (2006) 339–349. http://dx.doi.org/10.1016/j.nima.2005.10.018. URL http://www.sciencedirect.com/ science/article/pii/S0168900205019376. [11] S. Ota, H. Tokieda, C. Lee, M. Donzono, H. Matsubara, S. Kawase, Y. Kubota, S. Michimasa, H. Yamaguchi, M. Sako, T. Gunji, Y. Kikuchi, T. Hashimoto, H. Hamagaki, T. Uesaka, S. Kubono, H. Otsu, Y. Maeda, E. Takada, Annu. Report, Cent. Nucl. Study, Univ. Tokyo CNS-REP-90 (2011) 70–71. [12] T. Furuno, T. Kawabata, S. Adachi, Y. Ayyad, T. Baba, T. Hashimoto, Y. Ishii, S. Kabuki, H. Kubo, Y. Matsuda, et al., Development of Maiko: the Active Target with µ-pic for ri Beam Experiments, in: Journal of Physics: Conference Series, vol. 569, IOP Publishing, 2014, p. 012042. [13] D. Suzuki, M. Ford, D. Bazin, W. Mittig, W. Lynch, T. Ahn, S. Aune, E. Galyaev, A. Fritsch, J. Gilbert, et al., Nucl. Instrum. Methods Phys. Res. A 691 (2012) 39–54. [14] W. Mittig, S. Beceiro-Novo, A. Fritsch, F. Abu-Nimeh, D. Bazin, T. Ahn, W. Lynch, F. Montes, A. Shore, D. Suzuki, et al., Nucl. Instrum. Methods Phys. Res. A 784 (2015) 494–498. [15] A. Laird, P. Amaudruz, L. Buchmann, S. Fox, B. Fulton, D. Gigliotti, T. Kirchner, P. Mumby-Croft, R. Openshaw, M. Pavan, J. Pearson, G. Ruprecht, G. Sheffer, P. Walden, Nucl. Instrum. Methods Phys. Res. A 573 (1–2) (2007) 306–309. http://dx.doi.org/10.1016/j.nima.2006.10.384. URL http://www.sciencedirect.com/science/article/pii/S0168900206021553. [16] L. Linhardt, L. Baby, D. Bardayan, J. Blackmon, H. Gardiner, E. Johnson, E. Koschiy, M. Macon, M. Matos, B. Rasco, et al., Measurement of 17f+ p Reactions with Anasen, in: Journal of Physics: Conference Series, vol. 403, IOP Publishing, 2012, p. 012036. [17] A. Obertelli, A. Delbart, S. Anvar, L. Audirac, G. Authelet, H. Baba, B. Bruyneel, D. Calvet, F. Château, A. Corsi, et al., Eur. Phys. J. A 50 (2014) 8. [18] M. Gai, M. Ahmed, S. Stave, W. Zimmerman, A. Breskin, B. Bromberger, R. Chechik, V. Dangendorf, T. Delbar, R. France III, et al., J. Instrum. 5 (12) (2010) P12004. [19] T. Hashimoto, S. Kubono, H. Yamaguchi, S. Hayakawa, D.M. Kahl, S. Ota, S. Michimasa, H. Tokieda, H. Ishiyama, Y.X. Watanabe, Y. Hirayama, N. Imai, H. Miyatake, S.C. Jeong, K. Yamaguchi, I. Arai, T. Komatsubara, T. Fukuda, Y. Mizoi, S.K. Das, N. Iwasa, T. Yamada, A.A. Chen, J.J. He, Direct measurement of the [sup 18]Ne(α , p)[sup 21]Na reaction, in: Orig. MATTER Evol. GALAXIES 2011, vol. 1484, AIP Publishing, 2012, pp. 454–456. http://dx.doi.org/10.1063/1.4763451. URL http://scitation.aip.org.proxy.library.nd.edu/content/aip/proceeding/aipcp/10.1063/1.4763451. [20] D. Kahl, T. Hashimoto, N.N. Duy, S. Kubono, H. Yamaguchi, D.N. Binh, A.A. Chen, S. Cherubini, S. Hayakawa, J.J. He, H. Ishiyama, N. Iwasa, L.H. Khiem, Y.K. Kwon, S. Michimasa, T. Nakao, S. Ota, T. Teranishi, H. Tokieda, Y. Wakabayashi, T. Yamada, L.Y. Zhang, Active target studies of the α p-process at CRIB, in: Orig. MATTER Evol. GALAXIES 2013 Proc. 12th Int. Symp. Orig. Matter Evol. Galaxies, vol. 1594, AIP Publishing, 2014, pp. 163–170. http://dx.doi.org/10.1063/1.4874062. URL http://scitation.aip.org.proxy.library.nd.edu/content/aip/proceeding/aipcp/10.1063/1.4874062. [21] K. Yamaguchi, H. Ishiyama, M. Tanaka, Y. Watanabe, H. Miyatake, Y. Hirayama, N. Imai, H. Makii, Y. Fuchi, S. Jeong, T. Nomura, Y. Mizoi, S. Das, T. Fukuda, T. Hashimoto, I. Arai, Nucl. Instrum. Methods Phys. Res. A 623 (1) (2010) 135–137. http://dx.doi.org/10.1016/j.nima.2010.02.173. URL http://www.sciencedirect.com/science/article/pii/S0168900210004407. [22] H. Ishiyama, K. Yamaguchi, Y. Mizoi, Y.X. Watanabe, S.K. Das, T. Hashimoto, H. Miyatake, Y. Hirayama, N. Imai, M. Oyaizu, S.C. Jeong, T. Fukuda, S. Mitsuoka, H. Makii, T.K. Sato, J. Instrum. 7 (03) (2012) C03036. http://dx.doi.org/10.1088/1748-0221/7/03/C03036. URL http://stacks.iop.org/17480221/7/i=03/a=C03036. [23] A. Vorobyov, G. Korolev, V. Schegelsky, G. Solyakin, G. Sokolov, Y. Zalite, Nucl. Instrum. Methods 119 (1974) 509–519. http://dx.doi.org/10.1016/0029554X(74)90801-5. URL http://www.sciencedirect.com/science/article/pii/0029554X74908015. [24] S. Karataglidis, K. Amos, B.A. Brown, P.K. Deb, Phys. Rev. C 65 (4) (2002) 044306. http://dx.doi.org/10.1103/PhysRevC.65.044306. URL http://link.aps.org/doi/10.1103/PhysRevC.65.044306. [25] S. Ilieva, F. Aksouh, G. Alkhazov, L. Chulkov, A. Dobrovolsky, P. Egelhof, H. Geissel, M. Gorska, A. Inglessi, R. Kanungo, A. Khanzadeev, O. Kiselev, G. Korolev, X. Le, Y. Litvinov, C. Nociforo, D. Seliverstov, L. Sergeev, H. Simon, V. Volkov, A. Vorobyov, H. Weick, V. Yatsoura, A. Zhdanov, Nucl. Phys. A 875 (2012) 8–28. http://dx.doi.org/10.1016/j.nuclphysa.2011.11.010. URL http://www.sciencedirect.com/science/article/pii/S037594741100666X. [26] O. Kiselev, F. Aksouh, A. Bleile, O. Bochkarev, L. Chulkov, D. Cortina-Gil, A. Dobrovolsky, P. Egelhof, H. Geissel, M. Hellström, N. Isaev, B. Komkov, M. Matoš, F. Moroz, G. Münzenberg, M. Mutterer, V. Mylnikov, S. Neumaier, V. Pribora, D. Seliverstov, L. Sergueev, A. Shrivastava, K. Sümmerer, H. Weick, M. Winkler, V. Yatsoura, Nucl. Instrum. Methods Phys. Res. A 641 (1) (2011) 72–86. http://dx.doi.org/10.1016/j.nima.2011.03.046. URL http://www.sciencedirect.com/science/article/pii/S0168900211006826. [27] E. Ibraeva, O. Imambekov, A. Dzhazairov-Kahramanov, Internat. J. Modern Phys. E 22 (04) (2013) 1350017. http://dx.doi.org/10.1142/S0218301313500171. URL http://www.worldscientific.com/doi/abs/10.1142/S0218301313500171. [28] G. Alkhazov, M. Andronenko, A. Dobrovolsky, P. Egelhof, G. Gavrilov, H. Geissel, H. Irnich, A. Khanzadeev, G. Korolev, A. Lobodenko, G. Münzenberg, M. Mutterer, S. Neumaier, F. Nickel, W. Schwab, D. Seliverstov, T. Suzuki, J. Theobald, N. Timofeev, A. Vorobyov, V. Yatsoura, Phys. Rev. Lett. 78 (12) (1997) 2313–2316. http://dx.doi.org/10.1103/PhysRevLett.78.2313. URL http://link.aps.org/doi/10.1103/PhysRevLett.78.2313. [29] F. Aksouh, O.A. Kisselev, A. Bleile, O.V. Bocharev, L.V. Chulkov, D. Cortina-Gil, A. Dobrovolsky, P. Egelhof, H. Geissel, M. Hellström, N. Isaev, B. Komkov, M. Matoš, F. Moroz, G. Münzenberg, M. Mutterer, V. Mylnikov, S. Neumaier, V. Pribora, D. Seliverstov, L. Sergueev, A. Shrivastava, K. Sümmerer, H. Weick, M. Winkler, V. Yatsoura, Ric. Sci. Ed Educ. Perm. Suppl. No. 122 (2003) 157–166. [30] A.C. Villari, et al., Nucl. Phys. A 693 (1) (2001) 465–476. [31] http://srim.org. [32] K. Lau, J. Pyrlik, Nucl. Instrum. Methods Phys. Res. A 366 (2) (1995) 298–309. [33] M. Akatsu, K. Fujimoto, Y. Fujita, H. Hirano, M. Hirose, K. Inami, S. Itami, A. Ishikawa, Y. Iwasaki, K. Joo, et al., Nucl. Instrum. Methods Phys. Res. A 454 (2) (2000) 322–333. [34] A. Andronic, H. Appelshäuser, C. Blume, P. Braun-Munzinger, D. Bucher, O. Busch, A.C. Ramirez, V. Cătănescu, M. Ciobanu, H. Daues, et al., Nucl. Instrum. Methods Phys. Res. A 498 (1) (2003) 143–154. [35] J. Santiard, et al., Cern-ecp/94-17, Nucl. Instr. and Meth. A 360. [36] T. Roger, M. Caamaño, C. Demonchy, W. Mittig, H. Savajols, I. Tanihata, Nucl. Instrum. Methods Phys. Res. A 638 (1) (2011) 134–142. [37] S. Sambi, R. Raabe, M. Borge, M. Caamano, S. Damoy, B. Fernández-Domínguez, F. Flavigny, H. Fynbo, J. Gibelin, G. Grinyer, et al., Eur. Phys. J. A 51 (3) (2015) 1–7. [38] F. de Oliveira Santos, P. Himpe, M. Lewitowicz, I. Stefan, N. Smirnova, N. Achouri, J. Angélique, C. Angulo, L. Axelsson, D. Baiborodin, et al., Eur. Phys. J. A 24 (2) (2005) 237–247. [39] H. Savajols, C. Demonchy, W. Mittig, M. Caamano, M. Chartier, D. Cortina-Gil, A. Gillibert, T. Roger, P. Roussel-Chomaz, I. Tanihata, Nuclear Instruments and Methods in Physics Research Section B: Beam Interactions with Materials and Atoms 266 (19) (2008) 4583–4588. [40] I. Tanihata, M. Alcorta, D. Bandyopadhyay, R. Bieri, L. Buchmann, B. Davids, N. Galinski, D. Howell, W. Mills, S. Mythili, et al., Phys. Rev. Lett. 100 (19) (2008) 192502. [41] G. Potel, F. Barranco, E. Vigezzi, R.A. Broglia, Phys. Rev. Lett. 105 (17) (2010) 172502. [42] M. Caamano, D. Cortina-Gil, W. Mittig, H. Savajols, M. Chartier, C. Demonchy, B. Fernández, M.G. Hornillos, A. Gillibert, B. Jurado, et al., Phys. Rev. Lett. 99 (6) (2007) 062502. [43] D. Cortina-Gil, W. Mittig, Europhys. News 41 (2) (2010) 23–26. [44] L. Grigorenko, I. Mukha, C. Scheidenberger, M.V. Zhukov, Phys. Rev. C 84 (2) (2011) 021303. [45] T. Roger, H. Savajols, I. Tanihata, W. Mittig, M. Alcorta, D. Bandyopadhyay, R. Bieri, L. Buchmann, M. Caamano, B. Davids, et al., Phys. Rev. C. 79 (3) (2009).

S. Beceiro-Novo et al. / Progress in Particle and Nuclear Physics 84 (2015) 124–165

163

[46] C. Bachelet, G. Audi, C. Gaulard, C. Guénaut, F. Herfurth, D. Lunney, M. de Saint Simon, C. Thibault, Phys. Rev. Lett. 100 (18) (2008) 182501. [47] M. Smith, M. Brodeur, T. Brunner, S. Ettenauer, A. Lapierre, R. Ringle, V. Ryjkov, F. Ames, P. Bricault, G. Drake, et al., Phys. Rev. Lett. 101 (20) (2008) 202501. [48] G. Audi, A. Wapstra, C. Thibault, Nucl. Phys. A 729 (1) (2003) 337–676. [49] C. Monrozeau, E. Khan, Y. Blumenfeld, C. Demonchy, W. Mittig, P. Roussel-Chomaz, D. Beaumel, M. Caamano, D. Cortina-Gil, J. Ebran, et al., Phys. Rev. Lett. 100 (4) (2008) 042501. [50] M. Vandebrouck, J. Gibelin, E. Khan, N. Achouri, H. Baba, D. Beaumel, Y. Blumenfeld, M. Caamaño, L. Càceres, G. Colò, et al., Phys. Rev. Lett. 113 (3) (2014) 032504. [51] E. Khan, N. Paar, D. Vretenar, Phys. Rev. C 84 (2011) 051301. http://dx.doi.org/10.1103/PhysRevC.84.051301. URL http://link.aps.org/doi/10.1103/ PhysRevC.84.051301. [52] G. Charpak, J. Derré, Y. Giomataris, P. Rebourgeard, Nucl. Instrum. Methods Phys. Res. A 478 (1–2) (2002) 26–36. http://dx.doi.org/10.1016/S01689002(01)01713-2. URL http://www.sciencedirect.com/science/article/pii/S0168900201017132. [53] J. Pancin, S. Damoy, D.P. Loureiro, V. Chambert, F. Dorangeville, F. Druillole, G. Grinyer, A. Lermitage, A. Maroni, G. Noël, et al., Nucl. Instrum. Methods Phys. Res. A 735 (2014) 532–540. [54] E. Pollacco, S. Anvar, H. Baba, P. Baron, D. Bazin, C. Belkhiria, B. Blank, J. Chavas, P. Chomaz, E. Delagnes, et al., Phys. Procedia 37 (2012) 1799–1804. [55] D. Perez-Loureiro, Private Communication. [56] W. Christie, J. Romero, F. Brady, C. Tull, C. Castaneda, E. Barasch, M. Webb, J. Drummond, H. Crawford, I. Flores, D. Greiner, P. Lindstrom, H. Sann, J. Young, Nucl. Instrum. Methods Phys. Res. A 255 (3) (1987) 466–476. http://dx.doi.org/10.1016/0168-9002(87)91213-7. URL http://www. sciencedirect.com/science/article/pii/0168900287912137. [57] K. Kimura, Y. Akiba, Y. Miake, S. Nagamiya, Nucl. Instrum. Methods Phys. Res. A 297 (1–2) (1990) 190–198. http://dx.doi.org/10.1016/01689002(90)91366-J. URL http://www.sciencedirect.com/science/article/pii/016890029091366J. [58] R.N. Boyd, I. Tanihata, N. Inabe, T. Kubo, T. Nakagawa, T. Suzuki, M. Yonokura, X.X. Bai, K. Kimura, S. Kubono, S. Shimoura, H.S. Xu, D. Hirata, Phys. Rev. Lett. 68 (9) (1992) 1283–1286. http://dx.doi.org/10.1103/PhysRevLett.68.1283. URL http://link.aps.org/doi/10.1103/PhysRevLett.68.1283. [59] X. Gu, R. Boyd, M. Farrell, J. Kalen, C. Mitchell, J. Kolata, M. Belbot, K. Lamkin, K. Ashktorab, F. Becchetti, J. Brown, D. Roberts, K. Kimura, I. Tanihata, K. Yoshida, M. Islam, Phys. Lett. B 343 (1–4) (1995) 31–35. http://dx.doi.org/10.1016/0370-2693(94)01480-Z. URL http://www.sciencedirect.com/ science/article/pii/037026939401480Z. [60] P.F.F. Carnelli, S. Almaraz-Calderon, K.E. Rehm, M. Albers, M. Alcorta, P.F. Bertone, B. Digiovine, H. Esbensen, J.O.F. Niello, D. Henderson, C.L. Jiang, J. Lai, S.T. Marley, O. Nusair, T. Palchan-Hazan, R.C. Pardo, M. Paul, C. Ugalde, Phys. Rev. Lett. 112 (19) (2014) 192701. http://dx.doi.org/10.1103/PhysRevLett.112.192701. URL http://link.aps.org/doi/10.1103/PhysRevLett.112.192701. [61] R. Allemand, G. Thomas, Nucl. Instrum. Methods 137 (1) (1976) 141–149. http://dx.doi.org/10.1016/0029-554X(76)90260-3. URL http://www. sciencedirect.com/science/article/pii/0029554X76902603. [62] P. Némethy, P.J. Oddone, N. Toge, A. Ishibashi, Nucl. Instrum. Methods Phys. Res. 212 (1–3) (1983) 273–280. http://dx.doi.org/10.1016/01675087(83)90702-0. URL http://www.sciencedirect.com/science/article/pii/0167508783907020. [63] F. Sauli, Nucl. Instrum. Methods Phys. Res. A 386 (2–3) (1997) 531–534. http://dx.doi.org/10.1016/S0168-9002(96)01172-2. URL http://www. sciencedirect.com/science/article/pii/S0168900296011722. [64] K. Yamaguchi, H. Ishiyama, Y. Mizoi, Y.X. Watanabe, T. Hashimoto, M.H. Tanaka, H. Miyatake, Y. Hirayama, N. Imai, Y. Fuchi, S.C. Jeong, T. Nomura, S.K. Das, T. Fukuda, H. Makii, S. Mitsuoka, I. Arai, H. Yamaguchi, S. Kubono, Y. Wakabayashi, S. Hayakawa, I. Tanihara, H.J. Ong, A. Tamii, T. Kishimoto, T. Kajino, S. Kubono, T. Shima, Development of the GEM-MSTPC for Studies of Astrophysical Nuclear Reaction Rates, 2010, pp. 433–435. http://dx.doi.org/10.1063/1.3485192. URL http://scitation.aip.org/content/aip/proceeding/aipcp/10.1063/1.3485192. [65] A. Orthen, H. Wagner, H. Besch, R. Menk, A. Walenta, Nucl. Instrum. Methods Phys. Res. A 492 (1–2) (2002) 160–177. http://dx.doi.org/10.1016/S01689002(02)01276-7. URL http://www.sciencedirect.com/science/article/pii/S0168900202012767. [66] F.D. Amaro, C.A.O. Henriques, M.R. Jorge, A.L.M. Silva, E.D.C. Freitas, J. Instrum. 9 (12) (2014) P12002. http://dx.doi.org/10.1088/17480221/9/12/P12002. URL http://stacks.iop.org/1748-0221/9/i=12/a=P12002. [67] D. Kahl, T. Hashimoto, S. Kubono, S. Ota, H. Yamaguchi, Algorithms for active target data analysis, Annu. Report, Cent. Nucl. Study, Univ. Tokyo CNS-REP-92 (2012) 55–56. [68] Y. Mizoi, T. Fukuda, Y. Matsuyama, T. Miyachi, H. Miyatake, N. Aoi, N. Fukuda, M. Notani, Y. Watanabe, K. Yoneda, M. Ishihara, H. Sakurai, A. Yoshida, Phys. Rev. C 62 (6) (2000) 065801. http://dx.doi.org/10.1103/PhysRevC.62.065801. URL http://link.aps.org/doi/10.1103/PhysRevC.62.065801. [69] H. Miyatake, H. Ishiyama, M.-H. Tanaka, Y. Watanabe, N. Yoshikawa, S. Jeong, Y. Matsuyama, Y. Fuchi, T. Nomura, T. Hashimoto, T. Ishikawa, K. Nakai, S. Das, P. Saha, T. Fukuda, K. Nishio, S. Mitsuoka, H. Ikezoe, S. Ichikawa, M. Matsuda, Y. Mizoi, T. Furukawa, H. Izumi, T. Shimoda, M. Terasawa, Nucl. Phys. A 738 (2004) 401–405. http://dx.doi.org/10.1016/j.nuclphysa.2004.04.070. URL http://www.sciencedirect.com/science/ article/pii/S0375947404006116. [70] H. Ishiyama, T. Hashimoto, T. Ishikawa, Y. Watanabe, S. Das, H. Miyatake, Y. Mizoi, T. Fukuda, M.-H. Tanaka, Y. Fuchi, N. Yoshikawa, Y. Hirayama, S. Jeong, T. Nomura, I. Katayama, S. Mitsuoka, K. Nishio, M. Matsuda, P. Saha, S. Ichikawa, H. Ikezoe, T. Furukawa, H. Izumi, T. Shimoda, K. Nakai, Phys. Lett. B 640 (3) (2006) 82–85. http://dx.doi.org/10.1016/j.physletb.2006.07.036. URL http://www.sciencedirect.com/science/article/pii/ S0370269306009208. [71] S.K. Das, A New Measurement of the 8Li(α , n)11B Reaction for Astrophysical Interest, in: AIP Conf. Proc., vol. 847, AIP, 2006, pp. 374–376. http://dx.doi.org/10.1063/1.2234430. URL http://scitation.aip.org.proxy.library.nd.edu/content/aip/proceeding/aipcp/10.1063/1.2234430. [72] M. La Cognata, A. Del Zoppo, Astrophys. J. 736 (2) (2011) 148. http://dx.doi.org/10.1088/0004-637X/736/2/148. URL http://iopscience.iop.org/0004637X/736/2/148/article/. [73] T. Ishikawa, H. Ishiyama, T. Hashimoto, T. Kawamura, H. Miyatake, M. Tanaka, Y. Fuchi, N. Yoshikawa, S. Jeong, Y. Watanabe, H. Kawakami, I. Katayama, T. Nomura, S. Mitsuoka, K. Nishio, S. Ichikawa, H. Ikezoe, T. Furukawa, H. Yano, H. Izumi, Y. Mizoi, M. Terasawa, T. Fukuda, K. Nakai, T. Shimoda, Nucl. Phys. A 718 (2003) 484–486. http://dx.doi.org/10.1016/S0375-9474(03)00864-9. URL http://www.sciencedirect.com/science/ article/pii/S0375947403008649. [74] H. Ishiyama, Study of Astrophysical (α , n) Reactions Using Light-Neutron Rich Radioactive Nuclear Beams, in: AIP Conf. Proc., vol. 847, AIP, 2006, pp. 249–254. http://dx.doi.org/10.1063/1.2234409. URL http://scitation.aip.org.proxy.library.nd.edu/content/aip/proceeding/aipcp/10.1063/1.2234409. [75] N.N. Duy, L.H. Khiem, S. Kubono, H. Yamaguchi, D.M. Kahl, T. Hashimoto, S. Ota, Y. Wakabayashi, T. Komatsubara, T. Teranishi, S. Kato, N. Iwasa, T. Yamada, Y.K. Kwon, A. Kim, Y.H. Kim, J.S. Song, J. Hu, J. Ito, Annu. Report, Cent. Nucl. Study, Univ. Tokyo (2011) 9–10. [76] D.H. Youngblood, Y.-W. Lui, H.L. Clark, Phys. Rev. C 60 (6) (1999) 067302. http://dx.doi.org/10.1103/PhysRevC.60.067302. URL http://link.aps.org/ doi/10.1103/PhysRevC.60.067302. [77] M. Itoh, H. Sakaguchi, M. Uchida, T. Ishikawa, T. Kawabata, T. Murakami, H. Takeda, T. Taki, S. Terashima, N. Tsukahara, Y. Yasuda, M. Yosoi, U. Garg, M. Hedden, B. Kharraja, M. Koss, B.K. Nayak, S. Zhu, H. Fujimura, M. Fujiwara, K. Hara, H.P. Yoshida, H. Akimune, M.N. Harakeh, M. Volkerts, Phys. Rev. C 68 (6) (2003) 064602. http://dx.doi.org/10.1103/PhysRevC.68.064602. URL http://link.aps.org/doi/10.1103/PhysRevC.68.064602. [78] S. Ota, S. Michimasa, T. Gunji, H. Yamaguchi, R. Akimoro, T. Hashimoto, K. Devid, S. Hayakawa, H. Tokieda, S. Kawase, T. Tsuji, S. Kubono, H. Hamagaki, T. Uesaka, Annu. Report, Cent. Nucl. Study, Univ. Tokyo CNS-REP-86 (2009) 57–58. [79] S. Ota, S. Michimasa, T. Gunji, H. Yamaguchi, H. Tokieda, T. Hashimoto, R. Akimoro, D. Kahl, M. Dozono, S. Hayakawa, S. Kawase, Y. Kikuchi, Y. Maeda, H. Matsubara, H. Otsu, S. Kubono, H. Hamagaki, T. Uesaka, CNS Active Target Development and Test Experiment at HIMAC, Annu. Report, Cent. Nucl. Study, Univ. Tokyo CNS-REP-88, 2010, pp. 47–48.

164

S. Beceiro-Novo et al. / Progress in Particle and Nuclear Physics 84 (2015) 124–165

[80] R. Akimoto, S. Ota, S. Michimasa, T. Gunji, H. Yamaguchi, T. Hashimoto, H. Tokieda, T. Tsuji, S. Kawase, H. Hamagaki, T. Uesaka, S. Kubono, T. Isobe, T. Kawabata, A. Ozawa, H. Suzuki, D. Nagae, T. Moriguchi, Y. Ito, Y. Ishibashi, H. Ooishi, Y. Abe, Development and Performance Evaluation of a GEM-TPC for the Experiment of Highly Excited State of Nuclei with High Rate and High Energy Unstable Nuclear Beam, Annu. Report, Cent. Nucl. Study, Univ. Tokyo CNS-REP-86, 2009, pp. 59–60. [81] S. Ota, H. Tokieda, C.S. Lee, Y.N. Watanabe, J. Radioanal. Nucl. Chem. (2015) 1–5. [82] C.S. Lee, S. Ota, H. Tokieda, R. Kojima, Y.N. Watanabe, T. Uesaka, J. Instrum. 9 (05) (2014) C05014. http://dx.doi.org/10.1088/1748-0221/9/05/C05014. URL http://iopscience.iop.org/1748-0221/9/05/C05014. [83] H. Tokieda, S. Ota, M. Dozono, C.S. Lee, T. Uesaka, Measurements of Gas Gain of Gas Electron Multiplier with 1-atm Deuterium Gas, Annu. Report, Cent. Nucl. Study, Univ. Tokyo CNS-REP-92, 2012, pp. 51–52. [84] C.S. Lee, S. Ota, H. Tokieda, T. Uesaka, Properties of Thick-GEM in Low Pressure Deuterium for Low Pressure Gaseous Active Target, Annu. Report, Cent. Nucl. Study, Univ. Tokyo CNS-REP-92, 2012, pp. 53–54. [85] K. Sato, K. Endo, E. Mashiro, T. Hattori, A. Itano, T. Kanai, M. Kanazawa, K. Kawachi, T. Kohno, S. Matsumoto, M. Yoshitoshi, A. Noda, K. Noda, H. Ogawa, Y. Sato, F. Soga, H. Suzuki, S. Watanabe, S. Yamada, T. Yamada, Y. Hirao, Part. Accel. 33 (1990) 147–152. [86] C.S. Lee, S. Ota, H. Tokieda, M. Dozono, H. Matsubara, S. Kawase, Y. Kubota, S. Michimasa, H. Yamaguchi, H. Hamagaki, M. Sako, Y. Maeda, T. Uesaka, H. Otsu, K. S., Annu. Report, Cent. Nucl. Study, Univ. Tokyo CNS-REP-90, 2011, pp. 72–73. [87] S. Ota, H. Tokieda, A. Corsi, M. Dozno, J. Gibelin, T. Gunji, H. Hamagaki, T. Kawabata, S. Kawase, S. Kubono, Y. Kubota, C.S. Lee, Y. Maeda, H. Matsubara, Y. Matsuda, S. Michimasa, T. Nakao, T. Nishi, A. Obertelli, K. Okochi, H. Otsu, T. Ryan, C. Santamaria, M. Sasano, M. Takaki, Y. Tanaka, T. Uesaka, K. Yako, H. Yamaguchi, J. Zenihiro, E. Takada, Annu. Report, Cent. Nucl. Study, Univ. Tokyo CNS-REP-92 (2012) 49–50. [88] M. Ito, Phys. Rev. C 83 (4) (2011) 044319. [89] A. Ochi, T. Nagayoshi, S. Koishi, T. Tanimori, T. Nagae, M. Nakamura, Nucl. Instrum. Methods Phys. Res. A 471 (1) (2001) 264–267. [90] Y. Mizumura, T. Tanimori, H. Kubo, A. Takada, J. Parker, T. Mizumoto, S. Sonoda, D. Tomono, T. Sawano, K. Nakamura, et al., J. Instrum. 9 (05) (2014) C05045. [91] D. Suzuki, D. Bazin, W. Mittig, W. Lynch, C. Hewko, A. Roux, D.B. Ali, J. Browne, E. Galyaev, M. Ford, et al., Nucl. Instrum. Methods Phys. Res. A 660 (1) (2011) 64–68. [92] D. Suzuki, A. Shore, W. Mittig, J. Kolata, D. Bazin, M. Ford, T. Ahn, F. Becchetti, S.B. Novo, D.B. Ali, et al., Phys. Rev. C 87 (5) (2013) 054301. [93] A. Fritsch, The search for cluster structure in the 14 c with the prototype at-tpc (Ph.D. dissertation). [94] T. Suhara, Y. Kanada-En’yo, Progr. Theoret. Phys. 123 (2) (2010) 303–325. [95] M. Ito, Modern Phys. Lett. A 21 (31n33) (2006) 2429–2438. [96] Y. Kanada-En’yo, H. Horiuchi, A. Doté, Phys. Rev. C 60 (6) (1999) 064304. [97] K. Arai, Phys. Rev. C 69 (1) (2004) 014309. [98] N. Itagaki, S. Okabe, Phys. Rev. C 61 (4) (2000) 044306. [99] L.H. De La Peña, P. Hess, G. Lévai, A. Algora, J. Phys. G: Nucl. Phys. 27 (10) (2001) 2019. [100] P. Descouvemont, Nucl. Phys. A 699 (3) (2002) 463–478. [101] S. Darmodjo, R.D. Alders, D.G. Martin, P. Dyer, S. Ali, S.A.A. Zaidi, Phys. Rev. C 4 (3) (1971) 672–683. http://dx.doi.org/10.1103/PhysRevC.4.672. URL http://link.aps.org/doi/10.1103/PhysRevC.4.672. [102] H. Scott, W. Galati, J. Weil, M. McEllistrem, Phys. Rev. 172 (4) (1968) 1139. [103] R.C. Pardo, G. Savard, S. Baker, C. Davids, E.F. Moore, R. Vondrasek, G. Zinkann, Nucl. Instrum. Methods Phys. Res. B 261 (1) (2007) 965–968. [104] O. Kester, D. Bazin, C. Benatti, J. Bierwagen, G. Bollen, S. Bricker, A. Crawford, S. Chouhan, C. Compton, K. Davidson, et al., Rea3–the rare isotope reaccelerator at msu, LINAC2010, Tsukuba, Japan. [105] A. Laird, P. Amaudruz, L. Buchmann, S. Fox, B. Fulton, D. Gigliotti, T. Kirchner, P. Mumby-Croft, R. Openshaw, M. Pavan, et al., Nucl. Instrum. Methods Phys. Res. A 573 (1) (2007) 306–309. [106] S. Wanajo, Y. Ishimaru, Nucl. Phys. A 777 (2006) 676–699. [107] T. Hashimoto, T. Ishikawa, T. Kawamura, K. Nakai, H. Ishiyama, Y. Watanabe, H. Miyatake, M. Tanaka, N. Yoshikawa, S. Jeong, et al., Nucl. Phys. A 746 (2004) 330–334. [108] F. Sauli, L. Ropelewski, P. Everaerts, Nucl. Instrum. Methods Phys. Res. A 560 (2) (2006) 269–277. http://dx.doi.org/10.1016/j.nima.2005.12.239. URL http://www.sciencedirect.com/science/article/pii/S0168900205026781. [109] S.P. Fox, P.A. Amaudruz, P. Bruskiewich, L. Buchmann, K.A. Chipps, U. Hager, A.M. Laird, L. Martin, G. Ruprecht, A.C. Shotter, P. Walden, J. Phys. Conf. Ser. 312 (5) (2011) 052007. http://dx.doi.org/10.1088/1742-6596/312/5/052007. URL http://stacks.iop.org/1742-6596/312/i=5/a=052007. [110] M. Matos, J. Blackmon, H. Gardiner, L. Linhardt, K. Macon, L. Mondello, L. Baby, E. Johnson, E. Koshchiy, G. Rogachev, et al., The Array for Nuclear Astrophysics Studies with Exotic Nuclei (Anasen), in: Capture Gamma-Ray Spectroscopy and Related Topics, vol. 1, 2013, pp. 481–486. [111] A. Kutchera, Clustering phenomena in the a = 10, t = 1, isobaric multiplet (Ph.D. dissertation). [112] G. Engel, N. Duggireddi, V. Vangapally, J. Elson, L. Sobotka, R. Charity, Nucl. Instrum. Methods Phys. Res. A 652 (1) (2011) 462–465. [113] http://www.physics.fsu.edu/nuclear/foxlab/nuclear-foxlab.html. [114] J.C. Blackmon, Private Communication. [115] A. Gade, T. Glasmacher, Prog. Part. Nucl. Phys. 60 (1) (2008) 161–224. [116] T. Otsuka, T. Suzuki, R. Fujimoto, H. Grawe, Y. Akaishi, Phys. Rev. Lett. 95 (23) (2005) 232502. [117] B. Brown, B. Wildenthal, Annu. Rev. Nucl. Part. Sci. 38 (1) (1988) 29–66. [118] S. Takeuchi, et al., RIKEN Accel. Prog. Rep. 36 (2002) 148–149. [119] F. Recchia, Installation of the Agata Demonstrator and Commissioning Experiments at lnl, in: Journal of Physics: Conference Series, vol. 205, IOP Publishing, 2010, p. 012045. [120] I. Lee, R. Clark, M. Cromaz, M. Deleplanque, M. Descovich, R. Diamond, P. Fallon, A. Macchiavelli, F. Stephens, D. Ward, Nucl. Phys. A 746 (2004) 255–259. [121] M. Pfützner, M. Karny, L. Grigorenko, K. Riisager, Rev. Modern Phys. 84 (2) (2012) 567. [122] K. Miernik, W. Dominik, Z. Janas, M. Pfützner, C. Bingham, H. Czyrkowski, M. Ćwiok, I. Darby, R. Dabrowski, T. Ginter, et al., Phys. Rev. C 76 (4) (2007) 041304. [123] W. Zimmerman, M. Ahmed, B. Bromberger, S. Stave, A. Breskin, V. Dangendorf, T. Delbar, M. Gai, S. Henshaw, J. Mueller, et al., Phys. Rev. Lett. 110 (15) (2013) 152502. [124] H.R. Weller, M.W. Ahmed, H. Gao, W. Tornow, Y.K. Wu, M. Gai, R. Miskimen, Prog. Part. Nucl. Phys. 62 (1) (2009) 257–303. http://dx.doi.org/10.1016/j.ppnp.2008.07.001. URL http://www.sciencedirect.com/science/article/pii/S0146641008000434. [125] F. Hoyle, D. Dunbar, W. Wenzel, et al., Phys. Rev. 92 (1953) 1095. [126] F. Sauli, Gaseous Radiation Detectors: Fundamentals and Applications, Vol. 36, Cambridge University Press, 2014. [127] W. Blum, W. Riegler, L. Rolandi, Particle Detection with Drift Chambers, Springer Science & Business Media, 2008. [128] W.R. Leo, Techniques for Nuclear and Particle Physics Experiments: A How-to Approach, Springer Science & Business Media, 2012. [129] H. Hilke, Rep. Progr. Phys. 73 (11) (2010) 116201. [130] A. Peisert, F. Sauli, Drift and Diffusion of Electrons in Gases: a Compilation, Tech. Rep., European Organization for Nuclear Research, Geneva (Switzerland), 1984. [131] S. Fox, P. Amaudruz, P. Bruskiewich, L. Buchmann, K. Chipps, U. Hager, A. Laird, L. Martin, G. Ruprecht, A. Shotter, et al., Tactic: A New Detector for Nuclear Astrophysics Experiments, in: Journal of Physics: Conference Series, vol. 312, IOP Publishing, 2011, p. 052007. [132] J. Pancin, J. Gibelin, M. Goth, P. Gangnant, J.-F. Libin, R. Raabe, T. Roger, P. Roussel-Chomaz, J. Instrum. 7 (01) (2012) P01006. [133] G. Rogachev, M. Avila, A. Kuchera, L. Baby, J. Belarge, J. Blackmon, V. Goldberg, E. Johnson, K. Kemper, E. Koshchiy, et al., Clustering in Non-SelfConjugate Nuclei 10be and 18o, in: Journal of Physics: Conference Series, vol. 569, IOP Publishing, 2014, p. 012004.

S. Beceiro-Novo et al. / Progress in Particle and Nuclear Physics 84 (2015) 124–165

165

[134] M. Anderson, J. Berkovitz, W. Betts, R. Bossingham, F. Bieser, R. Brown, M. Burks, M.C. de la Barca Sánchez, D. Cebra, M. Cherney, et al., Nucl. Instrum. Methods Phys. Res. A 499 (2) (2003) 659–678. [135] P. Baron, D. Besin, D. Calvet, C. Coquelet, X. De La Broise, E. Delagnes, F. Druillole, A. Le Coguie, E. Monmarthe, E. Zonca, Architecture and implementation of the front-end electronics of the time projection chambers in the t2k experiment, in: Real Time Conference, 2009. RT’09. 16th IEEE-NPSS, IEEE, 2009, pp. 43–48. [136] M. Anderson, F. Bieser, R. Bossingham, D. Cebra, E. Hjort, S. Klein, S. Kleinfelder, C. Vu, H. Wieman, Nucl. Instrum. Methods Phys. Res. A 499 (2) (2003) 679–691. [137] M. Dixit, A. Rankin, Nucl. Instrum. Methods Phys. Res. A 566 (2) (2006) 281–285. [138] R.E. Kalman, J. FluidsJ. Fluids Eng. 82 (1) (1960) 35–45. [139] C. Höppner, S. Neubert, B. Ketzer, S. Paul, Nucl. Instrum. Methods Phys. Res. A 620 (2) (2010) 518–525. [140] P. Hough, Method and means for recognising complex patterns., us patent 3069654, 1962, Appendix i Development of Equation 5. [141] C. Cheshkov, Nucl. Instrum. Methods Phys. Res. A 566 (1) (2006) 35–39. [142] http://www.frib.msu.edu. [143] https://www.gsi.de/en/research/fair.htm.