Prospect of the NUCLEUS Experiment at Chooz for Coherent Elastic Neutrino-Nucleus Scattering and New Physics Searches

arXiv:2603.24450 · hep-ex, hep-ph · Submitted 2026-03-25 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: High Energy Physics: Experiment. The day in high-energy experiment: detectors, searches and limits.

Kai: We're looking at the paper: "Prospect of the NUCLEUS Experiment at Chooz for Coherent Elastic Neutrino-Nucleus Scattering and New Physics Searches".

Mira: Here's the gist of the paper. The NUCLEUS experiment aims to measure coherent elastic neutrino–nucleus scattering (CEνNS) at unprecedentedly low nuclear recoil energies using gram-scale cryogenic calorimeters operated at the Chooz nuclear power plant in France.

Kai: The NUCLEUS experiment projects a 4.7 σ observation of CEνNS with a statistical precision of about 20% in 1 year for the Physics Run, enabling a determination of the weak mixing angle at the lowest momentum transfer probed to date with CEνNS and leading CEνNS-based constraints on the neutrino charge radius and new mediator models.

Mira: Experimental Setup and Background Mitigation The experiment deploys gram-scale cryogenic calorimeters in close proximity to the Chooz nuclear power plant in France.

Kai: The NUCLEUS will employ CaWO4 target crystals operated at 10 mK and instrumented with tungsten transitionedge sensors (W-TES), targeting nuclear-recoil energy thresholds down to O(10 eV) thanks to the excellent intrinsic energy resolution of cryogenic calorimetry.

Title and authors: Kai: So we’re talking about the paper "Prospect of the NUCLEUS Experiment at Chooz for Coherent Elastic Neutrino-Nucleus Scattering and New Physics Searches." This paper sets the stage by describing exactly what this experiment is designed to do.

Mira: Right, it outlines the core concept: CEnuNS, which is a Standard Model process involving coherent scattering of a neutrino off an entire nucleus via zero exchange. It's predicted by Freedman back in one thousand nine hundred seventy-four one <ref:2603.24450#pg2>.

Lev: The paper points out that COHERENT provided the first experimental observation of this using a cesium iodine detector and neutrinos from the spallation neutron source. They've also done measurements on argon nuclei and more recently on germanium five six seven eight <ref:2603.24450#pg1>.

Kai: But the real push here is looking beyond just Standard Model physics. It sets up a search for effects that go beyond the SM, like nonstandard neutrino interactions or the neutrino magnetic moment.

Mira: The paper highlights how this process is sensitive to things like light vector mediator scenarios because of the enhancement at low momentum transfer when you have a new mediator four <ref:2603.24450#pg1>.

The paper's summary: Kai: This section summarizes the main goals of the NUCLEUS experiment. It boils down to using ultra-low nuclear recoil energies to probe physics that higher energy experiments can't reach.

Mira: It specifically mentions that during the technical run, sensitivity is limited by a low-energy excess or LEE that dominates the recoil energy spectrum at sub-kilo electron volts.

Lev: They hypothesize that a significant fraction of this LEE might be stress-induced, coming from relaxation processes in the detector crystal or surrounding mechanical structures

thirty-one–thirty-three: <ref:2603.24450#pg3,from relaxation processes in the detector crystal or surrounding mechanical structures>.

Kai: The paper says that their double transition edge sensor technology has shown it can discriminate a distinct LEE component that couples more to the sensor than to the absorber phonon system.

Mira: For the physics run, they assume complete suppression of this LEE, making the particle-induced background from simulations the main thing they need to worry about

thirty-one–thirty-three: <ref:2603.24450#pg3>.

The paper's improvements: Kai: The authors suggest several key improvements for the experiment. They focus on how the physics run will handle those low-energy challenges we just talked about.

Mira: They discuss using reactor power variation to disentangle signal and background in a regime where the signal-to-background ratio is quite low two <ref:2603.24450#pg1>.

Lev: The paper mentions that during the physics run, they will integrate the detectors into an instrumented holder, which is a key upgrade with respect to the technical run because it's expected to provide additional event rejection capabilities against LEE from holder-related issues

thirty-six–thirty-seven: <ref:2603.24450#pg3,a key upgrade with respect to the technical run>.

Kai: They also project how this setup will allow them to achieve a median discovery significance of four point seven sigma for the Standard Model CEnuNS signal during the physics run, using only seven grams of calcium tungstate.

Conclusion: Mira: So, to wrap up this paper "Prospect of the NUCLEUS Experiment at Chooz for Coherent Elastic Neutrino-Nucleus Scattering and New Physics Searches," it establishes that NUCLEUS is uniquely positioned by accessing the lowest nuclear recoil energies achieved to date.

Kai: It shows that even with a small target mass, NUCLEUS can be competitive for both Standard Model and new-physics scenarios. The results highlight the strong complementarity between these ultra-low threshold cryogenic detectors and larger-mass experiments operating at higher recoil energy thresholds.

Lev: From an engineering standpoint, the paper shows that improving statistical precision by a factor of two for the Standard Model CEnuNS measurement requires increasing the target mass by about a factor six, which is around forty grams of calcium tungstate Appendix B.

Mira: And they project that this experiment will constrain the diagonal electronic neutrino charge radius within negative eight point three to five point six times ten to negative thirty two centimeters squared at ninety percent confidence level in the physics run.

Kai: The paper also gives a projected sensitivity for the effective electron flavor neutrino magnetic moment, putting it below two point zero times ten to negative ten mu-B at ninety percent confidence level, which is better than some current limits.

Lev: And they flag that the energy scale uncertainty introduces an asymmetric bias of plus zero point five to minus four point zero percent on the reconstructed CEnuNS signal, but this is still much smaller than the expected twenty percent statistical uncertainty in the physics run Appendix B.

Atominstitut · Max-Planck-Institut für Physik der Osterreichischen Akademie der Wissenschaften · Istituto Nazionale di Fisica Nucleare – Seziona di Roma “Tor Vergata” · IRFU, CEA, Université Paris-Saclay · Physik-Department, TUM School of Natural Sciences · Marietta-Blau-Institut für Teilchenphysik der Osterreichischen Akademie der Wissenschaften · Istituto Nazionale di Fisica Nucleare – Seziona di Roma · Dipartimento di Fisica, Sapienza Université di Roma “Tor Vergata” · Istituto Nazionale di Fisica Nucleare – Seziona di Ferrara · Dipartimento di Fisica, Université di Ferrara

hep-ex, hep-ph

Submitted: 2026-03-25

Updated: 2026-08-13

Journal ref: Phys. Rev. D 114, 012016 (2026)

DOI: 10.1103/dnyj-xzr1

License: http://creativecommons.org/licenses/by/4.0/

Importance score: 81/100

The gist: The NUCLEUS experiment aims to measure coherent elastic neutrino–nucleus scattering (CEνNS) at unprecedentedly low nuclear recoil energies using gram-scale cryogenic calorimeters operated at the…

Key concepts

Coherent Elastic Neutrino–Nucleus Scattering (CEνNS)
This is a rare interaction where a neutrino scatters off an entire nucleus coherently, rather than just one nucleon. Measuring this process allows physicists to probe fundamental properties of neutrinos and the weak interaction at very low momentum transfers.
Low-Energy Excess (LEE)
This is a spectral feature dominating the recoil energy spectrum at sub-keV energies, which limits sensitivity in the Technical Run. The experiment plans to model this excess using a double-exponential function and aims to suppress it completely during the Physics Run to observe the true CEνNS signal.
Light Vector Mediator Scenarios
These are new physics models involving a light vector boson (like a Z prime) that can interact with neutrinos. NUCLEUS is particularly sensitive here because the interaction strength is enhanced at the ultra-low recoil energies it probes, allowing for stringent constraints on these hypothetical mediators.

Terminology

Summary

The NUCLEUS experiment aims to measure coherent elastic neutrino–nucleus scattering (CEνNS) at unprecedentedly low nuclear recoil energies using gram-scale cryogenic calorimeters operated at the Chooz nuclear power plant in France. p. 0

Where the paper says it

Wex , 8 and L. Wienke 8 (NUCLEUS Collaboration) 1Atominstitut, Technische Universit¨at Wien, Stadionallee 2, Wien, A-1020, Austria 2Max-Planck-Institut f¨ur Physik, Boltzmannstraße 8, Garching, D-85748, Germany 3Marietta-Blau-Institut f¨ur Teilchenphysik der Osterreichischen Akademie ¨ der Wissenschaften, Dominikanerbastei 16, Wien, A-1010, Austria 4 Istituto Nazionale di Fisica Nucleare – Sezione di Roma “Tor Vergata”, Via della Ricerca Scientifica 1, Roma, I-00133, Italy 5 IRFU, CEA, Universit´e Paris-Saclay, Bˆatiment 141, Gif-sur-Yvette, F-91191, France 6 Istituto Nazionale di Fisica Nucleare – Sezione di Roma, Piazzale Aldo Moro 2, Roma, I-00185, Italy 7Dipartimento di Fisica, Sapienza Universit`a di Roma, Piazzale Aldo Moro 5, Roma, I-00185, Italy 8Physik-Department, TUM School of Natural Sciences, Technische Universit¨at M¨unchen, James-Franck-Straße 1, Garching, D-85748, Germany 9Dipartimento di Fisica, Universit`a di Roma “Tor Vergata”, Via della Ricerca Scientifica 1, Roma, I-00133, Italy 10Istituto Nazionale di Fisica Nucleare – Sezione di Ferrara, Via Giuseppe Saragat 1c, Ferrara, I-44122, Italy 11Dipartimento di Fisica, Universit`a di Ferrara, Via Giuseppe Saragat 1, Ferrara, I-44122, Italy The NUCLEUS experiment aims to measure coherent elastic neutrino–nucleus scattering (CEνNS) at unprecedentedly low nuclear recoil energies using gram-scale cryogenic calorimeters operated at the Chooz nuclear power plant in France.

The NUCLEUS experiment projects a 4. p. 0

Where the paper says it

Prospect of the NUCLEUS Experiment at Chooz for Coherent Elastic Neutrino-Nucleus Scattering and New Physics Searches H. Abele , 1 G.

7 σ observation of CEνNS with a statistical precision of about 20% in 1 year for the Physics Run, enabling a determination of the weak mixing angle at the lowest momentum transfer probed to date with CEνNS and leading CEνNS-based constraints on the neutrino charge radius and new mediator models. p. 0
Where the paper says it

For the Physics Run, assuming complete suppression of the LEE, we project a 4. 7 σ observation of CEνNS with a statistical precision of about 20% in 1 year, enabling a determination of the weak mixing angle at the lowest momentum transfer probed to date with CEνNS and leading CEνNS-based constraints on the neutrino charge radius and new mediator models.

Experimental Setup and Background Mitigation

The experiment deploys gram-scale cryogenic calorimeters in close proximity to the Chooz nuclear power plant in France. The NUCLEUS will employ CaWO4 target crystals operated at 10 mK and instrumented with tungsten transitionedge sensors (W-TES), targeting nuclear-recoil energy thresholds down to O(10 eV) thanks to the excellent intrinsic energy resolution of cryogenic calorimetry. p. 1

Where the paper says it

The NUCLEUS Collaboration aims to measure the CEνNS cross section in the fully coherent regime with high precision by deploying gram-scale cryogenic calorimeters in close proximity to the Chooz nuclear power plant in France [19]. NUCLEUS will employ CaWO4 target crystals operated at 10 mK and instrumented with tungsten transitionedge sensors (W-TES), targeting nuclear-recoil energy thresholds down to O(10 eV) thanks to the excellent intrinsic energy resolution of cryogenic calorimetry [20].

The experiment will be hosted in Chooz at the so-called very near site (VNS), a ∼ 24 m2 experimental room located between the two reactor cores at distances of 72 m and 102 m, respectively. p. 1
Where the paper says it

A precise measurement of reactor CEνNS with NUCLEUS will also provide important input for future dark-matter searches with similar detectors, as coherent neutrino scattering will ultimately constitute an irreducible background once sensitivities reach the neutrino fog [23]. The experiment will be hosted in Chooz at the so-called very near site (VNS), a ∼ 24 m2 experimental room located between the two reactor cores at distances of 72 m and 102 m, respectively.

Background mitigation involves passive shielding with multiple active veto systems, including a cryogenic muon veto and a cryogenic outer veto (COV) based on high-purity Ge detectors. p. 1
Where the paper says it

The background mitigation concept combines passive shielding with multiple active veto systems, whose performance and interplay have been extensively studied using detailed Monte Carlo simulations [25]. In particular, muon-induced backgrounds are suppressed by a muon-veto system [26] that includes a cryogenic muon veto [27], while external γ rays and secondary particles are further mitigated by a cryogenic outer veto (COV) based on high-purity Ge detectors providing close to 4π coverage of the cryogenic target [28, 29].

Signal Modeling and LEE Handling

The expected CEνNS recoil spectrum is obtained by convolving the differential cross section with the reactor antineutrino flux, integrating over all possible neutrino energies, and folding in the detector energy response. During the Technical Run, the sensitivity is limited by the low-energy excess (LEE), which dominates the recoil-energy spectrum at sub-keV energies. p. 1

Where the paper says it

The primary goals of this phase are to validate detector performance under VNS conditions and to characterize the particle background in situ. Based on the commissioning results, the sensitivity during the Technical Run is expected to remain limited by the low-energy excess (LEE), an excess 2

The LEE spectral shape is parameterized by a double-exponential function. p. 6
Where the paper says it

The analysis range extends up to 500 eV. Within this range, the LEE spectral shape is parameterized by a double-exponential function, SLEE(Tnr) = B × h f e−Tnr/ε1 + (1 − f) e −Tnr/ε2 i , (12) where SLEE(Tnr) describes the normalized energy distribution of LEE events in the analysis window.

For the Physics Run, an optimistic scenario assumes complete suppression of the LEE, making the dominant background the particle-induced component estimated from simulations. p. 7
Where the paper says it

investigation. For the sensitivity projections presented in this work, we consider an optimistic scenario in which the LEE is suppressed to a negligible level, such that the dominant background during the Physics Run is given by the particle-induced component estimated from simulations [25].

Sensitivity to New Physics Scenarios

The ultra-low nuclear recoil energy threshold provides enhanced sensitivity to new-physics scenarios that modify the CEνNS spectrum at low energies. The study considers nonstandard neutrino interactions (NSI) and the neutrino magnetic moment as well. p. 3

Where the paper says it

Beyond the Standard Model Scenarios The ultra-low nuclear recoil energy threshold targeted by NUCLEUS provides enhanced sensitivity to new-physics scenarios that modify the CEνNS spectrum at low energies. In this work, we consider two well-motivated classes of BSM scenarios: nonstandard neutrino interactions (NSI) and the neutrino magnetic moment.

Light vector mediator scenarios are particularly sensitive to the ultra-low recoil-energy regime probed by NUCLEUS due to the 1/(⃗q2 + m2Z′) enhancement of the mediator propagator at low momentum transfer. p. 4
Where the paper says it

Other gauge structures, such as B − L or B − 3Le [59, 61, 62], or scenarios with a scalar mediator [63], can be straightforwardly accommodated once experimental data become available. Owing to the 1/(|⃗q| 2 + m2 Z′ ) enhancement of the mediator propagator at low momentum transfer, light vector mediator scenarios are particularly sensitive to the ultra-low recoil-energy regime probed by NUCLEUS.

Statistical Framework and Results

The analysis framework utilizes a likelihood framework that incorporates reactor-power variation to disentangle signal and background in a low signal-to-background regime. The combined time-energy likelihood yields the largest median test statistic, reflecting the complementarity of spectral and reactor-power information. p. 9

Where the paper says it

For comparison, a likelihood analysis based solely on spectral-shape information yields a substantially weaker sensitivity, at the level of about 200 times the SM rate, due to the strong degeneracy between the CEνNS spectrum and the LEE shape. An analysis based only on the time variation of the total event rate performs better, reaching a sensitivity of roughly 50 times the SM expectation at the 90% CL.

In the Physics Run, a median discovery significance of 4. p. 9
Where the paper says it

In contrast, during the Physics Run – where the LEE contribution is assumed to be suppressed to negligible levels – a clear observation of the SM CEνNS signal becomes feasible. For the reference configuration considered here, a median discovery significance of 4.

7 σ is expected for SM CEνNS with a statistical uncertainty of about 20% using only 7 g of CaWO4. p. 9
Where the paper says it

For the reference configuration considered here, a median discovery significance of 4. 7 σ is expected, corresponding to approximately 58 expected CEνNS events and a statistical uncertainty of around 20% on the measured cross section.

The experiment is expected to constrain the diagonal electronic neutrino charge radius within (−8. p. 3
Where the paper says it

In fact, radiative corrections induce a flavor-dependent contribution to the neutrinoproton coupling through the neutrino charge radius ⟨r 2 νe ⟩ (NCR), which represents the only nonvanishing electromagnetic property of neutrinos within the SM. For the electronic flavor, this is equal to ⟨r 2 νe ⟩ ≃ −0.

3 < ⟨r2νe⟩ < 5.6) × 10−32 cm2 at 90% CL in the Physics Run. p. 3
Where the paper says it

For the electronic flavor, this is equal to ⟨r 2 νe ⟩ ≃ −0. 83 × 10−32 cm2 [49].

p. 3
Where the paper says it

For the electronic flavor, this is equal to ⟨r 2 νe ⟩ ≃ −0. 83 × 10−32 cm2 [49].

Conclusion and Future Outlook

The NUCLEUS experiment occupies a unique position in the landscape of CEνNS measurements by accessing the lowest nuclear recoil energies achieved to date. The study demonstrates that NUCLEUS can compensate for its small target mass and provide competitive sensitivity to both SM and new-physics scenarios. p. 5

Where the paper says it

Owing to its characteristic 1/Tnr enhancement, this contribution is particularly relevant at very low recoil energies, making NUCLEUS sensitive to competitive constraints despite its small target mass. An example of the corresponding theoretical prediction, assuming a neutrino magnetic moment of µνe = 10−9 µB, is shown in Figure 1.

p. 5
Where the paper says it

Owing to its characteristic 1/Tnr enhancement, this contribution is particularly relevant at very low recoil energies, making NUCLEUS sensitive to competitive constraints despite its small target mass. An example of the corresponding theoretical prediction, assuming a neutrino magnetic moment of µνe = 10−9 µB, is shown in Figure 1.

The experiment is expected to achieve a clear observation of SM CEνNS with a statistical uncertainty of about 20% using only 7 g of CaWO4 in the Physics Run. p. 9
Where the paper says it

Because the sensitivity of the Technical Run is limited by the dominant LEE background, it improves only slowly with exposure, and extending the data-taking period beyond one year would not lead to a substantial gain in sensitivity within realistic timescales. In contrast, during the Physics Run – where the LEE contribution is assumed to be suppressed to negligible levels – a clear observation of the SM CEνNS signal becomes feasible.

The results highlight the strong complementarity between ultra-lowthreshold cryogenic detectors and larger-mass experiments operating at higher recoil-energy thresholds. p. 6
Where the paper says it

75 g of each Technical Run detector, we assume that the LEE rate does not scale with detector mass. This assumption is supported by observations from other cryogenic experiments operating at comparable energy thresholds [32, 35, 76].

p. 6
Where the paper says it

75 g of each Technical Run detector, we assume that the LEE rate does not scale with detector mass. This assumption is supported by observations from other cryogenic experiments operating at comparable energy thresholds [32, 35, 76].

p. 6
Where the paper says it

75 g of each Technical Run detector, we assume that the LEE rate does not scale with detector mass. This assumption is supported by observations from other cryogenic experiments operating at comparable energy thresholds [32, 35, 76].

p. 6
Where the paper says it

75 g of each Technical Run detector, we assume that the LEE rate does not scale with detector mass. This assumption is supported by observations from other cryogenic experiments operating at comparable energy thresholds [32, 35, 76].

p. 6
Where the paper says it

75 g of each Technical Run detector, we assume that the LEE rate does not scale with detector mass. This assumption is supported by observations from other cryogenic experiments operating at comparable energy thresholds [32, 35, 76].

The NUCLEUS experiment can compensate for its small target mass and provide competitive sensitivity to both SM and new-physics scenarios. p. 5
Where the paper says it

Owing to its characteristic 1/Tnr enhancement, this contribution is particularly relevant at very low recoil energies, making NUCLEUS sensitive to competitive constraints despite its small target mass. An example of the corresponding theoretical prediction, assuming a neutrino magnetic moment of µνe = 10−9 µB, is shown in Figure 1.

p. 5
Where the paper says it

Owing to its characteristic 1/Tnr enhancement, this contribution is particularly relevant at very low recoil energies, making NUCLEUS sensitive to competitive constraints despite its small target mass. An example of the corresponding theoretical prediction, assuming a neutrino magnetic moment of µνe = 10−9 µB, is shown in Figure 1.

p. 5
Where the paper says it

Owing to its characteristic 1/Tnr enhancement, this contribution is particularly relevant at very low recoil energies, making NUCLEUS sensitive to competitive constraints despite its small target mass. An example of the corresponding theoretical prediction, assuming a neutrino magnetic moment of µνe = 10−9 µB, is shown in Figure 1.

p. 5
Where the paper says it

Owing to its characteristic 1/Tnr enhancement, this contribution is particularly relevant at very low recoil energies, making NUCLEUS sensitive to competitive constraints despite its small target mass. An example of the corresponding theoretical prediction, assuming a neutrino magnetic moment of µνe = 10−9 µB, is shown in Figure 1.

p. 5
Where the paper says it

Owing to its characteristic 1/Tnr enhancement, this contribution is particularly relevant at very low recoil energies, making NUCLEUS sensitive to competitive constraints despite its small target mass. An example of the corresponding theoretical prediction, assuming a neutrino magnetic moment of µνe = 10−9 µB, is shown in Figure 1.

The results highlight the strong complementarity between ultra-lowthreshold cryogenic detectors and larger-mass experiments operating at higher recoil-energy thresholds. The NUCLEUS experiment can compensate for its small target mass and provide competitive sensitivity to both SM and new-physics scenarios. The results highlight the strong complementarity between ultra-lowthreshold cryogenic detectors and larger-mass experiments operating at higher recoil-energy thresholds. The NUCLEUS experiment can compensate for its small target mass and provide competitive sensitivity to both SM and new-physics scenarios.

Impact of Target Mass Scaling

Improving the CaWO4 target mass offers substantial margin to improve the precision of the CEνNS cross-section measurement. Increasing the CaWO4 target mass offers substantial margin to improve the precision of the CEνNS cross-section measurement. p. 1

Where the paper says it

The NUCLEUS Collaboration aims to measure the CEνNS cross section in the fully coherent regime with high precision by deploying gram-scale cryogenic calorimeters in close proximity to the Chooz nuclear power plant in France [19]. NUCLEUS will employ CaWO4 target crystals operated at 10 mK and instrumented with tungsten transitionedge sensors (W-TES), targeting nuclear-recoil energy thresholds down to O(10 eV) thanks to the excellent intrinsic energy resolution of cryogenic calorimetry [20].

Pushing the precision further will eventually bring the experiment into a regime where systematic uncertainties dominate over the statistical ones. p. 2
Where the paper says it

The appendices provide details on the validation of the statistical framework (App. A) and on the treatment of systematic uncertainties (App.

A mass up to several orders of magnitude smaller than that employed by other CEνNS experiments operating at higher recoil-energy thresholds. p. 7
Where the paper says it

The unbinned log-likelihood function (L) used in this work is defined as − log L = " α Z T2 T1 P(t) dt + Z T2 T1 β(t) dt# − X i log [α P(ti) fs(Ei) + β(ti) fb(Ei)] , (15) where each term has a well-defined physical interpretation: • α is the rate of the CEνNS signal, expressed in counts per day per MWth; • P(t) is the time-dependent reactor power, proportional to the instantaneous antineutrino flux at the detector, and ti denotes the timestamp of the i-th event; • fs(E) is the CEνNS recoil-energy probability density function (PDF) within the analysis window, including detector response effects; 2 The much lower event rate in NUCLEUS is primarily a consequence of the gram-scale target mass, in contrast to the ∼ 8. 3 t liquid scintillator target mass of each Double Chooz detector [79], even though the CEνNS cross section at reactor energies is orders of magnitude larger than the one of the inverse beta decay.

The results are shown in Fig. p. 9
Where the paper says it

An analysis based only on the time variation of the total event rate performs better, reaching a sensitivity of roughly 50 times the SM expectation at the 90% CL. These results highlight the substantial gain achieved by combining temporal and spectral information in the likelihood, as illustrated in Fig.

7, which reports the 1 σ median statistical precision obtained through the sensitivity procedure described in Sec. p. 9
Where the paper says it

The distributions of the test statistic obtained from ensembles of pseudo-experiments are used to determine median expected discovery significances for the SM CEνNS signal or, in the absence of a signal, to derive projected upper limits on the parameters of interest at the desired confidence level. Additional details on the statistical procedure and its validation are provided in App.

III B.

Systematic Uncertainty Evaluation

The impact of the energy-scale uncertainty is modeled by introducing a linear relation between the observed and true recoil energies Tnr,obs = Tnr,true(1 + δ). The resulting systematic bias is an asymmetric bias of +0. p. 5

Where the paper says it

CEνNS Signal Prediction The expected CEνNS recoil spectrum is obtained by convolving the differential cross section with the reactor antineutrino flux, integrating over all possible neutrino energies, and folding in the detector energy response. The resulting differential event rate is given by dR dTnr =NT X N Z T ′max nr 0 dT′ nr R(Tnr, T′ nr) Z E max ν Emin ν (T′ nr) dEν× dΦ dEν (Eν) dσν¯e−N dT′ nr (Eν, T′ nr) , (10) where the sum runs over the nuclear species N in CaWO4, and NT denotes the number of target nuclei in the detector.

p. 5
Where the paper says it

CEνNS Signal Prediction The expected CEνNS recoil spectrum is obtained by convolving the differential cross section with the reactor antineutrino flux, integrating over all possible neutrino energies, and folding in the detector energy response. The resulting differential event rate is given by dR dTnr =NT X N Z T ′max nr 0 dT′ nr R(Tnr, T′ nr) Z E max ν Emin ν (T′ nr) dEν× dΦ dEν (Eν) dσν¯e−N dT′ nr (Eν, T′ nr) , (10) where the sum runs over the nuclear species N in CaWO4, and NT denotes the number of target nuclei in the detector.

5 −4.0% on the reconstructed CEνNS signal, with a similar variation observed in the median expected discovery significance. p. 9
Where the paper says it

For the reference configuration considered here, a median discovery significance of 4. 7 σ is expected, corresponding to approximately 58 expected CEνNS events and a statistical uncertainty of around 20% on the measured cross section.

p. 9
Where the paper says it

For the reference configuration considered here, a median discovery significance of 4. 7 σ is expected, corresponding to approximately 58 expected CEνNS events and a statistical uncertainty of around 20% on the measured cross section.

Both systematic effects are therefore subdominant compared to the expected ∼ 20% statistical uncertainty of the CEνNS measurement in the Physics Run. p. 2
Where the paper says it

For both runs, a total measurement period of one year is assumed. If the LEE cannot be fully suppressed, but is nevertheless reduced compared to the currently observed level, an intermediate physics reach between the Technical Run and Physics Run scenarios can be expected.

p. 2
Where the paper says it

For both runs, a total measurement period of one year is assumed. If the LEE cannot be fully suppressed, but is nevertheless reduced compared to the currently observed level, an intermediate physics reach between the Technical Run and Physics Run scenarios can be expected.

p. 2
Where the paper says it

For both runs, a total measurement period of one year is assumed. If the LEE cannot be fully suppressed, but is nevertheless reduced compared to the currently observed level, an intermediate physics reach between the Technical Run and Physics Run scenarios can be expected.

The uncertainty on the reactor thermal power determination is approximately 5% when using publicly available reactor power information. p. 8
Where the paper says it

This results in an overall effective reactor duty cycle of approximately 85%. The assumed schedule is representative of typical Chooz operating conditions and is consistent with historical reactor operation patterns inferred from publicly available electricity production records averaged over the past decade [80].

p. 8
Where the paper says it

This results in an overall effective reactor duty cycle of approximately 85%. The assumed schedule is representative of typical Chooz operating conditions and is consistent with historical reactor operation patterns inferred from publicly available electricity production records averaged over the past decade [80].

p. 8
Where the paper says it

This results in an overall effective reactor duty cycle of approximately 85%. The assumed schedule is representative of typical Chooz operating conditions and is consistent with historical reactor operation patterns inferred from publicly available electricity production records averaged over the past decade [80].

Sensitivity to BSM Scenarios

The projected sensitivity to a universal light vector mediator model exhibits a characteristic degeneracy band corresponding to regions where destructive interference occurs between the SM nuclear weak charge and the contribution from the new vector mediator. The right panel of Figure 5 shows the projected NUCLEUS sensitivity in the (ϵuVee, ϵdVee) plane achievable during the Physics Run, compared with 90% CL constraints from a selection of existing CEνNS experiments. p. 9

Where the paper says it

The green point indicate the expected precision for the Physics Run, assuming complete suppression of the LEE. This is compared to the precision from atomic parity violation (APV) on cesium [48] and existing measurements from CEνNS experiments, specifically: CONUS+ [82], COHERENT CsI [83], Ar [84], Ge [85], a combined analysis of XENONnT [10] and PandaX [9] data [86] and the determination from LUX-ZEPLIN [11].

NUCLEUS is expected to place a constraint on the effective electron-flavor neutrino magnetic moment that is more stringent than the current limit obtained from the combined COHERENT CsI+Ar dataset. p. 4
Where the paper says it

The magnetic-moment contribution does not interfere with the SM CEνNS amplitude and adds incoherently to Eq. (1), namely dσMM νℓ-N dTnr = πα2 EM m2 e 1 Tnr − 1 Eν Z 2F 2 Z(|⃗q| 2 ) µνℓ µB 2 , (9) where µνℓ is the effective magnetic moment of the neutrino flavor νℓ and µB is the Bohr magneton [49].

The Physics Run yields a projected sensitivity of µνe < 2. p. 9
Where the paper says it

23873(5) COH CsI COH Ar COH Ge PandaX -4T + XENONnT LUX -ZEPLIN CONUS + NUCLEUS Sensitivity Physics Run (LEE-free) FIG. Projected sensitivity of the NUCLEUS experiment to the weak mixing angle sin2 θW (Q)MS as a function of the momentum transfer |⃗q| .

0 × 10−10µB at 90% CL.

SM CEνNS Sensitivity

A detection of the SM CEνNS signal is not expected during the upcoming Technical Run because the LEE exceeds the expected CEνNS signal by several orders of magnitude. Using reactor-power variation analysis, a median 90% CL sensitivity corresponding to a signal normalization approximately 35 times larger than the SM prediction is obtained. p. 9

Where the paper says it

III A 2, we obtain a median 90% confidence level (CL) sensitivity corresponding to a signal normalization approximately 35 times larger than the SM prediction. To put this number into context, we can compare it with results obtained from other reactor CEνNS experiments.

During the Physics Run – where the LEE contribution is assumed to be suppressed to negligible levels – a clear observation of the SM CEνNS signal becomes feasible with a median discovery significance of 4. p. 9
Where the paper says it

In contrast, during the Physics Run – where the LEE contribution is assumed to be suppressed to negligible levels – a clear observation of the SM CEνNS signal becomes feasible. For the reference configuration considered here, a median discovery significance of 4.

7 σ. This precision enables a competitive determination of the weak mixing angle at low momentum transfer probed to date with CEνNS and leading CEνNS-based constraints on the neutrino charge radius. p. 0
Where the paper says it

For the Physics Run, assuming complete suppression of the LEE, we project a 4. 7 σ observation of CEνNS with a statistical precision of about 20% in 1 year, enabling a determination of the weak mixing angle at the lowest momentum transfer probed to date with CEνNS and leading CEνNS-based constraints on the neutrino charge radius and new mediator models.

The expected 1σ (68. p. 1
Where the paper says it

In particular, muon-induced backgrounds are suppressed by a muon-veto system [26] that includes a cryogenic muon veto [27], while external γ rays and secondary particles are further mitigated by a cryogenic outer veto (COV) based on high-purity Ge detectors providing close to 4π coverage of the cryogenic target [28, 29]. Together, these systems are designed to compensate for the limited overburden at the VNS and to suppress particle-induced backgrounds in the CEνNS region of interest to a level that preserves sensitivity to the expected signal.

p. 1
Where the paper says it

In particular, muon-induced backgrounds are suppressed by a muon-veto system [26] that includes a cryogenic muon veto [27], while external γ rays and secondary particles are further mitigated by a cryogenic outer veto (COV) based on high-purity Ge detectors providing close to 4π coverage of the cryogenic target [28, 29]. Together, these systems are designed to compensate for the limited overburden at the VNS and to suppress particle-induced backgrounds in the CEνNS region of interest to a level that preserves sensitivity to the expected signal.

p. 1
Where the paper says it

In particular, muon-induced backgrounds are suppressed by a muon-veto system [26] that includes a cryogenic muon veto [27], while external γ rays and secondary particles are further mitigated by a cryogenic outer veto (COV) based on high-purity Ge detectors providing close to 4π coverage of the cryogenic target [28, 29]. Together, these systems are designed to compensate for the limited overburden at the VNS and to suppress particle-induced backgrounds in the CEνNS region of interest to a level that preserves sensitivity to the expected signal.

p. 1
Where the paper says it

In particular, muon-induced backgrounds are suppressed by a muon-veto system [26] that includes a cryogenic muon veto [27], while external γ rays and secondary particles are further mitigated by a cryogenic outer veto (COV) based on high-purity Ge detectors providing close to 4π coverage of the cryogenic target [28, 29]. Together, these systems are designed to compensate for the limited overburden at the VNS and to suppress particle-induced backgrounds in the CEνNS region of interest to a level that preserves sensitivity to the expected signal.

27% CL) interval for sin2 θW (q ∼ 5 MeV) is 0. p. 9
Where the paper says it

23873(5) COH CsI COH Ar COH Ge PandaX -4T + XENONnT LUX -ZEPLIN CONUS + NUCLEUS Sensitivity Physics Run (LEE-free) FIG. Projected sensitivity of the NUCLEUS experiment to the weak mixing angle sin2 θW (Q)MS as a function of the momentum transfer |⃗q| .

p. 9
Where the paper says it

23873(5) COH CsI COH Ar COH Ge PandaX -4T + XENONnT LUX -ZEPLIN CONUS + NUCLEUS Sensitivity Physics Run (LEE-free) FIG. Projected sensitivity of the NUCLEUS experiment to the weak mixing angle sin2 θW (Q)MS as a function of the momentum transfer |⃗q| .

p. 9
Where the paper says it

23873(5) COH CsI COH Ar COH Ge PandaX -4T + XENONnT LUX -ZEPLIN CONUS + NUCLEUS Sensitivity Physics Run (LEE-free) FIG. Projected sensitivity of the NUCLEUS experiment to the weak mixing angle sin2 θW (Q)MS as a function of the momentum transfer |⃗q| .

187 < sin2 θW (q ∼ 5 MeV) < 0. p. 9
Where the paper says it

23873(5) COH CsI COH Ar COH Ge PandaX -4T + XENONnT LUX -ZEPLIN CONUS + NUCLEUS Sensitivity Physics Run (LEE-free) FIG. Projected sensitivity of the NUCLEUS experiment to the weak mixing angle sin2 θW (Q)MS as a function of the momentum transfer |⃗q| .

p. 9
Where the paper says it

23873(5) COH CsI COH Ar COH Ge PandaX -4T + XENONnT LUX -ZEPLIN CONUS + NUCLEUS Sensitivity Physics Run (LEE-free) FIG. Projected sensitivity of the NUCLEUS experiment to the weak mixing angle sin2 θW (Q)MS as a function of the momentum transfer |⃗q| .

p. 9
Where the paper says it

23873(5) COH CsI COH Ar COH Ge PandaX -4T + XENONnT LUX -ZEPLIN CONUS + NUCLEUS Sensitivity Physics Run (LEE-free) FIG. Projected sensitivity of the NUCLEUS experiment to the weak mixing angle sin2 θW (Q)MS as a function of the momentum transfer |⃗q| .

286.

SM CEνNS Sensitivity to Charge Radius

NUCLEUS is expected to constrain the NCR within (−8.3 < ⟨r2νe⟩ < 5.6) × 10−32 cm2 at 90% CL in the Physics Run. This would represent one of the most stringent constraints to date, when compared to the existing CEνNS measurement shown in Figure 4. p. 5

Where the paper says it

Owing to its characteristic 1/Tnr enhancement, this contribution is particularly relevant at very low recoil energies, making NUCLEUS sensitive to competitive constraints despite its small target mass. An example of the corresponding theoretical prediction, assuming a neutrino magnetic moment of µνe = 10−9 µB, is shown in Figure 1.

The expected 1σ (68.27% CL) interval for sin2 θW (q ∼ 5 MeV) is 0.187 < sin2 θW (q ∼ 5 MeV) < 0.286. The expected precision for the Physics Run is compared to the precision from CONUS+ [68] and COHERENT CsI+Ar [45]. p. 9
Where the paper says it

The green point indicate the expected precision for the Physics Run, assuming complete suppression of the LEE. This is compared to the precision from atomic parity violation (APV) on cesium [48] and existing measurements from CEνNS experiments, specifically: CONUS+ [82], COHERENT CsI [83], Ar [84], Ge [85], a combined analysis of XENONnT [10] and PandaX [9] data [86] and the determination from LUX-ZEPLIN [11].

The expected 1σ (68.27% CL) interval for sin2 θW (q ∼ 5 MeV) is 0.187 < sin2 θW (q ∼ 5 MeV) < 0.286.

SM CEνNS Sensitivity to Magnetic Moment

The projected sensitivity to the effective electron-flavor neutrino magnetic moment µνe at the 90% CL is µνe < 22.3 × 10−10µB. For the Physics Run, an additional improvement of approximately one order of magnitude is expected, yielding a projected sensitivity of µνe < 2. p. 7

Where the paper says it

For the sensitivity projections presented in this work, we consider an optimistic scenario in which the LEE is suppressed to a negligible level, such that the dominant background during the Physics Run is given by the particle-induced component estimated from simulations [25]. Because the detectors are located at distances of order 70-100 m from the reactor cores, reactor-correlated particle backgrounds are expected to be negligible, and this component is therefore assumed to be constant in time.

0 × 10−10µB at 90% CL. This places NUCLEUS among the most sensitive CEνNS-based probes of the effective electron neutrino magnetic moment, even though CONUS+ constraints remain stronger when νES is included on top of the CEνNS signal. p. 4
Where the paper says it

This allows NUCLEUS to set competitive constraints on light mediator models even in the absence of a direct CEνNS observation. A fundamental BSM neutrino property is provided by the neutrino magnetic moment (µν), which is expected to arise because neutrinos are massive [49, 64, 65].

The expected 1σ (68.27% CL) interval for µνe at 90% CL is 0. p. 5
Where the paper says it

Owing to its characteristic 1/Tnr enhancement, this contribution is particularly relevant at very low recoil energies, making NUCLEUS sensitive to competitive constraints despite its small target mass. An example of the corresponding theoretical prediction, assuming a neutrino magnetic moment of µνe = 10−9 µB, is shown in Figure 1.

187 < µνe [10-10µB] < 2. p. 5
Where the paper says it

Owing to its characteristic 1/Tnr enhancement, this contribution is particularly relevant at very low recoil energies, making NUCLEUS sensitive to competitive constraints despite its small target mass. An example of the corresponding theoretical prediction, assuming a neutrino magnetic moment of µνe = 10−9 µB, is shown in Figure 1.

8 × 10−10µB.

SM CEνNS Sensitivity to Light Mediators

The sensitivity to a universal light vector mediator model exhibits a characteristic degeneracy band, corresponding to regions where destructive interference occurs between the SM nuclear weak charge and the contribution from the new vector mediator. The Physics Run significantly extends the reach of NUCLEUS, allowing previously unexplored regions of the parameter space to be probed, particularly for mediator masses between approximately 0. p. 4

Where the paper says it

1, where the red dot-dashed curve shows the theoretical prediction for a representative point in the (gZ′ , mZ′ ) parameter space, which features a strong enhancement at low recoil energies. This allows NUCLEUS to set competitive constraints on light mediator models even in the absence of a direct CEνNS observation.

1 and 10 MeV when compared to existing CEνNS experiments. p. 9
Where the paper says it

The green point indicate the expected precision for the Physics Run, assuming complete suppression of the LEE. This is compared to the precision from atomic parity violation (APV) on cesium [48] and existing measurements from CEνNS experiments, specifically: CONUS+ [82], COHERENT CsI [83], Ar [84], Ge [85], a combined analysis of XENONnT [10] and PandaX [9] data [86] and the determination from LUX-ZEPLIN [11].

Technical Run Performance

During the Technical Run at Chooz, the detector module will consist of four CaWO4 double-TES detectors, for a total target mass of 7 g. The analysis range extends up to 500 eV. p. 6

Where the paper says it

Since an energy threshold of 35 eV was achieved during commissioning, the same analysis threshold is adopted for the technical-run sensitivity projections. The analysis range extends up to 500 eV.

The median profile-likelihood ratio test statistic for the Technical Run configuration is obtained from ensembles of ∼ 500 pseudo-experiments fitted with the energy-only, time-only, and combined likelihoods. p. 8
Where the paper says it

Figure 2 shows an example pseudo-experiment for the Technical Run configuration, illustrating the resulting time and energy distributions of signal and background events. For each pseudo-experiment, parameter inference is performed using a profile likelihood ratio test statistic [81], q(θ) = −2 log L(θ, ˆνˆ) L( ˆθ, νˆ) , (16) where θ denotes the parameter of interest (e.

Validation of Statistical Framework

The empirical 90% quantile of the simulated distribution for the Technical Run configuration is found to be 2.67 ± 0.04, compared to the asymptotic expectation of 2. p. 8

Where the paper says it

The first term in square brackets in Eq. (15) represents the Poisson expectation value for the total number of events in the observation time window [T1, T2], given by the sum of the signal contribution, proportional to the time-integrated reactor power, and the time-integrated background rate.

p. 8
Where the paper says it

The first term in square brackets in Eq. (15) represents the Poisson expectation value for the total number of events in the observation time window [T1, T2], given by the sum of the signal contribution, proportional to the time-integrated reactor power, and the time-integrated background rate.

71 for a χ2 distribution with one degree of freedom. p. 8
Where the paper says it

Middle: Representative Monte Carlo pseudo-experiment for the Technical Run configuration, generated assuming a CEνNS signal normalized to 35 times the SM prediction for illustrative purposes. Time distribution of events over one year, showing the signal and background components separately, together with their expected PDFs.

p. 8
Where the paper says it

Middle: Representative Monte Carlo pseudo-experiment for the Technical Run configuration, generated assuming a CEνNS signal normalized to 35 times the SM prediction for illustrative purposes. Time distribution of events over one year, showing the signal and background components separately, together with their expected PDFs.

The empirical 68.27% (1σ) quantile of the simulated distribution for the Physics Run configuration is found to be 1. p. 8
Where the paper says it

Middle: Representative Monte Carlo pseudo-experiment for the Technical Run configuration, generated assuming a CEνNS signal normalized to 35 times the SM prediction for illustrative purposes. Time distribution of events over one year, showing the signal and background components separately, together with their expected PDFs.

p. 8
Where the paper says it

Middle: Representative Monte Carlo pseudo-experiment for the Technical Run configuration, generated assuming a CEνNS signal normalized to 35 times the SM prediction for illustrative purposes. Time distribution of events over one year, showing the signal and background components separately, together with their expected PDFs.

009 ± 0.019, in good agreement with the asymptotic expectation of unity. p. 5
Where the paper says it

This additional interaction channel is particularly relevant in scenarios predicting an enhanced low-energy signal, and has been successfully exploited in several experiments employing Ge [66–69], CsI [50, 69, 70] and Xe [71, 72] targets. In fact, the neutrino-electron scattering (νES) signal in the O(keV) recoil range and can be modeled with good theoretical control by scaling the free electron-neutrino cross section by the number of ionized electrons for a given energy deposit [73].

p. 5
Where the paper says it

This additional interaction channel is particularly relevant in scenarios predicting an enhanced low-energy signal, and has been successfully exploited in several experiments employing Ge [66–69], CsI [50, 69, 70] and Xe [71, 72] targets. In fact, the neutrino-electron scattering (νES) signal in the O(keV) recoil range and can be modeled with good theoretical control by scaling the free electron-neutrino cross section by the number of ionized electrons for a given energy deposit [73].

The median of this distribution defines the expected discovery significance, corresponding to the sensitivity achieved in 50% of repeated experiments under the SM hypothesis. p. 9
Where the paper says it

For the reference configuration considered here, a median discovery significance of 4. 7 σ is expected, corresponding to approximately 58 expected CEνNS events and a statistical uncertainty of around 20% on the measured cross section.

p. 9
Where the paper says it

For the reference configuration considered here, a median discovery significance of 4. 7 σ is expected, corresponding to approximately 58 expected CEνNS events and a statistical uncertainty of around 20% on the measured cross section.

The combined likelihood yields the largest median test statistic, reflecting the complementarity of spectral and reactor-power information. p. 9
Where the paper says it

signal [12]. For comparison, a likelihood analysis based solely on spectral-shape information yields a substantially weaker sensitivity, at the level of about 200 times the SM rate, due to the strong degeneracy between the CEνNS spectrum and the LEE shape.

Conclusion

The NUCLEUS experiment occupies a unique position in the landscape of CEνNS measurements by accessing the lowest nuclear recoil energies achieved to date. The study demonstrates that NUCLEUS can compensate for its small target mass and provide competitive sensitivity to both SM and new-physics scenarios. The results highlight the strong complementarity between ultra-lowthreshold cryogenic detectors and larger-mass experiments operating at higher recoil-energy thresholds. The NUCLEUS experiment can compensate for its small target mass and provide competitive sensitivity to both SM and new-physics scenarios. The results highlight the strong complementarity between ultra-lowthreshold cryogenic detectors and larger-mass experiments operating at higher recoil-energy thresholds. The NUCLEUS experiment can compensate for its small target mass and provide competitive sensitivity to both SM and new-physics scenarios.

Acknowledgments

This work has been financed by the CEA, the INFN, the OAW and partially supported by the TU Munich and the MPI f¨ur Physik. NUCLEUS members acknowledge additional funding by the DFG through the SFB1258 and Excellence Cluster ORIGINS, by the European Commission through the ERC-StG2018-804228 “NU-CLEUS”, by the P2IO LabEx (ANR-10-LABX-0038) in the framework “Investissements d’Avenir” (ANR-11IDEX-0003-01) managed by the Agence Nationale de la Recherche (ANR), France, by the Austrian Science Fund (FWF) through the “P 34778-N, ELOISE”, and by Max-Planck-Institut f¨ur Kernphysik (MPIK), Germany. p. 0

Where the paper says it

bossio@cea. fr b Also at: LIBPhys-UC, Departamento de Fisica, Universidade de Coimbra, Rua Larga 3004-516, Coimbra, P3004- 516, Portugal c Now at: Max-Planck-Institut f¨ur Kernphysik, Saupfercheckweg 1, Heidelberg, D-69117, Germany d Now at: Istituto Nazionale di Fisica Nucleare – Laboratori Nazionali del Gran Sasso, Via Giovanni Acitelli 22, Assergi (L’Aquila), I-67100, Italy e Now at: kiutra GmbH, Fl¨oßergasse 2, D-81369 Munich, Germany arXiv:2603.

Appendix A: Validation of Statistical Framework

To validate the fit procedure based on the likelihood defined in Eq. 15, we use ensembles of pseudo-experiments generated from the nominal signal and background model. p. 8

Where the paper says it

Possible interruptions requiring a warm-up and subsequent cool-down cycle could be incorporated into the phenomenological LEE model and included in the likelihood analysis if needed. Monte Carlo pseudo-experiments are generated by randomly sampling Poisson-distributed toy datasets from the signal and background PDFs defined above, 0 50 100 150 200 250 300 350 Time from measurement start (days) 0 2 4 P e ff ( G Wth ) B2 OFF B1 OFF B1 and B2 ON B1 and B2 OFF 0 50 100 150 200 250 300 350 Time from measurement start (days) 10 1 10 3 Events per day Background time PDF Background events Signal time PDF (35xSM) Signal events (35xSM) All events Technical Run: 7 g CaWO4 0.

The distribution of the test statistic q(αtrue), with αtrue = 35 × αSM, for the Technical Run configuration is shown in the left panel of Fig. p. 8
Where the paper says it

Figure 2 shows an example pseudo-experiment for the Technical Run configuration, illustrating the resulting time and energy distributions of signal and background events. For each pseudo-experiment, parameter inference is performed using a profile likelihood ratio test statistic [81], q(θ) = −2 log L(θ, ˆνˆ) L( ˆθ, νˆ) , (16) where θ denotes the parameter of interest (e.

9. The empirical 90% quantile of the simulated distribution is found to be 2. p. 6
Where the paper says it

Further details on the definition of t0 and the LEE time evolution are provided in Ref. When the time difference t−t0 is expressed in days, the exponent was found to be k ≃ 0.

67 ± 0.04, compared to the asymptotic expectation of 2.71 for a χ2 distribution with one degree of freedom. The empirical 68.27% (1σ) quantile of the simulated distribution for the Physics Run configuration is found to be 1.009 ± 0.019, in good agreement with the asymptotic expectation of unity. The median of this distribution defines the expected discovery significance, corresponding to the sensitivity achieved in 50% of repeated experiments under the SM hypothesis.

Appendix B: Systematic Uncertainties

The impact of the energy-scale uncertainty is modeled by introducing a linear relation between the observed and true recoil energies Tnr,obs = Tnr,true(1 + δ). The resulting systematic bias is an asymmetric bias of +0.5 −4.0% on the reconstructed CEνNS signal, with a similar variation observed in the median expected discovery significance. Both systematic effects are therefore subdominant compared to the expected ∼ 20% statistical uncertainty of the CEνNS measurement in the Physics Run. The uncertainty on the reactor thermal power determination is approximately 5% when using publicly available reactor power information. The impact of increased target mass is examined, showing that improving the statistical precision of the SM CEνNS measurement by a factor of two requires an increase in target mass by approximately a factor six (about 40 g of CaWO4). p. 5

Where the paper says it

No additional energy-dependent trigger efficiency is applied beyond this resolution folding; only a flat overall efficiency factor is included in the overall analysis efficiency quoted below (Sec. Technical Run During the Technical Run at Chooz, the detector module will consist of four CaWO4 double-TES detectors, for a total target mass of 7 g.

The uncertainty on the reactor thermal power determination is approximately 5% when using publicly available reactor power information. For future higher-exposure phases and mass scaling, the use of operator-provided reactor information and a detailed treatment of the fission inventory evolution will become necessary. The uncertainty on the SM value of the weak mixing angle induces only a subpercent variation in the normalization. p. 7

Where the paper says it

The time dependence of the effective reactor power induces a corresponding variation of the antineutrino flux, which can be exploited to identify signal and background in the time domain. This strategy was successfully employed by the Double Chooz experiment to obtain a background–modelindependent measurement of the mixing angle θ13 with a precision comparable to that achieved using full spectral information [78].

These effects are therefore negligible compared to the statistical and experimental systematic uncertainties considered above. p. 2
Where the paper says it

The appendices provide details on the validation of the statistical framework (App. A) and on the treatment of systematic uncertainties (App.

The uncertainty on the reactor antineutrino flux results in a relative uncertainty of ±2% on the extracted signal normalization. p. 9
Where the paper says it

In contrast, the CONUS+ experiment reported a 3. 7 σ observation of reactor CEνNS with a total uncertainty of about 27% on the extracted neutrino 1 5 10 50 100 0.

Both systematic effects are therefore subdominant compared to the expected ∼ 20% statistical uncertainty of the CEνNS measurement in the Physics Run. The uncertainty on the reactor thermal power determination is approximately 5% when using publicly available reactor power information.

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. Improved simulation of low-energy excess (LEE) backgrounds by incorporating data-driven treatment of LEE observed during commissioning runs, as mentioned in incorporating a data-driven treatment of the low-energy excess (LEE) observed during commissioning. This allows AI systems to better distinguish between stress-induced LEE and other backgrounds.

  2. Likelihood framework for signal and background disentanglement using reactor power variation, exploiting reactor-power variation to disentangle signal and background in a low signal-to-background regime. This enables AI models to assess the impact of dominant systematic uncertainties more accurately across different operational power states.

  3. Enhanced sensitivity projection for the Physics Run by assuming complete suppression of the LEE and using a likelihood framework that exploits both the time dependence induced by reactor-power variation and the recoil-energy information of individual events. This allows AI to project a clearer SM CEnuNS signal observation with a projected 4.7 sigma significance.

  4. Sensitivity analysis for new physics scenarios by exploring parameter space for nonstandard neutrino interactions (NSI) mediated by a light vector boson Z prime, utilizing the enhancement at low momentum transfer described in light vector mediator scenarios are particularly sensitive to the ultra-low recoil-energy regime probed by NUCLEUS. This allows AI to set competitive constraints on new mediator models.

  5. Constraint derivation for neutrino properties by projecting sensitivity to the neutrino charge radius and effective electron flavor magnetic moment, as demonstrated in Figure 4 and Figure 6. This enables AI systems to determine the lowest momentum-transfer determination of the weak mixing angle and place stringent limits on fundamental neutrino properties like muen e.

Abstract

The NUCLEUS experiment aims to measure coherent elastic neutrino-nucleus scattering (CE ν NS) at unprecedentedly low nuclear recoil energies using gram-scale cryogenic calorimeters operated at the Chooz nuclear power plant in France. Access to recoil energies at the O(10 eV) scale enables CE ν NS studies at extremely low momentum transfer and provides enhanced sensitivity to new physics. In this work, we present sensitivity projections for the upcoming NUCLEUS technical and physics runs, incorporating a data-driven treatment of the low-energy excess (LEE) observed during commissioning. We develop a likelihood framework that exploits reactor-power variation to disentangle signal and background in a low signal-to-background regime and to assess the impact of the dominant systematic uncertainties. For the Technical Run with a 7 g CaWO 4 target, we find competitive sensitivity to several scenarios beyond the Standard Model, which do not require a CE ν NS observation. For the Physics Run, assuming complete suppression of the LEE, we project a 4.7 σ observation of CE ν NS with a statistical precision of about 20 % in 1 year, enabling a determination of the weak mixing angle at the lowest momentum transfer probed to date with CE ν NS and leading CE ν NS-based constraints on the neutrino charge radius and new mediator models.

Sources

Related papers