betaFeedback

WorksPhysics and Astronomy

Sub-keV energy calibration of CONUS+ via 71Ge M-shell neutron activation

Abstract

The CONUS+ experiment has recently reported the first detection of coherent elastic neutrino-nucleus scattering (CEvNS) of reactor antineutrinos on germanium nuclei and is now entering a precision phase. The dominant uncertainty in the first measurement was the energy scale, which contributed 14% to the uncertainty of the prediction of the combined signal. We present a dedicated neutron activation campaign in which one of the new 2.4 kg CONUS+ germanium detectors was irradiated with a strong 241AmBe source, demonstrating that a contribution below 4% to the uncertainty of signal prediction is achievable. For the first time, the 71Ge M-shell X-ray line was clearly resolved at (158.7+-1.4) eVee, validating the CONUS+ energy reconstruction down to the detection threshold. This validation includes the understanding of the energy scale, the energy resolution, the trigger efficiency, and the correct separation of physical from noise events. These results establish the foundation for a future activation campaign at the Kernkraftwerk Leibstadt reactor site, strengthening the CONUS+ energy calibration and extending its sensitivity to precision CEvNS and beyond Standard Model physics measurements.

Introduction

Coherent Elastic Neutrino-Nucleus Scattering (CEν\nuNS) is a Standard Model process in which neutrinos interact coherently with an entire nucleus via Z boson exchange. The scattering amplitudes from all nucleons add coherently, enhancing the cross-section by several orders of magnitude compared to other neutrino detection channels, such as inverse beta decay or elastic neutrino-electron scattering. Furthermore, the interaction is thresholdless because no charged leptons need to be produced. These properties make CEν\nuNS a particularly appealing channel for reactor monitoring and for probing physics beyond the Standard Model (BSM) [1, 2]. There is currently a broad worldwide effort to detect CEν\nuNS at reactors [3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14].

CEν\nuNS was theoretically proposed in 1974 [15, 16], yet more than four decades elapsed before its first experimental detection by the COHERENT collaboration [17, 18, 19], using neutrinos produced with a pion decay at rest source. The primary experimental challenge lies in the signature of the interaction: a nuclear recoil of only a few keV, requiring detectors with exceptionally low energy thresholds and backgrounds. More recently, CEν\nuNS has been observed with 8B^{8}\mathrm{B} solar neutrinos [20, 21, 22], and with reactor antineutrinos by the CONUS+ experiment [23]. A previous claim of reactor CEν\nuNS detection [5] remains in tension with the results of [8, 9, 14], underscoring the experimental difficulty of this measurement. Following the first CEν\nuNS detection, CONUS+ now enters a precision phase. To this end, three of the original 1 kg high-purity germanium (HPGe) detectors have been replaced by new ones with 2.4 kg diodes, significantly increasing the target mass and thus the signal rate.

The dominant uncertainty in the signal prediction in the first measurement was the energy scale, for which a conservative value of ±5\pm5 eVee_{\mathrm{ee}} (electron equivalent) was adopted. This resulted in an uncertainty of 14% in the prediction of the combined signal, nearly twice the impact of the next largest contribution, the quenching factor, with an impact at the level of 7%. Figure 1 shows the signal prediction uncertainty as a function of the energy scale uncertainty, evaluated at a reference energy of 150 eVee_{\mathrm{ee}}, for three realistic energy thresholds. The threshold dependence reflects the interplay between trigger efficiency and quenching. Reducing the energy scale uncertainty to ±3\pm3 eVee_{\mathrm{ee}} would render this contribution subdominant at the 4% level. Achieving a sub-percent impact on the signal prediction requires reaching an energy scale uncertainty below ±1\pm1 eVee_{\mathrm{ee}}.

Signal prediction uncertainty as a function of the energy scale uncertainty

Figure 1. Signal prediction uncertainty as a function of the energy scale uncertainty, evaluated at an energy reference of 150 eVee_{\mathrm{ee}} for three representative energy thresholds.

External sources emitting γ\gamma-rays cannot be used for low energy calibration in CONUS+, as the emitted low energy γ\gamma-rays are absorbed in the end cap of the few-millimeter-thick copper cryostat and the dead layer of the detector. The calibration therefore relies on X-ray lines produced directly within the germanium crystal through cosmogenic and neutron-induced activation. During detector fabrication and transport above ground, cosmic-ray spallation produces 68^{68}Ge (half-life 270.95 days), 65^{65}Zn (half-life 244.01 days) and 68^{68}Ga (half-life 67.7 minutes). The latter is continuously produced by the decay of 68^{68}Ge. The 68^{68}Ge isotope is produced by spallation reactions on 70^{70}Ge, which require fast neutrons of at least ∼20\sim20 MeV [24]:

70Ge+nfast→68Ge+3n(1){}^{70}\mathrm{Ge} + n_{\mathrm{fast}} \to{}^{68}\mathrm{Ge} + 3n \tag*{(1)}

Similarly, 65^{65}Zn is produced by spallation reactions on the surface of the Earth [24]:

70Ge+nfast→65Zn+2n+4He(2){}^{70}\mathrm{Ge} + n_{\mathrm{fast}} \to{}^{65}\mathrm{Zn} + 2n + {}^{4}\mathrm{He} \tag*{(2)}

In addition to these cosmogenic isotopes, 71^{71}Ge is continuously produced in situ within the detector via neutron capture on 70^{70}Ge (natural isotopic abundance 20.5%) by thermal, epithermal and fast neutrons with energies up to a few MeV:

70Ge+n→71Ge→71Ga (stable)(3){}^{70}\mathrm{Ge} + n \to{}^{71}\mathrm{Ge} \to{}^{71}\mathrm{Ga}\ \text{(stable)} \tag*{(3)}

Unlike 68^{68}Ge, which requires fast neutrons for its production through spallation and is therefore only activated above ground, 71^{71}Ge is produced continuously at the CONUS+ experimental site by muon-induced neutrons generated in the lead shielding, leading to a K-shell line count rate of ∼40\sim40 counts kg−1^{-1} day−1^{-1} [23]. It can also be produced in larger quantities using external neutron sources such as 252^{252}Cf or 241^{241}AmBe.

Electron capture (EC) from all these isotopes produces characteristic X-ray lines via atomic de-excitation. The dominant feature in the low-energy range is the K-shell line at 10.367 keVee_{\mathrm{ee}}, emitted with a branching ratio of 0.876 from both 68^{68}Ge and 71^{71}Ge (100% EC). The corresponding L-shell lines appear at 1.298 keVee_{\mathrm{ee}} (L1_1, branching ratio ≈0.119\approx0.119) and 1.143 keVee_{\mathrm{ee}} (L2_2, branching ratio ≈0.001\approx0.001) [25]. Due to the proximity in energy, the L-shell emissions from 68^{68}Ge, 71^{71}Ge, 68^{68}Ga and 65^{65}Zn are observed as a single broad peak even at the high energy resolution of the CONUS+ detectors. Additionally, 68^{68}Ga contributes a K-shell line at 9.659 keVee_{\mathrm{ee}}, at 11.1% of the 68^{68}Ge K-shell intensity, and 65^{65}Zn produces a K-shell line at 8.979 keVee_{\mathrm{ee}}, at 11.4% of the 68^{68}Ge K-shell intensity. The M-shell line at 158.1 eVee_{\mathrm{ee}} is emitted by both 68^{68}Ge and 71^{71}Ge, but has not been resolved in previous CONUS campaigns due to insufficient statistics in the M-shell line. Contributions from copper activation in the cryostat are negligible. Since 68^{68}Ge and 71^{71}Ge produce X-ray lines at identical energies, their contributions are disentangled by exploiting the large difference in their half-lives (270.95 d vs. 11.4645 d) through a time-dependent fit of the K-shell line count rate [26].

The energy calibration of CONUS+ uses the K-shell (10.367 keVee_{\mathrm{ee}}) and L-shell (1.298 keVee_{\mathrm{ee}}) X-ray lines from 68Ge^{68}\mathrm{Ge} and 71Ge^{71}\mathrm{Ge}, with a linear interpolation assumed to extend into the CEν\nuNS region of interest. The linearity of the data acquisition system and electronics has been studied via pulser scans [23], which previously revealed non-linearities at the level of up to 12 eVee_{\mathrm{ee}} in the sub-keV regime. For the second phase, these non-linearities have been mitigated through the optimization of the energy reconstruction parameters, and alternative energy reconstruction algorithms are under development to further improve it.

Figure 2 shows the energy scale uncertainty as a function of exposure, assuming a background rate consistent with CONUS+ operating conditions and a continuous 71Ge^{71}\mathrm{Ge} count rate of 40 counts kg−1^{-1} day−1^{-1}. Using the K- and L-shell lines, the precision floor is set by the uncertainty on the literature values of the peak positions [27], corresponding to a lower limit of approximately ±1.3\pm1.3 eVee_{\mathrm{ee}}. Reaching this floor requires exposures of 4000 kg·days, impractical given that the CONUS+ HPGe detectors have masses between 1 and 2.4 kg. Even achieving a ±3\pm3 eVee_{\mathrm{ee}} uncertainty requires more than 100 kg·days, while sub-2 eVee_{\mathrm{ee}} precision would demand exposures exceeding 400 kg·days. The neutron activation of the detector with external sources allows to enhance the 71Ge^{71}\mathrm{Ge} production rate, in order to reduce the exposure needed. Assuming an enhancement factor of 10 (green line in Figure 2), an energy scale uncertainty below ±2\pm2 eVee_{\mathrm{ee}} can be achieved with only approximately 30 days of exposure, translating to a signal prediction uncertainty contribution below 4%.

Energy scale uncertainty as a function of exposure

Figure 2. Energy scale uncertainty as a function of exposure for the nominal 71Ge^{71}\mathrm{Ge} production rate (blue) and an enhanced rate (green, ×10\times10 factor). The dashed horizontal line indicates the target precision level, which is equivalent to the literature uncertainty on the peak positions [27].

A further motivation for the neutron activation campaign is the enhancement of the M-shell X-ray line at 158.1 eVee_{\mathrm{ee}}. Observing this line serves multiple purposes: it directly validates the energy linearity of CONUS+ across the full CEν\nuNS region of interest, provides an independent constraint on the trigger efficiency at threshold, yields an essential input for the background model, and can be used as a proxy for different noise and background rejection techniques as the pulse shape discrimination [28].

In this work, we report the irradiation of one of the 2.4 kg CONUS+ HPGe detectors with an 241^{241}AmBe neutron source at the Max-Planck-Institut für Kernphysik (MPIK) in Heidelberg, Germany. We present the first clear observation of the 71Ge^{71}\mathrm{Ge} M-shell X-ray line, validating both the activation procedure and the sub-keV detector response. This demonstration establishes the foundation for a future high-statistics activation campaign at the Kernkraftwerk Leibstadt (KKL) reactor site, which will significantly strengthen the CONUS+ energy calibration, extend the sensitivity to BSM physics in the CEν\nuNS spectrum, and provide a robust validation of the trigger efficiency at the lowest detectable energies.

Experimental setup

The measurements were performed at the Low Level Laboratory (LLL) of the MPIK. The LLL provides an overburden of approximately 15 m of water equivalent (m w.e.), consisting of rock, soil and concrete, which significantly suppresses the flux of cosmic-ray-induced neutrons and the hadronic component of cosmic rays [29]. The residual cosmic muon flux at this depth is (65±2)(65 \pm2) muons s−1^{-1}m−2^{-2}, corresponding to a reduction factor of ∼3\sim3 with respect to the surface [29]. The laboratory is equipped with an air conditioning system that regulates temperature and humidity and maintains 222^{222}Rn concentration in the range of (30−70)(30 - 70) Bq m−3^{-3}.

The experimental setup consists of a p-type point-contact (PPC) high-purity germanium (HPGe) detector, named C8, with a crystal mass of 2.4 kg, produced in the same batch as the new detectors installed at the CONUS+ experiment at KKL [30]. The HPGe diode is cooled with a Cryo-Pulse 5 Plus (CP5+) electrically powered pulse tube cooler, upgraded with a liquid-cooled chiller system to suppress mechanical vibrations that would otherwise induce microphonic noise. The detector was operated under stable conditions, with the temperature maintained at (20±1∘C20 \pm1^\circ\mathrm{C}) by a dedicated air cooling system. The detector is equipped with a custom-built charge sensitive preamplifier (CSP) based on an application-specific integrated circuit (ASIC), providing improved noise performance and trigger efficiency at low energies. A pulsed-reset scheme generates rectangular inhibit signals (TRP) to veto spurious events after baseline resets. A full description of the detector and electronics upgrades is provided in [30]. The detector is surrounded by 10 cm of lead on all sides to reduce the environmental γ\gamma-ray background. Additionally, an EJ-200 plastic scintillator 50×50 cm250 \times50\ \mathrm{cm}^{2} plate is placed on top of the shield and operated as an active muon veto. A photograph of the full setup is shown in Figure 3. The data acquisition (DAQ) follows the same design as the CONUS+ experiment [30]. It is based on a 16-bit CAEN V1782 digitizer sampling at 100 MHz managed via the CoMPASS software, with TRP inhibit signals recorded by a separate CAEN V1740D module. An external pulse generator (Keysight 33500B) injects artificial signals of known amplitude directly into the preamplifier, enabling regular monitoring of the energy resolution and trigger efficiency. The achieved pulser resolution in terms of full-width-at-half-maximum (FWHM) is (56±256 \pm2) eVee_{\mathrm{ee}}, and the trigger efficiency above 90% is maintained down to 160 eVee_{\mathrm{ee}}.

Photograph of the C8 detector setup

Figure 3. Photograph of the C8 detector setup at the MPIK Low Level Laboratory. The 2.4 kg HPGe detector is enclosed in a 10 cm lead shield. The upper part of the shield is open with the muon veto plate, which is on top during operation, in front of the shield.

Several data quality selection cuts are applied to the acquired data. A muon veto cut rejects all events within a 170 μ\mus window following a muon trigger in the scintillator. A TRP cut vetoes events within 1000 μ\mus after each preamplifier reset. A cut in the time difference distribution (TDD) rejects events separated by less than 1 ms from the preceding event, suppressing clustered microphonic noise in the low-energy region. Additionally, different quality cuts are applied based on the waveforms to remove pile-up and mis-reconstructed events. The combined dead time from all these contributions is accounted for in the final data analysis.

Energy reconstruction

The signals produced by ionization events in the HPGe detector are processed through a chain of digital filtering steps to reconstruct the deposited energy. Figure 4 shows a representative detector pulse at three stages of this chain. The raw preamplifier output (blue) exhibits an exponential decay tail produced by the AC coupling of the ADC input, with a decay time constant of 10 μ\mus. A pole zero cancellation filter (green) is first applied to flatten this tail, yielding a step-like signal whose amplitude is proportional to the deposited energy. For each event, the baseline is estimated from the pre-trigger region of the waveform and subtracted prior to filtering,

Representative detector pulse at three stages of the energy reconstruction chain

Figure 4. Representative detector pulse at three stages of the energy reconstruction chain: the raw preamplifier output (blue), the pole-zero corrected signal (green), and the trapezoidal filter output (red). The flat-top amplitude of the trapezoidal pulse is used as the energy estimator.

ensuring a stable reference level for the energy reconstruction. A trapezoidal filter (red) is then applied to the baseline-subtracted, pole-zero corrected signal, producing a shaped pulse whose flat-top amplitude is used as the energy estimator. The energy of each event is reconstructed online by the CAEN V1782 digitizer using this trapezoidal filter, implemented at the FPGA level via the CoMPASS software [30].

The choice of the trapezoidal rise time is a critical parameter that determines the noise performance of the energy reconstruction. Figure 5 shows the squared energy resolution (FWHM2^{2}) as a function of the trapezoidal rise time, together with the individual noise contributions. Three components are identified: 1/f1/f noise (orange), which is independent of the rise time and sets a constant floor; series noise (blue dashed), which dominates at short rise times and decreases rapidly with increasing shaping time; and parallel noise (green), which increases linearly with the rise time. The total FWHM2^{2} (red) exhibits a clear minimum at approximately 3 μ\mus, where series and parallel noise are balanced. A trapezoidal rise time of 3.5 μ\mus is therefore adopted for all data presented in this work, consistent with the value used for the CONUS+ detectors at KKL [30]. For the flat top a value of 1 μ\mus is considered.

Squared energy resolution of the C8 detector as a function of the trapezoidal rise time

Figure 5. Squared energy resolution (FWHM2\mathrm{FWHM}^{2}) of the C8 detector as a function of the trapezoidal rise time, measured via pulser scans. The individual noise contributions are shown: 1/f1/f noise (orange dotted), series noise (blue dashed), and parallel noise (green). The total fit (red) exhibits a minimum at ∼\sim3.5 μ\mus, which is adopted as the optimal working point.

The energy scale is established using the K-shell and L-shell X-ray lines from 68Ge^{68}\mathrm{Ge} and 71Ge^{71}\mathrm{Ge}, with a linear calibration function. However, the assumption of a purely linear DAQ response does not hold across the full energy range, in particular at sub-keV energies [23]. The non-linearity is quantified via a pulser scan in the sub-keV range, defining the energy deviation as ΔE=Emeas−Ereal\Delta E = E_{\mathrm{meas}} - E_{\mathrm{real}}, where ErealE_{\mathrm{real}} is derived from the known pulse amplitude and the linear calibration. The C8 detector exhibits a maximum deviation of −12-12 eVee at ∼\sim160 eVee, which is corrected through spline interpolation of the measured non-linearity. The impact of this correction on the M-shell line position is discussed in § 5. For the CONUS+ Run-2 data collection at KKL, the CAEN DAQ settings were optimized, reducing the maximum non-linearity to ∼\sim5 eVee across the full sub-keV range.

A custom offline energy reconstruction algorithm has been developed for the CONUS+ experiment. By processing the full digitized waveform offline, this approach offers several advantages over the online CAEN reconstruction. Since the full waveform is available, advanced noise reduction techniques such as optimal filtering [31] can be applied, which are not accessible in the online FPGA-based reconstruction. The non-linearity observed in the CAEN online reconstruction is suppressed in this approach, as the energy is extracted directly from the waveform without the constraints imposed by the fixed FPGA filter implementation. The custom reconstruction applied to the data demonstrates a comparable energy resolution to the online CAEN reconstruction, validating the approach. This algorithm will be used as the standard energy reconstruction for future CONUS+ analyses, providing a more accurate and stable energy scale at sub-keV energies, which is essential for high-precision CEν\nuNS measurements.

Neutron activation

The detector was irradiated with an 241AmBe{}^{241}\mathrm{AmBe} neutron source to enhance the 71Ge{}^{71}\mathrm{Ge} production rate within the crystal and improve the statistical precision of the low-energy calibration lines. The 241AmBe{}^{241}\mathrm{AmBe} source produces neutrons through a two-step nuclear reaction process involving alpha decay from 241Am{}^{241}\mathrm{Am} and subsequent interaction with beryllium nuclei:

241Am→237Np+α9Be+α→12C+n(4){}^{241}\mathrm{Am} \rightarrow{}^{237}\mathrm{Np} + \alpha\\ {}^{9}\mathrm{Be} + \alpha\rightarrow{}^{12}\mathrm{C} + n \tag*{(4)}

Additional reaction channels, such as 9Be(α,n)8Be+α{}^{9}\mathrm{Be}(\alpha,n){}^{8}\mathrm{Be}+\alpha, contribute at the few percent level. The source has an activity of 3.84 GBq corresponding to a neutron emission rate of approximately 10510^{5} n/s. It emits neutrons with an average energy of about 4.5 MeV and a maximum energy up to 11 MeV [32, 33, 34]. This energy range is particularly well suited for the activation of 71Ge{}^{71}\mathrm{Ge} via neutron capture on 70Ge{}^{70}\mathrm{Ge}, while avoiding the production of long-lived radioactive isotopes in the surrounding detector materials, in particular copper. The activation products from copper, 64Cu{}^{64}\mathrm{Cu} (half-life 12.7 h) and 66Cu{}^{66}\mathrm{Cu} (half-life 5.12 min [35]), decay rapidly and their contributions are negligible for measurements performed several days after irradiation [36]. Similarly, the production of tritium via neutron-induced spallation of germanium requires neutron energies above ∼10\sim10 MeV [37] and is therefore strongly suppressed for the 241AmBe{}^{241}\mathrm{AmBe} spectrum. Among all possible activation products, 71Ge{}^{71}\mathrm{Ge} is therefore the only isotope that produces a significant contribution.

To maximize the neutron flux at the detector, the source-to-detector distance was reduced to 20 cm, compared to ∼1\sim1 m in the KKL irradiation campaigns, with fewer shielding layers between the source and the detector. This configuration led to an estimated neutron flux of approximately 18 n s−1 cm−218\ \mathrm{n\ s^{-1}\ cm^{-2}} at the detector position. The detector was continuously irradiated for one week, resulting in a total neutron fluence of approximately 1.1×107 n cm−21.1 \times10^{7}\ \mathrm{n\ cm^{-2}}, well below the 109 n cm−210^{9}\ \mathrm{n\ cm^{-2}} threshold for neutron-induced damage in germanium [38, 39, 40]. After irradiation, the detector was monitored for 31 days with daily spectra collected to study the time evolution of the activation products. Based on the detector mass of 2.4 kg, the 70Ge{}^{70}\mathrm{Ge} natural abundance and a neutron capture cross-section of 3 barn [41] for the 70Ge(n,γ)71Ge{}^{70}\mathrm{Ge}(\mathrm{n},\gamma){}^{71}\mathrm{Ge} reaction, approximately 10810^{8} 71Ge{}^{71}\mathrm{Ge} atoms were produced during the irradiation. Taking into account the 71Ge{}^{71}\mathrm{Ge} half-life of (11.4645±0.0036)(11.4645 \pm0.0036) days [42], this yields an expected emission of ∼107\sim10^{7} K-shell, ∼106\sim10^{6} L-shell and ∼105\sim10^{5} M-shell X-rays over the 31 days post-irradiation measurement period, representing an enhancement of more than three orders of magnitude compared to the ∼5000\sim5000 K-shell counts previously observed in the KKL data. Prior to the irradiation, 40 days of background data were collected under identical conditions to enable a precise background subtraction.

Figure 6 shows the comparison of the low-energy spectra up to 35 keVee_{\mathrm{ee}} acquired before (blue) and after (red) irradiation. The first 2 days after irradiation were not included to remove the short half-life products from copper. After irradiation, the 71Ge{}^{71}\mathrm{Ge} K-shell line at 10.367 keVee_{\mathrm{ee}} and L-shell line at 1.298 keVee_{\mathrm{ee}} become prominent, with count rates enhanced by more than three orders of magnitude with respect to the pre-irradiation background. The inset shows the region below 500 eVee_{\mathrm{ee}}, where a clear excess above the background is visible after irradiation, consistent with the expected M-shell X-ray emission from 71Ge{}^{71}\mathrm{Ge} at 158.1 eVee_{\mathrm{ee}}. The analysis of the M-shell line is presented in detail in § 5.

Figure 6. Comparison of the low-energy spectra acquired during 40 days before irradiation (blue) and 31 days after irradiation (red) with the 241AmBe{}^{241}\mathrm{AmBe} neutron source. The 71Ge{}^{71}\mathrm{Ge} K-shell and L-shell X-ray lines become prominent after irradiation. The inset shows the sub-500 eVee_{\mathrm{ee}} region, where an excess consistent with the 71Ge{}^{71}\mathrm{Ge} M-shell line at 158.1 eVee_{\mathrm{ee}} is visible after irradiation.

The K- and L-shell lines are used to establish the low-energy calibration of the C8 detector after irradiation. The spectra shown in Figure 7 are obtained after subtraction of the background before irradiation, isolating the contribution from the 71Ge{}^{71}\mathrm{Ge} decay. The L-shell region (left panel) is fitted with two Gaussian functions with a high energy tail to account for the L1_1 and L2_2 sub-shell capture lines, with fitted centroids of (1298.5±1.3)(1298.5 \pm1.3) eVee_{\mathrm{ee}} and (1148.2±6.1)(1148.2 \pm6.1) eVee_{\mathrm{ee}} [25, 43], respectively, in good agreement with the reference values of 1298 eVee_{\mathrm{ee}} and 1143.2 eVee_{\mathrm{ee}}. The asymmetric high-energy tail in both lines is attributed to pile-up events. The K-shell (right panel) peak at 10367 eVee_{\mathrm{ee}} is fitted with a Crystal Ball function, which accounts for the asymmetric low-energy tail arising from incomplete charge collection in the detector bulk and surface regions. As a cross-check, the fits are repeated without the background subtraction. In this case, a step-like background component is included for the K-shell region, while an exponential background is assumed for the L-shell. The resulting fit parameters are consistent within 1%. The total number of counts in the K-shell peak is of the order of 10710^{7}, consistent with the expectations from the activation calculation described above. The calibration achieved after irradiation yields an energy scale uncertainty of ±1.3\pm1.3 eVee_{\mathrm{ee}} at 150 eVee_{\mathrm{ee}}, a significant improvement with respect to the pre-irradiation value of more than ±5\pm5 eVee_{\mathrm{ee}}. For each line, the reconstructed peak position, energy resolution (FWHM), total count rate, and the ratio of the K-shell to L-shell count rates are summarized in Table 1, together with the comparison to the literature values [27].

Fits to the $^{71}\mathrm{Ge}$ L-shell and K-shell X-ray lines

Figure 7. Fits to the 71Ge^{71}\mathrm{Ge} L-shell (left) and K-shell (right) X-ray lines after subtracting the background before irradiation. The L-shell data (red) are fitted with two Gaussian functions (black) accounting for the L1_1 (1298.5 eVee_{\mathrm{ee}}) and L2_2 (1148.2 eVee_{\mathrm{ee}}) sub-shell capture lines. The K-shell data (red) are fitted with a Crystal Ball function (black), which accounts for the asymmetric low-energy tail arising from incomplete charge collection.

K-shellL1_1-shellL2_2-shellM-shell
Position [eVee]_{\mathrm{ee}}]This work10368.3±1.310368.3\pm1.31298.5±1.31298.5\pm1.31148.2±6.11148.2\pm6.1158.7±1.4158.7\pm1.4
Literature10367.1±0.510367.1\pm0.51297.7±1.11297.7\pm1.11143.21143.2158.1±0.5158.1\pm0.5
FWHM [eVee]_{\mathrm{ee}}]This work170.6±1.3170.6\pm1.376.4±1.376.4\pm1.373.8±7.473.8\pm7.458.2±3.458.2\pm3.4
Ratio to K-shellThis work110.118±0.0010.118\pm0.0010.0014±0.00030.0014\pm0.00030.0214±0.00040.0214\pm0.0004
Literature110.119±0.0030.119\pm0.0030.0010.0010.02050.0205
Half-life [days]This work11.46±0.0211.46\pm0.0211.49±0.1611.49\pm0.1611.42±0.2211.42\pm0.22
Literature11.4645±0.003611.4645\pm0.0036

Table 1.

To confirm that the observed lines originate from 71Ge{}^{71}\mathrm{Ge} decay, the time evolution of the K- and L-shell peak count rates was studied using daily spectra collected over the 31-days after the irradiation period. Both decay curves are well described by a single exponential function, yielding half-life values of (11.46±0.02)(11.46 \pm0.02) days for the K-shell and (11.49±0.16)(11.49 \pm0.16) days for the L-shell. These values are in excellent agreement with the most precise recent measurement of the 71Ge{}^{71}\mathrm{Ge} half-life, T1/2=(11.4645±0.0036)T_{1/2} = (11.4645 \pm0.0036) days [42], and with the other recent measurements [44, 45]. The precise knowledge of the 71Ge{}^{71}\mathrm{Ge} half-life is of broader interest beyond detector calibration, as it plays a direct role in the interpretation of the so-called gallium anomaly discussed in the context of the SAGE, GALLEX/GNO and BEST [46, 47, 48] experiments, which was proposed to be partially explained by an underestimation of the 71Ge{}^{71}\mathrm{Ge} half-life [49]. The recent precise measurements exclude this hypothesis. Our independent measurement, obtained as a byproduct of the calibration campaign, is fully consistent with these results and provides an additional cross-check of the 71Ge{}^{71}\mathrm{Ge} half-life in a completely different experimental context. The large statistics accumulated after irradiation also enable, for the first time in CONUS+, the study of the 71Ge{}^{71}\mathrm{Ge} M-shell X-ray line at ∼158.1\sim158.1 eVee_{\mathrm{ee}}, which was not accessible in previous campaigns due to the insufficient activation rate. The analysis of this line is presented in the following section.

M@汥瑀瑯步渠--shell line observation

The 71Ge{}^{71}\mathrm{Ge} M-shell X-ray line at ∼158.1\sim158.1 eVee_{\mathrm{ee}} lies in the most challenging region of the detector response, close to the energy threshold just above the region dominated by electronic noise. Its observation requires both a sufficient activation rate, as achieved by the 241AmBe{}^{241}\mathrm{AmBe} irradiation campaign described in § 4, and a precise energy reconstruction.

The M-shell peak is extracted from the post-irradiation spectrum without subtraction of the background to avoid effects related to the noise peak stability. The resulting spectrum in the [0,400][0, 400] eVee_{\mathrm{ee}} range is shown in the left panel of Figure 8. The spectrum is dominated at low energies by the noise peak, which is pre-fitted independently (fnoise(E)f_{\mathrm{noise}}(E)). The remaining spectrum is described by a fit model consisting of the trigger efficiency εtrig\varepsilon_{\mathrm{trig}} multiplied by the sum of an exponential background and a Gaussian peak for the M-shell line:

Fit of the $^{71}$Ge M-shell X-ray line and time evolution of the daily M-shell yield

Figure 8. Left: fit of the 71^{71}Ge M-shell X-ray line in the post-irradiation spectrum. The data (red) are described by a model consisting of a noise pre-fit (green), an exponential background weighted by the trigger efficiency (blue dashed), and the convolution of the M-shell signal with the trigger efficiency and electronic smearing (black), yielding μM=(158.7±1.4) eVee\mu_{\mathrm{M}}=(158.7\pm1.4)\ \mathrm{eV}_{\mathrm{ee}} and FWHMM=(58.2±3.4) eVee\mathrm{FWHM}_{\mathrm{M}}=(58.2\pm3.4)\ \mathrm{eV}_{\mathrm{ee}}. Right: time evolution of the daily M-shell yield NMN_{\mathrm{M}} over the 31-day post-irradiation period (blue points), fitted with a single exponential decay function (red), yielding a half-life of (11.42±0.22)(11.42\pm0.22) days, consistent with the known 71^{71}Ge half-life of (11.4645±0.0036)(11.4645\pm0.0036) days [42].

f(E)=fnoise(E)+εtrig(E)⋅ea−bE+A[εtrig⋅G(μM,σFano)]∗G(0,σnoise),(5)f(E) = f_{\mathrm{noise}}(E) + \varepsilon_{\mathrm{trig}}(E) \cdot e^{a-bE} + A [\varepsilon_{\mathrm{trig}} \cdot G(\mu_M,\sigma_{\mathrm{Fano}})] \ast G(0,\sigma_{\mathrm{noise}}), \tag*{(5)}

where aa and bb are the exponential background parameters, μM\mu_M is the M-shell peak position, AA the amplitude, σnoise\sigma_{\mathrm{noise}} the contribution of the electronic noise to the M-shell width and σFano\sigma_{\mathrm{Fano}} the contribution from the Fano factor. The trigger efficiency εtrig\varepsilon_{\mathrm{trig}} acts on the true energy E′E' before the electronic smearing, as there is no correlation between the DAQ trigger filter and the trapezoidal energy reconstruction. Since σFano≪σnoise\sigma_{\mathrm{Fano}} \ll\sigma_{\mathrm{noise}}, the peak shape is nearly symmetric. The total energy resolution (FWHMM_M) is obtained by combining the two contributions in quadrature. The fit yields a peak position of μM=(158.7±1.4)\mu_M = (158.7 \pm1.4) eVee_{\mathrm{ee}} and an energy resolution of FWHMM=(58.2±3.4)\mathrm{FWHM}_M = (58.2 \pm3.4) eVee_{\mathrm{ee}}, in good agreement with the literature value of (158.1±0.5)(158.1 \pm0.5) eVee_{\mathrm{ee}} [27]. The stability of the fit and the uniqueness of the minimum were verified through a 2D Δχ2\Delta\chi^2 profile scan in the (μM,FWHMM)(\mu_M, \mathrm{FWHM}_M) plane, confirming well-defined confidence contours. The small correlation between the peak position and width is accounted for in the propagation of the energy calibration systematic into the resolution uncertainty.

To confirm the 71Ge^{71}\mathrm{Ge} origin of the observed M-shell peak, the time evolution of the M-shell yield was studied using daily spectra after irradiation, as shown in the right panel of Figure 8. The decay curve is well described by a single exponential function, yielding a half-life of (11.42±0.22)(11.42 \pm0.22) days, in good agreement with the known 71Ge^{71}\mathrm{Ge} half-life of (11.4645±0.0036)(11.4645 \pm0.0036) days [42] and consistent with the values obtained for the K- and L-shell lines, confirming that the observed peak originates from 71Ge^{71}\mathrm{Ge} decay.

The reconstructed peak positions and energy resolutions for the K-, L- and M-shell lines are summarized in Table 1, together with the comparison to the literature values [27]. The measured positions are in good agreement for all three shells, validating the energy calibration, trigger efficiency and the non-linearity correction across the full sub-keV range from 158.1 eVee_{\mathrm{ee}} to 10.367 keVee_{\mathrm{ee}}. The K/L and M/L ratios of the count rates are also consistent with the expected branching ratios [27], indicating a proper modeling of the trigger efficiency. As an additional cross-check, the fit was repeated leaving the trigger efficiency parameters free, yielding values consistent with the independently measured ones, further validating the event reconstruction at sub-keV energies.

The energy resolution of (58.2±3.4)(58.2 \pm3.4) eVee_{\mathrm{ee}} (FWHM) measured at the M-shell line at (158.7±1.4)(158.7 \pm1.4) eVee_{\mathrm{ee}} provides a direct experimental determination of the detector resolution at sub-keV energies. The energy resolution of a Ge detector is dominated by two main contributions, the electronic noise and the Fano fluctuations in the charge creation process. Other sources like microphonic noise are found to be negligible. Effects such as incomplete charge collection can be accounted for with a quadratic term at high energies [50, 51]. The total resolution is described by:

σ2(E)=σnoise2+F⋅ϵ⋅E+β2⋅E2,(6)\sigma^2(E) = \sigma_{\mathrm{noise}}^2 + F \cdot\epsilon\cdot E + \beta^2 \cdot E^2, \tag*{(6)}

where FF is the dimensionless Fano factor, ϵ=2.96\epsilon= 2.96 eV is the mean energy required to create an electron-hole pair in germanium [50], and β\beta is the dimensionless ballistic deficit parameter accounting for incomplete charge collection at higher energies [51]. In the absence of ballistic deficit (β=0\beta= 0), Eq. (6) reduces to the standard Fano model. The electronic noise contribution σnoise\sigma_{\mathrm{noise}} is fixed from the pulser measurement ((56±2)(56 \pm2) eVee_{\mathrm{ee}} FWHM). The total energy resolution as a function of energy is shown in Figure 9. Fitting the standard Fano model (β=0\beta= 0, blue solid line) to the M-, L- and K-shell data yields F=0.151±0.002F = 0.151 \pm0.002. Including the quadratic term (black dash-dotted line) significantly improves the Table 1: Summary of the fit results for the 71^{71}Ge K-, L- and M-shell X-ray lines, compared to literature values [25, 27, 42]. The reconstructed peak position, energy resolution (FWHM), and intensity ratios with respect to the K-shell are reported.

Energy resolution as a function of energy

Figure 9. Energy resolution (σ\sigma) as a function of energy measured at the 71Ge{}^{71}\mathrm{Ge} M-, L- and K-shell X-ray lines at 158.7, 1298.5 and 10368.3 eVee\mathrm{eV}_{\mathrm{ee}}, respectively. The solid blue line shows a fit of the standard Fano model (F=0.151±0.002F = 0.151 \pm0.002), and the black dash-dotted line includes the quadratic term for incomplete charge collection (F=0.123±0.011F = 0.123 \pm0.011). In both cases σnoise\sigma_{\mathrm{noise}} is fixed by the pulser measurement (FWHM (56±2) eVee(56 \pm2)\,\mathrm{eV}_{\mathrm{ee}}). The Fano factor including the quadratic term is in better agreement with the values in literature [52, 53, 54, 55, 56, 57].

agreement with the data, giving F=0.123±0.011F=0.123\pm0.011 and a ballistic deficit parameter of β=(2.9±0.5)×10−3\beta=(2.9\pm0.5)\times10^{-3}.

The Fano factor from the quadratic fit is in good agreement with values in the range 0.080.08–0.130.13 typically reported in the literature for HPGe detectors [52, 53, 54, 55, 56, 57]. It is also in good agreement with the value used in [23]. The value is significantly smaller than F=0.39F=0.39 reported by SuperCDMS [51], which may suggest a temperature dependence of the Fano factor, given that the present detector was operated at 73 K while SuperCDMS operates at ∼50\sim50 mK. The ballistic deficit term is similar in both detectors.

Conclusions

In this work, we have presented a dedicated neutron activation campaign aimed at improving the sub-keV energy calibration of the CONUS+ experiment and demonstrated the detector response at energies close to the detection threshold. The campaign was performed at the Low Level Laboratory of MPIK using the C8 detector, one of the new 2.4 kg HPGe detectors produced for the CONUS+ Phase-2, and a 241^{241}AmBe neutron source placed at 20 cm from the detector. The resulting neutron fluence of ∼1.1×107 n cm−2\sim1.1\times10^{7}\ \mathrm{n\,cm}^{-2} produced approximately 10810^{8} 71^{71}Ge atoms within the crystal, enhancing the K- and L-shell X-ray line rates by more than three orders of magnitude compared to the previous irradiation at KKL. The use of neutron-induced 71Ge{}^{71}\mathrm{Ge} activation provides several key advantages for low-energy calibration. Its relatively short half-life of about 11 days allows the detector to return to nominal background conditions within a few months, enabling repeated calibrations between physics runs. Additionally, these events originate mostly from bulk energy depositions, closely matching the topology of CEνNS\mathrm{CE}\nu\mathrm{NS} interactions and therefore providing an optimal proxy for signal-like events.

The enhanced 71Ge{}^{71}\mathrm{Ge} statistics allowed a significant improvement of the energy scale calibration. The energy scale uncertainty at 150 eVee\mathrm{eV}_{\mathrm{ee}}, the reference threshold for the CEνNS\mathrm{CE}\nu\mathrm{NS} analysis, was reduced from more than ±5 eVee\pm5\,\mathrm{eV}_{\mathrm{ee}} before irradiation to ±1.3 eVee\pm1.3\,\mathrm{eV}_{\mathrm{ee}} after irradiation. Propagating this improvement into the CONUS+ signal prediction, the contribution of the energy scale uncertainty to the combined signal prediction uncertainty is reduced from 14% to below 4%, making it comparable to other uncertainties. This demonstrates that a single activation campaign of one week is sufficient to reach the calibration precision required for the upcoming CONUS+ phases.

For the first time, the 71Ge{}^{71}\mathrm{Ge} M-shell X-ray line at ∼158 eVee\sim158\,\mathrm{eV}_{\mathrm{ee}} has been clearly resolved in a CONUS+ detector. The measured peak position of (158.7±1.4) eVee(158.7 \pm1.4)\,\mathrm{eV}_{\mathrm{ee}} is in good agreement with the literature value, and the energy resolution (FWHM) of (58.2±3.4) eVee(58.2 \pm3.4)\,\mathrm{eV}_{\mathrm{ee}} is consistent with the expected detector performance at this energy. The 71Ge{}^{71}\mathrm{Ge} origin of the line was confirmed by the time evolution of the M-shell yield, which follows an exponential decay with a half-life of (11.42±0.22)(11.42 \pm0.22) days, in good agreement with the known value of (11.4645±0.0036)(11.4645 \pm0.0036) days [42]. The M-shell signal is observed with a statistical significance greater than 10σ10\sigma. Beyond the validation of the energy scale, the clear observation of the M-shell line demonstrates the detector performance in the immediate vicinity of the CEνNS\mathrm{CE}\nu\mathrm{NS} region of interest, providing experimental access to the energy resolution, the trigger efficiency turn-on, and the capability to discriminate physical low-energy depositions from noise and background events at threshold.

The energy resolution measured at the M-, L- and K-shell energies is well described by the resolution model of Eq. (6). A Fano factor of F=0.123±0.011F = 0.123 \pm0.011 is estimated consistent with values reported in the literature [52, 53, 54, 55, 56, 57]. The non-linearity of the CAEN online energy reconstruction was characterized via dedicated pulser scans, revealing a maximum deviation of −12 eVee-12\,\mathrm{eV}_{\mathrm{ee}} at ∼160 eVee\sim160\,\mathrm{eV}_{\mathrm{ee}}. A spline interpolation correction was applied to all reconstructed energies prior to the spectral analysis, validated by the agreement of the measured M-shell peak position with the literature value.

The results presented in this work establish the full experimental and analysis foundation for a future high-statistics activation campaign at the KKL reactor site. Applying the same irradiation procedure with the 252Cf{}^{252}\mathrm{Cf} source available at KKL will allow the CONUS+ detectors to achieve an energy scale uncertainty below 2 eVee\mathrm{eV}_{\mathrm{ee}}, reducing the signal prediction uncertainty contribution to below 4% for all detectors simultaneously. Combined with the validated trigger efficiency and the demonstrated M-shell response, this will significantly strengthen the CONUS+ energy calibration and open the path to precision measurements of the CEνNS\mathrm{CE}\nu\mathrm{NS} cross-section in the fully coherent regime. In addition, the high-statistics sample of 71Ge{}^{71}\mathrm{Ge} decays, dominated by bulk events, provides an excellent dataset to further develop pulse shape discrimination techniques, enabling improved separation of surface and bulk events in future high-precision CEνNS\mathrm{CE}\nu\mathrm{NS} measurements.

Acknowledgements

We thank Mirion Technologies (Canberra) in Lingolsheim for their highly professional support.

References

  1. [1]M. Lindner, T. Rink and M. Sen. Light vector bosons and the weak mixing angle in the light of future germanium-based reactor CEνNS experiments. JHEP, 08:171, 2024. doi:10.1007/JHEP08(2024)171. arXiv:2401.13025.DOI
  2. [2]S. Centelles Chuliá, M. Lindner and T. Rink. Testing lepton non-unitarity with the next generation of (Germanium-based) CEνNS reactor experiments. 2025. arXiv:2512.09027.DOI
  3. [3]A. A. Aguilar-Arevalo et al. Searches for CEνNS and Physics beyond the Standard Model using Skipper-CCDs at CONNIE. 2024. arXiv:2312.05976.
  4. [4]D. Mondal et al. CEνNS Search with Cryogenic Sapphire Detectors at MINER: Results from the TRIGA reactor data and Future Sensitivity at HFIR. 2025. arXiv:2510.15999.arxiv.org/abs/2510.15999
  5. [5]J. Colaresi, J. I. Collar, T. W. Hossbach, C. M. Lewis and K. M. Yocum. Measurement of Coherent Elastic Neutrino-Nucleus Scattering from Reactor Antineutrinos. Phys. Rev. Lett., 129(21):21802, 2022. doi:10.1103/PhysRevLett.129.211802. arXiv:2202.09672.DOI
  6. [6]J. J. Choi et al. Exploring coherent elastic neutrino-nucleus scattering using reactor electron antineutrinos in the NEON experiment. Eur. Phys. J. C, 83(3):226, 2023. doi:10.1140/epjc/s10052-023-11352-x. arXiv:2204.06318.DOI
  7. [7]H. Abel et al. Prospect of the NUCLEUS Experiment at Chooz for Coherent Elastic Neutrino-Nucleus Scattering and New Physics Searches. 2026. arXiv:2603.24450.
  8. [8]V. Belov et al. New constraints on coherent elastic neutrino-nucleus scattering by the νGeN experiment*. Chin. Phys. C, 49(5):053004, 2025. doi:10.1088/1674-1137/adb9c8. arXiv:2502.18502.DOI
  9. [9]N. Ackermann et al. Final CONUS Results on Coherent Elastic Neutrino-Nucleus Scattering at the Brokdorf Reactor. Phys. Rev. Lett., 133(25):251802, 2024. doi:10.1103/PhysRevLett.133.251802. arXiv:2401.07684.DOI
  10. [10]D. Y. Akimov et al. First constraints on the coherent elastic scattering of reactor antineutrinos off xenon nuclei. Phys. Rev. D, 111(7):072012, 2025. doi:10.1103/PhysRevD.111.072012. arXiv:2411.18641.
  11. [11]L. Bailly-Salins. Status of the RICOCHET experiment at the ILL. PoS, EPS-HEP2025:150, 2026. doi:10.22323/1.485.0150.
  12. [12]L. Yang, Y. Liang and Q. Yue. RECODE program for reactor neutrino CEνNS detection with PPC Germanium detector. PoS, TAUP2023:296, 2024. doi:10.22323/1.441.0296.
  13. [13]C. Cai et al. Reactor neutrino liquid xenon coherent elastic scattering experiment. Phys. Rev. D, 110(7):072011, 2024. doi:10.1103/PhysRevD.110.072011. arXiv:2405.05554.DOI
  14. [14]S. Karmakar et al. New Limits on Coherent Neutrino Nucleus Elastic Scattering Cross Section at the Kuo-Sheng Reactor Neutrino Laboratory. 2024. arXiv:2411.18812.DOI
  15. [15]D. Z. Freedman. Coherent Neutrino Nucleus Scattering as a Probe of the Weak Neutral Current. Phys. Rev. D, 9:1389–1392, 1974. doi:10.1103/PhysRevD.9.1389.
  16. [16]V. B. Kopeliovich and L. L. Frankfurt. Isotopic and chiral structure of neutral current. JETP Lett., 19:145–147, 1974.
  17. [17]D. Akimov et al. Observation of Coherent Elastic Neutrino-Nucleus Scattering. Science, 357(6356):1123–1126, 2017. doi:10.1126/science.aao0990. arXiv:1708.01294DOI
  18. [18]D. Akimov et al. First Measurement of Coherent Elastic Neutrino-Nucleus Scattering on Argon. Phys. Rev. Lett., 126(1):012002, 2021. doi:10.1103/PhysRevLett.126.012002. arXiv:2003.10630.DOI
  19. [19]M. Adhikari et al. Measurement of coherent elastic neutrino nucleus scattering on germanium by COHERENT. 2026. arXiv:2603.17951.arxiv.org/abs/2603.17951
  20. [20]E. Aprile et al. Probing the Solar 8 B Neutrino Fog with XENONnT. 2026. arXiv:2604.06002.DOI
  21. [21]Z. Bo et al. First Indication of Solar B8 Neutrinos through Coherent Elastic Neutrino-Nucleus Scattering in PandaX-4T. Phys. Rev. Lett., 133(19):191001, 2024. doi:10.1103/PhysRevLett.133.191001. arXiv:2407.10892.DOI
  22. [22]D. S. Akerib et al. Searches for Light Dark Matter and Evidence of Coherent Elastic Neutrino-Nucleus Scattering of Solar Neutrinos with the LUX-ZEPLIN (LZ) Experiment. 2025. arXiv:2512.08065.DOI
  23. [23]N. Ackermann et al. Direct observation of coherent elastic antineutrino–nucleus scattering. Nature, 643(8074):1229–1233, 2025. doi:10.1038/s41586-025-09322-2. arXiv:2501.05206.DOI
  24. [24]I. R. Barabanov et al. Cosmogenic activation of germanium and its reduction for low background experiments. Nucl. Instrum. Meth. A, 569(3):877–884, 2006. doi:10.1016/j.nima.2006.05.247.
  25. [25]R. Agnese et al. Production rate measurement of tritium and other cosmogenic isotopes in germanium with CDMSlite. Astropart. Phys., 104:1–12, 2019. doi:10.1016/j.astropartphys.2018.08.006. arXiv:1806.07043.DOI
  26. [26]H. Bonet et al. Full background decomposition of the CONUS experiment. Eur. Phys. J. C, 83(3):195, 2023. doi:10.1140/epjc/s10052-023-11240-4. arXiv:2112.09585.DOI
  27. [27]Laboratoire National Henri Becquerel. Recommended Data: Table of Radionuclides. http://www.lnhb.fr/nuclear-data/. Accessed 2025.
  28. [28]H. Bonet et al. Pulse shape discrimination for the CONUS experiment in the keV and sub-keV regime. Eur. Phys. J. C, 84(2):139, 2024. doi:10.1140/epjc/s10052-024-12470-w. arXiv:2308.12105.DOI
  29. [29]G. Heusser. Low-radioactivity background techniques. Ann. Rev. Nucl. Part. Sci., 45:543–590, 1995. doi:10.1146/annurev.ns.45.120195.002551.DOI
  30. [30]N. Ackermann et al. CONUS+ Experiment. Eur. Phys. J. C, 84(12):1265, 2024. doi:10.1140/epjc/s10052-024-13551-6. arXiv:2407.11912.DOI
  31. [31]M. R. Anderson et al. Performance of a convolutional autoencoder designed to remove electronic noise from p-type point contact germanium detector signals. Eur. Phys. J. C, 82(12):1084, 2022. doi:10.1140/epjc/s10052-022-11000-w. arXiv:2204.06655.DOI
  32. [32]E. A. Lorch. Neutron energy spectra from Am-Be and Pu-Be sources. International Journal of Applied Radiation and Isotopes, 24(10):585–589, 1973. doi:10.1016/0020-708X(73)90016-3.
  33. [33]D. H. Vijaya and A. Kumar. The neutron spectrum of Am-Be neutron sources. Nucl. Instrum. Methods, 111:435–440, 1973.
  34. [34]H. Kluge and K. Weise. The neutron energy spectrum of a 241Am-Be(α,n) source and resulting mean fluence to dose equivalent conversion factors. Radiat. Prot. Dosim., 2(2):85–93, 1982.DOI
  35. [35]National Nuclear Data Center (NNDC). Evaluated Nuclear Structure Data File (ENSDF), 2022. doi:10.18139/ nndc.ensdf/1845010. Data extracted from the NNDC On-Line Data Service.
  36. [36]National Nuclear Data Center. NuDat 3.0 Nuclear Structure and Decay Data. https://www.nndc.bnl.gov/nudat3/. Accessed 2025.
  37. [37]R. Agnese et al. Production Rate Measurement of Tritium and Other Cosmogenic Isotopes in Germanium with CDMSlite. Astropart. Phys., 104:1–12, 2019. doi: 10.1016/j.astropartphys.2018.08.006. arXiv:1806.07043.DOI
  38. [38]M. Descoivich et al. Effects of neutron damage on the performance of large volume segmented germanium detectors. Nuclear Instruments and Methods in Physics Research Section A, 545(1):199–209, 2005. doi:10.1016/j.nima.2005.01.308.
  39. [39]T. W. Raudorf and R. H. Pehl. Effect of charge carrier trapping on germanium coaxial detector line shapes. Nuclear Instruments and Methods in Physics Research Section A, 255(3):538–551, 1987. doi:10.1016/0168-9002(87)91225-3.
  40. [40]R. H. Pehl et al. Radiation damage resistance of reverse electrode Ge coaxial detectors. IEEE Transactions on Nuclear Science, 26(1):321–323, 1979. doi: 10.1109/TNS.1979.4329652.DOI
  41. [41]S. F. Mughabghab. Atlas of Neutron Resonances: Resonance Parameters and Thermal Cross Sections. Elsevier, 5 edition, 2006.
  42. [42]A. V. Derbin et al. New Precision Measurements of 71Ge. Phys. At. Nucl., 88(1), 2025. doi: 10.1134/S1063778825700152.
  43. [43]A. C. Thompson et al. X-Ray Data Booklet, Section 1, Table 1-1: Characteristic X-Ray Energies. Lawrence Berkeley National Laboratory, https://xdb.lbl.gov/Section1/Table_1-1.pdf, 2009. Accessed: April 2026.
  44. [44]J. J. Ikal and S. G. Yoon. Measurement of the 71Ge half-life. Phys. Rev. C, 108:L021602, 2023. doi:10.1103/PhysRevC.108.L021602.
  45. [45]E. B. Norman et al. Precision measurement of the 71Ge half-life. Phys. Rev. C, 109:055501, 2024. doi:10.1103/PhysRevC.109.055501.
  46. [46]J. N. Abdurashitov et al. Measurement of the solar neutrino capture rate with gallium metal. III. Results for the 2002–2007 data-taking period. Phys. Rev. C, 80:015807, 2009. doi:10.1103/PhysRevC.80.015807.
  47. [47]F. Kaether, W. Hampel, G. Heusser, J. Kiko and T. Kirsten. Reanalysis of the GALLEX solar neutrino flux and source experiments. Phys. Lett. B, 685:47, 2010. doi:10.1016/j.physletb.2010.01.030.DOI
  48. [48]V. V. Barinov et al. Results from the Baksan Experiment on Sterile Transitions (BEST). Phys. Rev. Lett., 128:232501, 2022. doi:10.1103/PhysRevLett.128.232501.DOI
  49. [49]C. Giunti, Y. F. Li, C. A. Ternes and Z. Xin. Constraining light sterile neutrinos with the gallium anomaly. Phys. Lett. B, 842:137983, 2023. doi:10.1016/j.physletb.2023.137983.
  50. [50]M. C. Lépy et al. Experimental study of the response of germanium detectors to low-energy photons. Nucl. Instrum. Meth. A, 439(2):239–246, 2000. doi:10.1016/S0168-9002(00)00210-8.DOI
  51. [51]R. Agnese et al. Search for Low-Mass Dark Matter with CDMSlite Using a Profile Likelihood Fit. Phys. Rev. D, 99(6):062001, 2019. doi:10.1103/PhysRevD.99.062001. arXiv:1809.09098.
  52. [52]S. Croft and D. S. Bond. A determination of the Fano factor for germanium at 77.4 K from measurements of the energy resolution of a 113 cm3 HPGe gamma-ray spectrometer taken over the energy range from 14 to 6129 keV. Appl. Radiat. Isot., 1996.
  53. [53]B. G. Lowe. Measurements of Fano factors in silicon and germanium in the low-energy X-ray region. Nucl. Instrum. Meth. A, 1997. doi:10.1016/S0168-9002(97)00965-0.DOI
  54. [54]T. Papp et al. A new approach to the determination of the Fano factor for semiconductor detectors. X-Ray Spectrometry, 34:106–111, 2005.DOI
  55. [55]A. Wiens et al. The 71Ge decay and its use for the calibration of germanium detectors. European Physical Journal A, 49:1–10, 2013.
  56. [56]M. Sánchez. Precision Measurement of the Fano Factor in Germanium. Thesis, Universität Heidelberg, 2015.
  57. [57]M. Agostini et al. Characterization of GERDA Phase II detectors. European Physical Journal C, 79(11):978, 2019. doi:10.1140/epjc/s10052-019-7353-8.

Paper details

vlnc:W_gwdwdkrg
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents