Measurement of the 241 Am(n, γ ) cross section at the n_TOF facility at CERN

. The neutron capture cross section of 241 Am is an important quantity for nuclear energy production and fuel cycle scenarios. Several measurements have been performed in recent years with the aim to reduce existing uncertainties in evaluated data. Two previous measurements, performed at the 185m flight-path station EAR1 of the neutron time-of-flight facility n_TOF at CERN, have permitted to substantially extend the resolved resonance region, but su ff ered in the near-thermal energy range from the unfavorable signal-to-background ratio resulting from the combination of the high radioactivity of 241 Am and the rather low thermal neutron flux. The here presented 241 Am(n, γ ) measurement, performed with C 6 D 6 liquid scintillator gamma detectors at the 20m flight-path station EAR2 of the n_TOF facility, took advantage of the much higher neutron flux. The current status of the analysis of the data, focussed on the low-energy region, will be described here.


Introduction
The cross section of the 241 Am(n,γ) reaction is an important quantity for a number of nuclear technology applications. In recent years several efforts on evalutions and measurements have been undertaken worldwide to improve the knowledge of the 241 Am(n,γ) cross section, in particular in the thermal and epithermal energy region. The 241 Am(n,γ) cross section is listed in the NEA High Priority Request List [1,2]. Work by the WPEC subgroup 41 [3] and a critical review of measured thermal cross sections [4] are examples of evaluations. As for time-offlight measurements, previously two capture experiments, one with C 6 D 6 detectors [5] and one with BaF 2 [6], have been performed at the long 185 m flight path EAR1 of the n_TOF facility. For both experiments the same sample was used. Those measurements have substantially increased the resolved energy region to 320 eV while evaluations previously went only up to 150 eV. The thermal neutron flux at this long flight path is rather low because of the use of borated water, strongly suppressing the gamma-ray background from neutron capture in hydrogen, but at the same time removing slow neutrons from the beam by capture in boron.
The two capture measurements [5,6] suffer from a large systematic uncertainty at low neutron energy caused by the large background component due to the radioactivity of 241 Am, in spite of the use of a 2 mm lead screen. The gamma-ray background at thermal energies was about 90% of the signal. After corrections, both measurements were found to be inconsistent at low neutron energies. To resolve this discrepancy, a new capture measurement with C 6 D 6 detectors was performed using the same sample, but now at the new 20 m vertical flight path EAR2. Because of the shorter flight path, the thermal flux per unit of time of flight is about a factor 10 higher. In addition, the neutrons in the vertical direction are moderated by normal water, resulting in a supplementary factor of about 40 for the thermal flux relative to the long flight path [7]. The 20 m flight path EAR2 therefore seems to be well adapted for capture measurements in the thermal and epithermal region, where the focus of the present experiment lies. Time-offlight measurements covering the full thermal region are a welcome complement to integral techniques typically used at reactors in this energy region. Note that the spallation target in use for this experiment was still the second generation target and not the recent upgraded target [8] with optimized performances. Special care has to be taken to extract the correct resolution function from the simulated data [9].

Experimental setup and preliminary spectra
The present experiment measuring data to extract the 241 Am(n,γ) cross section in the thermal region and the first few resonances at low energy has been carried out at the EAR2 station of the n_TOF facility at CERN [10]. Three neutron-insensitive C 6 D 6 liquid scintillator detectors have been used to measure gamma rays following neutron capture as a function of the neutron time of flight. Proton pulses with a nominal intensity of 7 × 10 12 protons per pulse at a minimum repetition rate of 1.2 s generated neutron by hitting the 40 cm thick and 60 cm diameter cylindrical lead target. The water cooling circuit served as a moderator resulting in a neutron beam with energies from thermal up to a few GeV. The sample with a mass of 32.23 ± 0.19 mg of Am and an activity of 4.1 GBq, the same as used in the previous experiments, consists of americium oxide ( 241 AmO 2 ), infiltrated and immobilized in a pressed pellet of aluminum oxide (Al 2 O 3 ) forming a rigid disk of 12.26 mm diameter, encapsulated in a sealed aluminum container. A second, similar sample with a mass of 40.98 ± 0.25 mg was also used, as well as a dummy sample consisting of a similar aluminum oxide pellet but without americium, used for background measurements. In addition to those samples we also measured several other samples in the same configuration inside an identical aluminum container, among which 197 Au and 103 Rh for the absolute normalization using their low-energy saturated resonances.
The distance from the sample to the detectors of about 40 cm was chosen such that the high radioactivity did not notably impact the bias current of the photomultipliers. We could therefore avoid the need for a lead shielding which allowed for the use a lower gamma-ray threshold in the data analysis, in particular the pulse height weighting function as used with the total energy method.
The setup of the three detectors including the sample inside a sample holder were modelled in GEANT4 in order to simulate the gamma-ray response to mono-energetic gamma-rays. Those responses are needed to calculate the pulse height weighting function, which makes the detector efficiency of the gamma-ray cascade independent of the gamma-ray energy [11,12]. The setup is shown in figure 1.
Upstream of the sample, the SiMon2 detector [13], consisting of a thin in-beam 6 Li foil and four off-beam sil- icon detectors, was installed to measure the neutron flux at the same time. This detector was also an important ingredient of the evaluated neutron flux typically used for experiments [7].
The used data acquisition system was based on Teledyne SP Devices digitizers, sampling the detector waveforms with steps of 1 ns and a resolution of 12 bits for a time window of 100 ms after each TOF pulse. The waveforms, after zero suppression, were transferred to CERN's long-term storage facility for off-line event building. Detector events, consisting of the time of flight and the pulse height extracted from the waveforms, were constructed for each detector and used for further analysis.
Energy calibrations of the C 6 D 6 detectors were done without beam using standard gamma-ray sources 137 Cs, 88 Y, 60 Co and a mixed Am-Be source. The spectra from the sources showed gain shifts over time during the measurement period. We therefore used a continuous calibration on a run-by-run basis using either the radioactivity of   the 241 Am sample, or the response to the ambient background, in particular the 1.460 MeV gamma ray from 40 K, present in the concrete of the walls of the experimental area. The radioactive background from the 241 Am sample mainly originates from the strong 59.5 keV gamma ray of 237 Np following the α-decay of 241 Am. Smaller background gamma rays come from α-induced reactions, mainly on 27 Al present in the aluminum oxide sample pellet.

= " > A A A C G 3 i c b V D L S s N A F J 3 4 r P U V d e l m s A g t 0 p K U o m 6 E 2 m 5 c V r A P a G q Y T C b t 0 M m D m Y m k h P 6 H G 3 / F j Q t F X A k u / B u n j 4 W 2 H r h w O O d e 7 r 3 H i R g V 0 j C + t Z X V t f W N z c x W d n t n d 2 9 f P z h s i T D m m D R x y E L e c Z A g j A a k K a l k p B N x g n y
The ambient background spectrum shown in figure 2 was measured with a high purity germanium detector and clearly shows a typical natural background including the 40 K peak and other smaller peaks due to the 232 Th and 238 U decay series. This response, since it is rather weak, becomes only visible when the neutrons from the beam are practically absent. This occurs at very long times of flight corresponding to an equivalent neutron energy below 1 meV given the energy distribution of the neutron beam. We used the last 50 ms of the recorded part of each time-of-flight cycle to obtain the spectrum measured by the C 6 D 6 detectors in which the Compton edge of the 40 K peak is fairly visible. We fitted the shape of the spectra with a logistic-based sigmoid function on top of a smooth background for each run with nonradioactive samples. From the fitted sigmoid parameters we deduced the channel-energy calibration factor of the deposited energy spectra. An example of such a pulse height spectrum measured with the C 6 D 6 detectors and the fit in the vicinity of the Compton edge is shown in figure 3.
In figure 4 we show the low-energy raw counting spectra for the 241 Am sample, together with a spectrum of the dummy sample and that of a 197 Au and 103 Rh sample. The spectra are measured as time of flight and here represented as a function of approximate energy. The Au and Rh spectra each show a saturated resonance which serves as normalization of the capture yield. In all four spectra the shape of the thermal peak is clearly visible.

Conclusion
The here presented experiment shows the capability of n_TOF EAR2 to peform neutron capture measurements in the low energy range, covering at the same time both the full thermal region and resonances. The measured nucleus 241 Am has a very high radioactivity hindering neutron capture time-of-flight measurements. In the described setup we used the high instantaneous flux of the EAR2 station of the n_TOF facility for a favorable signal to noise ratio, in combination with an increased distance between detector and sample to overcome the strong influence of the radioactivity on the photomultiplier bias. In this way there was no need for additional detector shielding and the total energy method could be employed with a low gammaray threshold. Gain shift issues were eventually solved by a run-by-run calibration using 40 K present in the natural background. A full re-analysis of the data with this much improved energy calibration is currently ongoing.