A simple decay-spectroscopy station at CRIS-ISOLDE

A new decay-spectroscopy station (DSS2.0) has been designed by the CRIS collaboration for use at the radioactive ion beam facility, ISOLDE. With the design optimised for both charged-particle and γ-ray detection, the DDS2.0 allows high-efficiency decay spectroscopy to be performed. The DSS2.0 complements the existing decay-spectroscopy system at the CRIS experiment, and together provide the ability to perform laser-assisted nuclear decay spectroscopy on both ground state and long-lived isomeric species. This paper describes the new decay-spectroscopy station and presents the characterisation studies that have recently been performed.


Introduction
The CERN-ISOLDE facility [1,2] provides beams of radioactive isotopes, ranging from 2 He to 92 U, to a variety of experimental setups. Many of these experiments perform nuclear decay spectroscopy, either as a direct measurement or for identification of the isotope of interest.
The collinear resonance ionization spectroscopy (CRIS) experiment [3][4][5] at ISOLDE is a laser spectroscopy setup with the ability to perform nuclear decay spectroscopy in parallel. By combining the high resolution of collinear laser spectroscopy with the high efficiency of resonance ionization and ion detection [6], high-resolution, highsensitivity studies of hyperfine structures and isotope shifts can be performed [7][8][9]. For hyperfine-structure studies, the frequency of the laser is scanned and the ion rate is measured as a function of applied frequency. With the added capability of studying the radioactive decay of the isotope under investigation, the CRIS experiment has the ability to identify complex hyperfine structures, and perform dedicated laserassisted nuclear decay spectroscopy.
When an isotope has a long-lived isomer (T ≥ 5 1/2 ms), its hyperfine structure and isomer shift can be studied alongside the ground state. In such cases, when each state's hyperfine structure cannot be identified with the usual methods of ion-counting (resonance peak location, separation, intensity, etc.), additional selection is necessary. Identification of the resonances can be achieved by studying the nuclear decay of the isotope of interest [10]. In a recent series of campaigns, the isomers in 202,204,206,214 Fr were studied via their characteristic α-particle energies [11][12][13], and the state observed in 218 Fr via its characteristic half-life [14]. Conversely, the high resolution of the CRIS experiment can be used to separate isotopes from isobaric contamination, and even isomers from each other. This was also successfully demonstrated and allowed the first measurement of the branching ratios in the decay of 204 m2 Fr [11] and 206 m2 Fr [12].
These campaigns were performed using the decay spectroscopy station DSS [15], which has been tailored to maintain the high vacuum required by the CRIS experiment, at the expense of γ-ray detection efficiency (<0.01% at 1 MeV in the aforementioned campaigns). Moreover, the limited number of implantation sites in the DSS (of 8 carbon foils) prevents the complete removal of long-lived activity, as would be achieved with a tape system.
In this article, we present the design of a new decay spectroscopy station, with the primary focus of increased γ-ray detection efficiency and versatility, alongside its offline-characterisation studies.

DSS2.0 Design
The new decay-spectroscopy station (DSS2.0) consists of a rectangular, aluminium chamber (dimensions: 10 × 10 × 25 cm) that can be mounted onto a DN100 vacuum flange. Three of the four side panels are braised, thin (2 mm) aluminium walls for minimising γ-ray absorption. The fourth panel is a base plate on which the sample holder, in-vacuum detectors, and associated connectors are placed. The back panel is a removable aluminium plate, which allows the future installation of a beryllium window or ancillary detectors, depending upon the physics case of interest. Two silicon detectors are mounted on the base plate for charged-particle spectroscopy, either side of an implantation site. A stainless steel mount holding two carbon foils and an insulated copper plate (used as a Faraday cup) is typically used as the implantation site. A Kapton ® shielded wire is soldered onto the copper plate and connected to a LEMO ® connection, allowing an electrical current to be measured and the plate to be used as a beammonitoring device. The Faraday cup is electrically insulated from the surrounding stainless steel by PEEK ® rings either side of the copper plate. Exchange of the mount for an aluminised Mylar ® tape is possible, with no modification to the base plate configuration necessary (see Fig. 1).

For α-decay spectroscopy
Two Canberra PIPS detectors are installed in the DSS2.0 for charged-particle detection, either side of a stainless steel mount containing two carbon foils and a Faraday cup. An annular PIPS (APIPS) detector (model: BKANPD 300-18 RM, thickness: 300 µm, active area: 300 mm 2 , hole diameter: 4 mm) is placed in front of the carbon-foil holder, and a standard PIPS detector (model: BKA 300-17 AM, thickness: 300 µm, active area: 300 mm 2 ) is installed behind. The detectors are connected to Canberra preamplifiers (model: 2003BT) via custom-made Microdot cables and BNC feedthroughs. Installed in the stainless-steel mount (dimensions: 12 × 10 × 2 mm 3 ) are two carbon foils (thickness: 20(1)μg cm −2 or 90 nm, diameter: 10 mm [16]) and a copper Faraday cup (thickness: 0.5 mm, diameter: 10 mm). Each carbon foil is attached to one side of a 1-mm thick copper o-ring, to aid manipulation. Each carbon-foil assembly is inserted into a circular groove in the stainless-steel mount, to fix in place. This mount is attached to a Kurt J. Lesker linear actuator (model: KLPDBAP) that allows its position to be varied to place either the carbon foils or Faraday cup in between the PIPS detectors (see Fig. 2).
The Faraday cup is used for tuning the beam through the aperture of the annular PIPS detector. For nuclear decay measurements, the radioactive ion beam is focused through the aperture of the annular PIPS detector and implanted into one of the carbon foils. Radioactive decay products (α or β particles, fission fragments, conversion electrons) emitted from the implanted isotopes are measured with the surrounding PIPS detectors.
GEANT4 [17] simulations were performed to determine the most appropriate orientation of the carbon foil and the optimal position of the annular PIPS detector with respect to the carbon foil holder (see Fig. 3). The α-particle source was spread uniformly across the carbon foil in the area that may be reached through the 4 mm hole in the APIPS detector. The distance between the PIPS detectors and the stainless steel mount was varied, and the position of the carbon foil flipped between facing closest to the annular PIPS (position 0) or the standard PIPS (position 1), see Fig. 4. This was achieved by overturning the copper o-ring that the carbon foil was attached to, inside the stainless steel mount. The PIPS detector is placed in close geometry to the carbon foil holder (between 0.4 and 1.6 mm), represented by the width of the curves in Fig. 3. It was concluded that the highest efficiency is obtained with the foil in position 1, facing closest to the standard PIPS, which is positioned at 0.4 mm from the carbon foil holder. The annular PIPS detector is mounted at an optimal distance of 1.5 mm from the carbon foil holder, resulting in an expected total solid angle of 65%.
SRIM [18] simulations were performed to investigate the depth of implantation of the ion beam into the carbon foil, the energy loss of the α particle upon exiting the foil, and the recoil of the daughter nucleus. A normally incident beam of 206 Fr + at 40 keV was chosen as a case study. The implantation profile can be seen in Fig. 5 (top). The implantation depth was found to be 21 nm, with a full width at half maximum of 7 nm.
This initial distribution of 206 Fr ions (21 nm deep in the carbon foil,   7 nm FWHM) was used as a starting sample for the α decay. An isotropic decay pattern was randomly generated from this sample, with α particles of energy 6792 keV. The trajectories of these α particles were then simulated through the carbon foil and extrapolated in order to determine the fraction of particles reaching the PIPS detectors. A total efficiency of 65% is found, in agreement with the GEANT4 simulations. The degraded α-particle energies are plotted in Fig. 6. Four αparticle energy spectra are shown: one for the APIPS detector and three for the PIPS detector when the 40-keV 206 Fr ions are implanted into a carbon foil of 90 nm, 200 nm and 500 nm, respectively. For each spectrum, the 6792-keV α-particle energy peak is displaced to a lower energy. Additionally, a low-energy tail, corresponding to the particles traveling with a large angle through the carbon foil, is present, which becomes more pronounced for thicker carbon foils. The line profiles were fit with a Crystal Ball function, which combines a Gaussian with a power-law low-energy tail [19,20], to reproduce the energy loss effects. For an intrinsic α-particle energy resolution of 30 keV, the energy straggling contributes less than 1 keV to the FWHM for the APIPS detector. For the PIPS detector, the degradation of the α-particle energy contributes 1 keV, 2 keV, and 8 keV to the energy resolution, for carbon foil thickness of 90 nm, 200 nm and 500 nm, respectively.
The distribution of the recoiling daughter nuclei 202 At (after emission of the 6792-keV α particle) were also investigated with SRIM, using the same distribution as that of the α-particle study. The recoil energy is 135 keV, which is greater than the 40 keV implantation energy, and a fraction of the recoils are accordingly ejected from the carbon foil. This effect was already identified, and accounted for, in studies with similar setups [21]. The distribution of the recoils remaining in the carbon foil can be seen in Fig. 5 (Bottom). The fraction of recoils remaining in the foil is 71%. Of the ejected recoils, a fraction is deposited on the surface of the annular silicon detector, amounting to 10% of all the recoils.
The thickness of the carbon foil is a compromise between beamimplantation depth, α-particle energy straggling and ease of use. At 90 nm, the carbon foils are thick enough to stop not only the incoming ion beam (with a nominal energy of 30 or 40 keV), but at least one αdecay particle recoil. Conversely, the foils are thin enough to let the emitted α particle through, to be detected by the PIPS detector, with minimal α-particle energy straggling.

For β-decay spectroscopy
When performing β-decay spectroscopy, the spectra become highly contaminated in time as isobars and daughter nuclei with longer halflives than the beam of interest accumulate at the measurement point. A traditional approach to minimise the impact of this contamination is to regularly remove the sample until it is completely out of sight of the detectors. The stainless steel mount with only two carbon-foil implantation sites (presented for α-decay spectroscopy) is therefore not appropriate, and a different geometry is required.
For a dedicated β-decay spectroscopy experiment, it is possible to   remove the linear-actuator assembly from the chamber and replace with the ISOLTRAP tape system [22,23]. The Mylar ® tape system allows the complete removal of accumulated β-emitting radioactivity.
With over 1 km of tape installed in the system, there are more than 1000 individual implantation sites available before the tape is reused. Additionally, the ISOLTRAP tape system is equipped with a secondary charged-particle and γ-ray detection point where the removed activity can be further studied.

For γ-ray spectroscopy
Up to three germanium detectors can be placed around the DSS2.0 chamber to detect the x and γ rays emitted from the implantation site. The back panel can also be changed for a beryllium window for the study of low-energy x and γ rays. In this configuration, the PIPS detector behind the implantation point is removed. This port may also be used to insert an ancillary detector, e.g. silicon lithium drifted detector (SiLi) for the high-resolution study of conversion electrons.

Beam tuning
For stable-beam tuning, the ion beam is focused through the 4 mm aperture of the annular PIPS detector and impinged upon the Faraday cup. A collimator with a 4 mm aperture is placed in front of the annular detector to avoid implanting beam directly onto the detector.
Additionally, the current of beam impinging upon the collimator upstream of the annular detector allows the fraction of the beam that does not pass within the aperture to be measured. A second electrical connection is made with Kapton ® shielded wire from the steel collimator to another LEMO ® connection. Highest transmission is achieved by maximising the signal on the Faraday cup and minimising it on the collimator.

Data acquisition
The data acquisition system used for the DSS2.0 characterisation comprises of a VME-based 8-channel digitiser (CAEN V1724) with a sampling rate of 100 MHz. Each detector is connected to a channel and the pre-amplified signal is analysed via on-board digital pulse processing for pulse height analysis (DPP-PHA). Each event is also attributed a time stamp for the post-processing of the time behavior of the signal and reconstructing coincidences between different detectors. The parameters for the DPP-PHA are tuned for each detector independently to offer the highest performance.
The data is acquired using Daresbury's Multi Instance Data Acquisition System (MIDAS) [24] and analyzed with a bespoke conversion and analysis code [25].

Offline characterisation
The performance of the DSS2.0 has been characterised off-line using sources of long-lived radio-isotopes.

α-Decay spectroscopy
The performance characteristics to detect α-decay were studied with an α source ( α 4 ) containing 1 kBq of four radioactive isotopes ( 148 Gd, 239 Pu, 241 Am, and 244 Cm) on a copper disk. Since the α particles can only be emitted in one direction, only the PIPS detector was installed in the chamber. An energy resolution of 30 keV at 5.5 MeV was achieved for the PIPS detector. This is sufficient to resolve the α-particle energies of the isotopes of interest in future studies. No absolute efficiency measurements could be made however, due to the geometry of the α 4 source and the difficulty placing it in the correct position of the carbon foil.

γ-Ray spectroscopy
The γ-decay performances were studied with single-isotope sources of 60 Co, 133 Ba, and 152 Eu, with respective activities 723 Bq, 11.6 kBq, Fig. 7. γ-ray efficiency curve for a single-crystal Canberra XtRa germanium detector, measured using 60 Co (723 Bq), 133 Ba (11.6 kBq) and 152 Eu (16.3 kBq).   9. Random-subtracted coincidence matrix between the silicon and germanium detectors using the α 4 source(1 kBq of 148 Gd, 239 Pu, 241 Am, 244 Cm). The high concentration of points corresponds to the α-particle emission (5486 keV) of 241 Am, followed by the 60-keV γ-ray decay. and 16.3 kBq. A Canberra extended range (XtRa) coaxial germanium detector (model: GX6020 ) was installed directly outside the chamber to record the emitted γ rays. An energy resolution of 3.5 keV at 1 MeV was interpolated from a fit of FWHM with respect to γ-ray energy. Fig. 7 shows the photo-peak γ-ray detection efficiency of the Canberra XtRa germanium detector. A peak efficiency of 2.7% at 100 keV compares favourably to the detection efficiency of the original DSS chamber of 0.007% [15].
The benefit of the increased γ-ray detection efficiency is illustrated with the example of the 531 keV γ-ray decay of 206 m2 Fr, assuming standard ISOLDE yields. At this energy, the original DSS has a γ-ray detection efficiency of 0.005%, resulting in a necessary measurement time of 1 h 57 min for an integrated number of 10,000 γ-ray events. For the same number of integrated counts, the DSS2.0, with a detection efficiency of 1.3% at 531 keV, would take only 27 s

Charged-particle-γ coincidences
Using the α 4 source, it was possible to study the coincidences between α decays and γ rays using the decay of 241 Am. The time difference between the silicon and germanium detector signals is shown in Fig. 8. A prompt peak is identified for time differences of t Δ < 150 ns, followed by random events.
Using the timing selection criterion of t Δ < 150 ns, a two-dimensional matrix was created, showing the energy dependence of the events according to both the α particle and the γ ray (see Fig. 9). From this coincidence matrix, a time component (10% of events that occurred in the window t 150 ns < Δ < 1650 ns) was subtracted to account for random events. A high concentration of points is identified around E = 5500 keV α and E = 50 keV γ , corresponding to the α-particle emission (5486 keV) of 241 Am, followed by the 60-keV γ-ray decay [26]. The efficiency of the α γ − -coincidences were found to be selfconsistent with the measured efficiency of α-particle and γ-ray detection.

Summary
A new decay spectroscopy station, named DSS2.0, has been designed and built for the CRIS experiment at ISOLDE. Its main purpose is the study of charged-particle and γ decays at CRIS and is therefore made in a compact geometry with thin aluminium walls.
The DSS2.0 has been characterised offline. Radioactive-decay energy spectra have been acquired with long-lived radioactive sources for the silicon PIPS and germanium detectors, and coincidences between the detectors investigated. The resolution and efficiency of the PIPS detectors are similar to the original decay spectroscopy station at CRIS. The γ-ray detection efficiency is nearly a factor of 400 better than the DSS. The performance of the DSS2.0 is found to be appropriate for the intended future studies.
While the DSS2.0 was initially intended for performing laserassisted nuclear decay spectroscopy at the CRIS experiment, the small size and portable nature of the device allows its installation at several experimental setups at ISOLDE. A commissioning experiment studying the α decay of the light mercury isotopes has been performed with the DSS2.0 at the GLM beamline at ISOLDE [27]. The chamber has also been utilised during an ISOLTRAP experiment studying the polonium isotopes. The DSS2.0 was installed after the ISOLTRAP multi-reflection time-of-flight mass separator (MR-TOF-MS) [28,29], where it was used to measure the ratio of ground-to isomeric-state in the odd-A neutron-deficient polonium beams.