Chasing the Light Sterile Neutrino Status of the STEREO Experiment

Chasing the light sterile neutrino Status of the STEREO experiment

Alessandro Minotti (IRFU - CEA Saclay) on behalf of the STEREO collaboration

16/03/2017 Outlook

• Neutrino physics and oscillation

• Reactor neutrinos and the Reactor Antineutrino Anomaly

• The STEREO experiment: principle, configuration, and timeline of the installation and commissioning phase

• The STEREO experiment: status of the analysis of first collected data

Alessandro Minotti (CEA - IRFU) Outlook

• Neutrino physics and oscillation

• Reactor neutrinos and the Reactor Antineutrino Anomaly

• The STEREO experiment: principle, configuration, and timeline of the installation and commissioning phase

• The STEREO experiment: status of the analysis of first collected data

Neutrino Physics and Oscillation Alessandro Minotti (CEA - IRFU) The Neutrino

• Neutrinos (ν) = neutral leptons

- A wide range of sources and energies

- Standard Model: 3 massless and only LH ν (RH ν̄)

νe νμ ντ W+ e+ W+ μ+ W+ τ+

• Neutrinos oscillate (change flavour) propagating α+ β+ - Energy-dependent deficit in solar ν flux να Flavour changing νβ W+ - Distance-dependent deficit for atmospheric ν’s

KOSMISK Electron-neutrinosSUN STRÅLNING are produced in the ATMOSFÄR Sun center.

SUPER- KAMIOKANDE 2015 SNO

Neutrino Physics and Oscillation Alessandro Minotti (CEA - IRFU) 4 Oscillation Formalism

• Neutrinos oscillate (change flavour) propagating α+ β+ να Flavour changing νβ Very small but non-zero different massesAs you can see, theW oscillatory+ behaviour comes from the difference in the energy eigenvalues of ν > and ν > (E WeakE ), Hamiltonian which we interpretFree Hamiltonian as coming fromWeak di Hamiltonianfferent masses for each of themass | 1 | 2 2 − 1 + Flavour eigenstates are a mix of eigenvalues.mass eigenstates A plot of this function is shown in Figure 7 for a particular setUnitary of parameters : ∆m2 =3 10−3eV 2, (like K̄ ⁰ K⁰ and KS KL) 2 × sin (2θ)=0.8andEν =1GeV.AtL = 0, the oscillation probability is zero and the corresponding 2 L π survival probability is one. As L increases the oscillatios begin to switch on until 1.27∆m E = 2 = Relative phases change while propagatingor L = 400 km. At this point the oscillation is a maximum. However, the mixing angle is just Two-Flavor mixingsin2 (2(forθ)=0 simplicity).8 so at maximal mixing, only 80% of the initial neutrinos have oscillated away. As L increases furthur, the oscillation dies down until, around L = 820 km, the beam is entirely composed 2 ! of the initial! neutrino flavour. If sin (2θ)=1.0, the oscillations would be referred to as maximal, " "#$$ $%&meaning$ that' at some point on the path to the detector 100% of the neutrinos have oscillated. • E.g.: 2 families ! $ Definite momentum p; same for ! # !#$%& $ "#$$ " ! ( all mass eigenstate components ! " ! "

( Time development for an initially pure |να> beam: ( ( ) # #$ ! $ #! . &)# ! .& Δm² = 0.003 eV² Amplitude Oscillation##$ % ! # & ) (, Eν = 1 GeV %!" ! ! " !"#$$ " %!' &$%& $ " %!( ( ( ( )'#)( %) ##$ ! ##$ ! $ # $ ! * ( % ( & ( ' (, ( $ ! " "#$ $ " &$%& $ " #$% !" ! -$$./%&0!, !%$!123!$-/3" ##$ ! ##$ ! # Squared mass % & ! !45 # 65 #*' / & "#$2 $2$%& $ ! " # " " $% ! # splitting m" 1 -m2 # ( Baseline, ν energy %) $ #$ !! * ! ( ' " ($ Mixing probability: • Sensitive to oscillation if L/E is ~ ΔmFigure2 7: The oscillation probability as a function of the baseline, L, for a given set of parameters : 2 ( −3 2 (2 ( $(#$' ' ! ! ' ! (!"!%(! %!∆!m! "=3% !(!10"#$$eV$%&, $sin" (2'#θ)=0"#$ .8andEν !=1GeV. " # # " × " ( # Neutrino Physics and Oscillation Alessandro Minotti (CEA - IRFU) 5 As a side( comment, the derivation of the oscillation( probability depends on two assumptions : that ( ( %) ( ( '*(+$%) " "+ # ' ! ! (! $%& (the! $%& neutrino flavour* $%& and( mass! $%& states are mixed and* that-) we create a coherent superposition of mass ! ! " ' # "! ! ! " # states at! ) the$ weak" vertex. This! coherent) $ " superposition,"+ # reflects" the fact that we can’t experimentally resolve which mass state was created at the vertex. One might ask oneself what we would expect to see if we did know which mass state was created at the vertex. If we knew that, we would know the mass of the neutrino state that propagates to the detector. There would be no superposition, no phase difference and no flavour oscillation. However there would be flavour change. Suppose that at the vertex we create a lepton of flavour α and a specific mass state, νk >. Mixing implies that we’ve picked out the kth mass state from the α flavour state. The probability| of doing this is just

Search for Neutrino Oscillations (PDG 1996)< ν ν > 2 = U 2 (47) | k| α | kα ( Exclusion plots ( ( %This) mass state then propagates to the detector, and is detected as a neutrino of flavour β with ' ! !" ' ! # (!"!$%& (! $%&probability* < ν ν > 2 = U 2. The flavour change probability is then the incoherent sum ! ) $ " | β| k | | βk| P (ν ν )mixing = < ν ν >e−iφk < ν ν > 2 = U 2 U 2 (48) ¥ Disappearance: reactor experiments. α → β | β| k k| α | | αk| | βk| Nuclear reactors are most intensive More k k statistics ! ! Ð In the two-flavour approximation, we would have a ν flavour transition probability of sources of νe on Eearth. e (I) With known neutrino flux: measure flux at distance L. excluded Only Ð flux is measured via νe 16 Ð + νe + p ' e + n (II) Measure neutrino flux at position 1 and verify flux after distance L.

More sensitive to small ∆m2 , as longer L can be used due to high flux Baseline longer ¥ Appearance: (Ð) Use neutrino beam of type A (νµ) and (Ð) search at distance L for neutrinos of type B (νe). More! sensitive to small sinθ due to appearance! Neutrino Oscillation

• Real three-families case: PMNS matrix (3 mixing angles, 1 CP-violating phase)

Majorana neutrino?

• 2 mass-squared differences: solar (Δm²₁₂) and atmospheric (Δm²₂₃), with |Δm²₂₃|≫|Δm²₁₂|

Atmospheric sector Super-Kamiokande 848 days Preliminary

2 4 2 4 • For atmospheric/accelerator νμ (L~10 -10 km, Eν~10 -10 MeV)

-

2 ➡ sin (2θ23) > 0.92 (90% CL)

2 -3 2 ➡ |Δm 23| = (2.44±0.06)x10 eV

Neutrino Physics and Oscillation Alessandro Minotti (CEA - IRFU) 6 Neutrino Oscillation

• Real three-families case: PMNS matrix (3 mixing angles, 1 CP-violating phase)

• 2 mass-squared differences: solar (Δm²₁₂) and atmospheric (Δm²₂₃), with |Δm²₂₃|≫|Δm²₁₂|

Solar sector

2 • For solar νe and medium-baseline reactor ν̄e (L~10 km, Eν~MeV)

2 ➡ sin (2θ12) = 0.846 ± 0.021

2 -5 2 ➡ Δm 12 = (7.53±0.18)x10 eV

Neutrino Physics and Oscillation Alessandro Minotti (CEA - IRFU) 7 Neutrino Oscillation

• Real three-families case: PMNS matrix (3 mixing angles, 1 CP-violating phase)

64 our @CERN result vs the world… DC-IV-PRELIMINARY @ CERN • 2 mass-squared differences: solar (Δm²₁₂) and atmospheric (Δm²₂₃), with |Δm²₂₃|≫|Δm²₁₂ CERN superseds MORIOND 64 our @CERN result vs the world… DC-IV-PRELIMINARY @ CERN CERN superseds MORIOND 13 sector 2 Δ σ (~+45%) θ sin (2θ13)=(0.119±0.016) (DYB:DC) ~2.2 ’s

• For short-baseline reactor ν̄e (L~km, Eν~MeV) 2 Δ(DYB:DC) ~2.2σ’s (~+45%) sin (2θ13)=(0.119±0.016)

3 • …and for accelerator νμ (L~10 km, Eν~GeV) NOvA Reactors PDG DC-IV @ CERN ➡ 2 Up to 2010: best limit by CHOOZ (sin (2θ13) < 0.14) NOvA Reactors PDG (Many thanks to NOvA: latest reference) Example:DC-IV NOvA @ CERN

Neutrino Physics and Oscillation Alessandro Minotti (CEA - IRFU) 8 arXiv:1601.05522 DC & beams might prefer a higher θ13? (accepted by PRL) (Many thanks to NOvA: latest reference) (beam “handicapped” by unknowns(δCP) / uncertainties)Example: NOvA

reactor-θ13 key to solve CP-violation & mass hierarchy→ redundancy fundamentalarXiv:1601.05522 DC & beams might prefer a higher θ13? (accepted by PRL) (reactor-(beam “handicapped”θ13 experiments by unknowns( work togetherδCP) / uncertainties) to resolve) Anatael Cabrera (CNRS-IN2P3 & APC) reactor-θ13 key to solve CP-violation & mass hierarchy→ redundancy fundamental

(reactor-θ13 experiments work together to resolve) Anatael Cabrera (CNRS-IN2P3 & APC) PMNS and Beyond

• Standard 3-families oscillation now well-established

- Mixing angles known @ ≲10%, Δm2 @ ~2%: now in the precision era

2 - Still need to find the sign of Δm 13 (mass hierarchy), δCP, octant of θ23

• Neutrino masses call for physics beyond the Standard Model

- Light right-handed neutrinos and Dirac mass terms

- Majorana mass terms, possibly with very heavy neutrinos + see-saw mechanism

• Moreover, some experimental anomalies in the standard neutrino oscillation

- Some quite old (Gallium anomaly, LSND-MiniBooNe results)

- Some more recent (reactor antineutrino anomaly)

Neutrino Physics and Oscillation Alessandro Minotti (CEA - IRFU) 9 Outlook

• Neutrino physics and oscillation

• Reactor neutrinos and the Reactor Antineutrino Anomaly

• The STEREO experiment: principle, configuration, and timeline of the installation and commissioning phase

• The STEREO experiment: status of the analysis of first collected data

Reactor Antineutrino Anomaly Alessandro Minotti (CEA - IRFU) 10 Reactor Antineutrinos

2 -5 2 2 2 20 ∆ m = 8×10 , sin (2θ ) = 0.9, ∆ m = 0.0025, sin (2θ ) = 0.1 • 12 12 13 13 Reactor neutrinos: 2⋅10 /s⋅GW pure ν̄ ₑ e ν 1 →

e ν P 0.8 • Long baseline (~50 km)

0.6

Far detector - “Measure” the oscillation around the Near detector 0.4 first global minimum to determine θ12 ν̄ ₑ ν̄ ₑ 2 ν̄ ₑ and Δm 12 (KamLAND) 0.2

0 3 4 5 10 10 10 • Short baseline (~1 km) L/EL/E (m/eV) (m/MeV)

- Compare oscillated and un-oscillated ν̄ ₑ rate/spectrum (disappearance) at first local minimum to measure θ13 (Double Chooz, Daya Bay, RENO)

• Very short baseline (~10 m)? Study sterile neutrinos

Reactor Antineutrino Anomaly Alessandro Minotti (CEA - IRFU) 11 Reactor Antineutrino Anomaly

• New computation of ν̄ spectrum for Double Chooz RAA 2011: Phys.Rev.D83, 073006 (2011) → 6% excess wrt results from previous reactor experiments

• Results confirmed by Double Chooz, RENO, Daya Bay near detectors Daya Bay 2016: Phys.Rev.Lett.116, 061801 (2016)

Global fit 3+1 arXiv:1703.00860

• Possible explanations

- Problems in the anti-neutrino spectrum prediction

- Oscillation with a sterile neutrino in the eV Δm² scale

Reactor Antineutrino Anomaly Alessandro Minotti (CEA - IRFU) 12 Chinese Physics C Vol. XX, No. X (201X) XXXXXX

the agreement is reasonable in other energy regions. A MeV prompt energy is clearly visible. Two approaches comparison to the Huber+Mueller model yields a 2/dof are adopted to evaluate the significance of this discrep- of 46.6/24 in the full energy range from 0.7 to 12 MeV, ancy. The first method evaluates the 2 contribution of corresponding to a 2.9 discrepancy. The ILL+Vogel each energy bin, model shows a similar level of discrepancy from the data. obs pred Another compatibility test was performed with a Ni Ni 2 i = , modified fitting algorithm. In this method, N(=number N obs N pred ij i i s j of prompt energy bins) free-floating nuisance parameters | | X 2 obs pred 1 obs pred e =(N N )(V ) (N N ). (30) are introduced to the oscillation parameter fit to adjust ij i i ij j j the normalization for each bin, as described in [65]. The 2 2 compatibility was tested by evaluating By definition, i i is equal to the value of defined in Eq. 28. As shown in Fig. 23C, an enhanced contribution 2 = 2(standard) 2(N extra parameters) (29) P Data is visible around 5e MeV. for 20000N degrees of freedom. WeFull obtained uncertainty2/N = In the second approach, the significance of the de- 50.1/25, which is consistent withReactor the results uncertainty obtained viation is evaluated based on the modified oscillation 15000 by the first method using Eq. 28. ILL+Vogel analysis similar to Eq. 29. Instead of allowing all the N nuisance parameters to be free floating, only parame- 10000 Data 80000 (A) Data ters within a selected energy window are varied in the fit. 20000 2 Entries / 250 keV Full uncertaintyIntegrated The di↵erence between minimum s before and after in- 5000 Full uncertainty Reactor uncertainty 60000 troducing these nuisance parameters within the selected 15000 Reactor uncertainty ILL+Vogel energy window was used to evaluate the p-value of the 1.2 2 Prompt Positron4 Energy (MeV)6 8 4000010000 local variation from the predictions. The p-values with Chinese Physics C Vol. XX, No. X (201X) XXXXXX 1.1 1 MeV sliding energy window are shown in Fig. 23C. The Entries / 250 keV Entries / 250 keV 1 IntegratedIntegrated Reactor Anti-Neutrino Anomalies200005000 local significance for a discrepancy is greater than 4 at To calculate the global average independent of the 0.9 the highest point around 5 MeV. In addition, the local 1.2 (Huber + Mueller) Spectral Anomaly

Flux Deficit Ratio to Prediction 0.8 model uncertainty used by the past measurements, we 2 Prompt Positron44 Energy (MeV)66 88 significance for the 2 MeV window between 4 and 6 MeV 1.2 (B) Prompt Positron Energy (MeV) 1 4 2 Prompt 4Energy (MeV) 6 8 2 follow the method described in Ref. [62] by first remov- −1 were evaluated. We obtained a /N value of 37.4/8, 1.1 10 1.0 2 6 ing model from both uncertainties, and define: 10−2 which corresponds to the p-value of 9.7 10 (4.4 ). ) i 1 ∼ χ −3 ⇥ Light Sterile Neutrinos 0 10 exp 2 2 Previous data ( Comparing with the ILL+Vogel model shows a similar = Data / Prediction 0.9 −4 err err model Daya Bay −2 10 0.8 contribution 2 level of local discrepancy between 4 and 6 MeV. World Average −5 Local p-value (Huber + Mueller) χ exp 2 2 (Huber + Mueller) 10 (1 MeV windows)

Ratio to Prediction 0.8 Ratio to Prediction 0.8 = . (18) 1-σ Exp. Unc. −4 cor p cor model −6 1- Flux Unc. 1011 The excess between 4 and 6 MeV was 1.5% of the • If we add a 4th neutrino with a 0.1-1 eV mass, @ very short baseline σ 4 (C) 2 Prompt 44Energy (MeV) 66 88 −11 ⇠ exp and exp now representp experimental uncertainties 0.6 Prompt Energy (MeV) 1010− total observed IBD candidates. An excess of events in err cor 2 3 2 exp 10 10 10 Daya Bay - arXiv:1607.05378 1010−−22 ) ) i only. We then build a covariance matrix V such that Distance (m) i a same energy range was not observed in the spallation ∼ χ ∼ χ 0 1010−−33 (

( 12 Fig. 17.Allowed The Regions: measured reactor⌫ ¯e rate as a function −−44 B beta decay spectrum, ruling out detector e↵ects as exp obs exp obs exp SBL Anomaly (Kopp), 95% CL −2 1010 contribution − V = R R , (19) of the distance from the reactor, normalized to the contribution 2 ij i i,cor j j,cor 2

All Disappearance Exps (Kopp), 95% CL Local p-value νe −−55 Local p-value

¥ χ · · · Fluxχ deficit remains after blinded analysis 1010 an explanation. Adding a simple beta-decay branch or a theoretical predictionGallium Anomaly of(Kopp), Huber+Mueller 95% CL model. −4 1 MeV window (1 MeV windows) obs SBL + Gallium Anomaly (RAA), 95% CL 1010−−66 where Ri is the “ratio” column in Table 11 corrected The rate is correctedDaya Bay Exclusion, by 95% 3-flavor CL neutrino oscil-¥ 2 44 66 88 mono-energetic peak cannot reproduce the observed ex-

] Spectral anomaly questions validity of models obs 2 PromptPrompt EnergyEnergy (MeV) 1.1 by the “Psur” column for the ✓13-oscillation e↵ect.3-familiesRi lations at the distance of each experiment. The cess, indicating that it cannot be explained by a simple

[eV 10 ¥ WeFig. need 23. new (A) data Comparison of predicted and mea- represents the observed rate from each measurement. purple2 14 shaded region represents the global aver- 1 4-families background contribution. Contributions from other in- past m sured prompt energy spectra. The prediction is age∆ and its 1 uncertainty. The 2.4% model un- 13 We then calculate the best-fit average ratio Rg by ¥ Spectral anomaly could point where to look teraction channels (e.g.⌫ ¯e+ C) were investigated and 0.9 2 certaintyReactor Anomaly is shown as a band around unity. The based on the Huber+Mueller model and normal-

pred,new minimizing the function defined as: ) were found to be too small to account for the excess. The 0.8 measurements at the same baseline are combined ¥ Allized !13 to measurements the number of measured at4 LEU power events. The error EXP 1 T.J. Langford - Yale University Date/Seminar 2 past past exp 1 past

/(N events in the energy region around 5 MeV are carefully (R )=(R Ri) (Vij ) (R Rj ), (20) together for clarity. The Daya Bay measurement reactorsbars on the data points represent the statistical 0.7 g g g OBS θee oscillation · N is shown at the flux-weighted baseline (573 m) of uncertainty. The hatched and red filled bands rep- examined: the neutron capture time, the delayed energy 1 Global Best Fits ¥ Model-independent searches are key 0.6 where V is the inverse of the covariance matrix V .This the two near halls. resent the square-root of diagonal elements of the spectrum, and the distance distribution for the delayed procedure yields the best-fit result Rpast =0.942 0.009, 10-1 0.5 g Daya Bay Exclusion ¥ HEUcovariance measurement matrix ( powerful(Vii)) for input the reactor to related neutron capture signal were found to match IBD event where the error is experimental only. ± 0.4 and the full systematic uncertainties, respectively. characteristics. The vertex distribution of the prompt 1 0 1 2 3 4 5 66 Measurement-2 of Reactor-1 Antineutrinospectral anomaly 10 Since10 we now10 use the10 Huber+Mueller10 10 model10 as the 10 10 10 2 1 p Reactor To Detector Distance (m) sin 2θ14 (B) Ratio of the measured prompt energy spec- reference model, we re-evaluate the model uncertainty Spectrum signal was found to be uniform and consistent with IBD T.J. Langford - Yale University 3 trum to the7/30/16 predicted - Neutrinos spectrumin Nuclear Physics (Huber+Mueller events. • Theusing existence the correlated of an and ~eV uncorrelated sterile uncertaintyneutrino is com- supported by other anomalies (LSND, 2 In this section, we extend the study from reactor an- model). (C) The defined distribution (i)of Figure 24 shows the event rate versus time in the MiniBooNE,ponents given Gallium by Ref. [24,anomaly) 25]. Using the weighted av- each bin (black solid curve) and local p-values for 235 238 tineutrino flux to its energy spectrum. The measured energy window of 4.5-5.5 MeV and other windows. erage fission fraction from all experiments ( U: U 1 MeV energy windows (magenta dashed curve). : 239Pu : 241Pu = 0.642 : 0.063 : 0.252 : 0.0425), the prompt energy spectra from the four near-site ADs were e The strong correlation indicates that the excess around • …but a global simple answer is disfavoured by disappearance results See Eq. 30 and relevant text for the definitions. model uncertainty is calculated to be 2.4%, and the final summed and compared with the predictions. The detec- 5 MeV is proportional to the reactor antineutrino flux. tor response of the Daya Bay ADs was studied and used result becomes: Therefore, it strongly suggests that the deviation is due • Current reactor neutrino experiments (Double Chooz,to convert Daya theBay, predicted Reno) antineutrinolittle sensitive spectrum to a to6.3 the Quantification of the Local Deviation 2 past prompt energy spectrum for comparison. A discrepancy to the imperfect modelling of the reactor antineutrino ∆m ~eV-drivenRg =0 .oscillation942 0.009 (exp.) - need0. 023to go (model) closer(21) and be able to see the oscillation ± ± was found in the energy range between 4 and 6 MeV withThe ratio of the measured to predicted energy spectra spectrum. A recent ab initio calculation of the antineu- Finally, we compare the Daya Bay result with the a maximum local significance of 4.4 . The discrepancyis shown in Fig. 23B. The spectral discrepancy around 5 trino spectrum showed a similar deviation from previous past global average. In the previous subsection, we ob- and possible reasons for it were investigated. Reactortained Antineutrino the Daya Anomaly Bay measured reactorAlessandro antineutrino Minotti flux (CEA - IRFU) 13 with respect to the Huber+Mueller model prediction: 6.1 Detector Response 010201-27 RDYB =0.946 0.020(exp.). This result is consistent with The predicted antineutrino flux and spectrum were the past global± average Rpast =0.942 0.009(exp.). If we g ± calculated via the procedure described in Sec. 2.At include the Daya Bay result in the global fit, the new each AD, the reactor antineutrino survival probability average is R =0.943 0.008(exp.) 0.023(model). The g ± ± was taken into account with the best fit oscillation pa- results of the global fit and the Daya Bay measurement 2 2 3 2 rameters, sin 2✓13 =0.084 and mee =2.42 10 eV , are shown in Fig. 17. based on the oscillation analysis| of the| same dataset⇥ [32]. The consistency between Daya Bay’s measurement The relation of the antineutrino spectrum S(E⌫¯ ) and the and past experiments suggests that the origin of the “re- e reconstructed prompt energy spectrum S(Ep) can be ex- actor antineutrino anomaly” is from the theoretical side. pressed as, Either the uncertainties of the theoretical models that predict the reactor antineutrino flux are underestimated S(Ep)= S(E⌫¯e )R(E⌫¯e ,Ep)dE⌫¯e (22) or more intriguingly, there exists an additional neutrino Z oscillation that suppresses the reactor antineutrino flux where R(E⌫¯e ,Ep) is the detector energy response and can within a few meters from the reactor. Such an oscillation be thought of as a response matrix, which maps each an- would imply the existence of one or more eV-mass-scale tineutrino energy to a spectrum of reconstructed prompt sterile neutrinos. To investigate this tantalizing possibil- energies. The energy response includes four main e↵ects: ity, future short baseline (10 m) experiments are required the IBD prompt energy shift, IAV e↵ect, non-linearity, to observe the L/E dependence of such an oscillation. and energy resolution, which are studied in the following.

010201-23 4

tures, and even other experiments. Measurement of the observed event rate as a function of position through detector movement or segmentation will be critical for understanding local background variations on the meter-scale.

III. EXPERIMENTAL SIGNATURE

A. Oscillations of Reactor Antineutrinos

Antineutrinos from reactors are produced as flavor-pure ⌫e in the decays of the neutron-rich fission products in the reactor fuel. More than 99.9% of all ⌫e emitted from commercial reactors are produced within the decay chains of four isotopes, 235U, 238U, 239Pu, and 241Pu. The thermal heat released in the nuclear decays is proportional to the number of emitted ⌫e and is thus a measure of the flux of expected antineutrinos. The spectrum of reactor antineutrinos detected via inverse beta decay has a mean energy of about 4 MeV and extends up to roughly 10 MeV. Neutrinos and antineutrinos are produced as a linear combination of mass eigenstates and their flavor is associated with the accompanying lepton. Due the di↵erence in the mass eigenstates the flavor of observed neutrinos oscillates as a function of baseline and energy. For the three active neutrino states the neutrino mixing parameters are well measured in atmospheric, solar, reactor, and accelerator based experiments. Re- actor ⌫e disappearance over baselines of 1-2 km and 180 km has been observed. The oscillation probability 2 can be parameterized in terms of the mass splitting mij and the mixing angle ✓ij between the ith and jth mass eigenstate. Additional sterile neutrino mass states with m2 1 eV2 beyond the 3 active neutrinos would yield an oscillation e↵ect over meter-long baselines with survival⇠ probability described by

2 ~ 2 2 1.27m41 L ~r Psur(E,L~ ) 1 sin 2✓ee sin | | , (1) Sterile Neutrino Search @' Very Short EBaseline! 2 2 with oscillation parameters m41 and sin 2✓ee. • With…

3 3 arXiv:1212.2182 ×10 ×10 1.4 - A source of neutrinos1.4 1.2 1.2 ✦ Reactor (pure1.0 ν̄ e) 1.0 0.8 0.8 ✦ 0.6 Powerful radioactive0.6 source (νe or ν̄ e) 3.5 3.5 4.0 4.0 0.4 4.5 0.4 4.5

5.0 Events, Unoscillated 5.0

Events, Unoscillated/Bin 0.2 5.5 0.2 5.5 - The capability to measure the rate at 6.0 6.0 0.0 6.5 0.0 6.5 7.0 different (very short)8 7baselines6 7.0 8 7 6 5 4 7.5 5 7.5 Baseline (m) Energy (MeV) 3 2 1 Baseline (m) Energy (MeV) 4 3 2 1 ✦ Movable detector 2 2 2 FIG. 1: Unoscillated (left) versus oscillated (right) detected ⌫e spectra at (m =1.8eV ,sin 2✓ee =0.5) for ✦ a reactor ⌫e detector with realistic parameters as described in Table I. Exaggerated oscillation values are chosen Segmentedfor detector illustrative purposes. or precise The change enough in spectral vertex shape reconstruction from baseline to baseline is a key signature of neutrino oscillations.

2 2 2 • …you can confirmFigure (disproof) 1 illustrates the oscillation oscillation e↵ect @ in Δ baselinem2~1eV and if energy you forobserve (m =1.8 (not eV ,sinobserve)2✓ee=0.5). a The characteristic L/E oscillation is pictured in Figure 2 for the sterile neutrino oscillation parameters significant

✦ Deviation from the 1/R2 flux variation (rate)

✦ Distortion in the energy spectrum @ different distances (shape)

Reactor Antineutrino Anomaly Alessandro Minotti (CEA - IRFU) 14 Sterile Neutrino Search with Sources: SOX

• Main challenges for reactor ν̄ @ very short baseline: cosmics (surface level) and reactor- induced background

• A ν source inside an underground detector profits from strong bkg suppression, but is 17 not as powerful as a reactor (up to the PBq compared to 10 ν̄ ₑ/s MWth)

• CeSOX uses 144Ce and the Borexino detector (β−, 285 days half life)

- LS purification → quasi background-free experiment (104 IBDs/1.5 y vs ~1 accidental/y)

-THEGood SIGNAL spatial IN(12 SOXcm @ (2) 2 MeV) and energy resolution (~3.5% @ 2 MeV)

SOX is at the same time a disappearance experiment and an oscillometry one

144Ce 12.25 m

4.25 m

Reactor Antineutrino Anomaly Alessandro144Ce Minotti (CEA - IRFU) 15

Neutrino 2016 - London July 5th, 2016 M. Pallavicini 12 / 23 Sterile Neutrino Search with Reactors

K. M. Heeger et al., Phys.Rev.D87, 073008 (2013)

Compact-Core Research Reactors Commercial Reactors Short baseline + compact core Lose sensitivity @ low energy Often HEU fuel Fuel evolution (burnup) ~102 MW thermal power ~GW thermal power Less space, more bkg from reactor Possibility of some overburden

Reactor Antineutrino Anomaly Alessandro Minotti (CEA - IRFU) 16 SpectralMo5va5on:&Anomalous&Reactor&Spectrum&Results& Distortion @ 5 MeV Data Mo5va5on:&Anomalous&Reactor&Spectrum&Results&Mo5va5on:&Anomalous&Reactor&Spectrum&Results&20000 Full uncertainty • Data Data Reactor uncertainty Anomalous20000 2000015000 spectralFull uncertainty distortionFull uncertaintyILL+Vogel @ 5MeV in θ13 reactor neutrino experiments DayaReactor Bay uncertaintyReactor uncertainty Reno Double Chooz 15000 1500010000 ILL+Vogel ILL+Vogel Data 20000 Daya DayaBay Bay Reno Reno DoubleDouble Chooz Chooz 10000 10000Entries / 250 keV Full uncertaintyIntegrated 5000 Data Data Reactor uncertainty 20000 Entries / 250 keV 20000 Full uncertainty Entries / 250 keV IntegratedFull uncertaintyIntegrated 5000 500015000 ILL+Vogel Reactor uncertaintyReactor uncertainty 2 4 6 8 15000 150001.2 Prompt ILL+VogelPositron EnergyILL+Vogel (MeV) 10000 2 Prompt Positron4 Energy (MeV)6 8 1.2 1.21.1 2 Prompt Positron4 Energy (MeV)6 8 10000 10000 1.1 Entries / 250 keV 1.1 1 Integrated 5000 Entries / 250 keV 1 Entries / 250 keV 1 Integrated Integrated 5000 50000.9 0.9

0.9(Huber + Mueller)

Ratio to Prediction 0.8 2 4 6 8

(Huber + Mueller) Prompt Positron Energy (MeV) (Huber + Mueller) 1.2

Ratio to Prediction 0.8

Ratio to Prediction 0.8 1 2 4Prompt Positron2 2Prompt4 Energy PositronPrompt4 (MeV)6 4EnergyEnergy (MeV) (MeV)6 8 6 8 8 1.2 1.22 4 6 8 1 1 −1 4 41.1 Prompt2 EnergyPrompt (MeV) 4Energy (MeV) 6 8 10 2 10−1 −1 1.1 1.1 10 10−2

2 ) 2 −2 i 1 10 −2

∼ 10 χ ) −3 i ) 1 i 1 0 10 1406.7763- ∼ χ −3 ∼ χ 0 ( 0 10 10−3 1406.7763-1406.7763- ( ( 0.9 −4 0.9 −2 −4 4 10 2 contribution 0.9 10 −

− 2 10 contribution −2 contribution 5 Local p-value 1511.05849-

2 − (Huber + Mueller) 2 χ 5 Local p-value 1511.05849- − 5 Local p-value 1511.05849- (Huber + Mueller) χ 1508.04233- − 10 (Huber + Mueller) χ (1 MeV windows) Ratio to Prediction 0.81508.04233- 10

4 (1 MeV windows) 10 Ratio to Prediction 0.8 − 1508.04233- (1 MeV windows)

Ratio to Prediction 0.8 −4 −4 −6 2 4 1016 −6 −6 8 101 HEU to LEU Ratio 4 2 4 Prompt2 2 4EnergyPrompt (MeV)Prompt 4Energy6 4Energy (MeV) (MeV)6 8 6 8 101 8 4 1 −1 Prompt EnergyPrompt (MeV)Prompt Energy Energy (MeV) (MeV) 10− 10−1 10 2 2 2 −2 2 10 10−2 10− ) ) i ) i i ⇒ ratio of HEU to LEU spectrum for different hypotheses ∼ χ −3 ∼ χ ∼ 3 0 χ 0 0 10 10− −3 ( 10

( Double Chooz 2014:

Daya-Bay( 2016: Phys.Rev.Lett. RENO 2016: Phys.Rev.Lett. 4 −2 10− 10−4 −4 contribution −2 contribution 2 10

2 − contribution 2 5 Local p-value

− Local p-value

2 5

χ − χ 10 −5 Local p-value

(1 MeV windows) 10 χ −4 −4 Provide-new-tests-of-reactor-models-by-making-precision-(1 MeV windows) 10 116, 061801 (2016)Provide-new-tests-of-reactor-models-by-making-precision-116, 211801(1 MeV windows) (2016) JHEP 10 (2014) 086 −4 Provide-new-tests-of-reactor-models-by-making-precision-10−6 −6 2 2 4 4 6 6 8 8 10 10−6 235 235 ◾ 2 y runtime Prompt2 EnergyPrompt (MeV) Energy4 (MeV) 6 8 235 measurements-of-novel-reactor-spectra,-especially-measurements-of-novel-reactor-spectra,-especially-measurements-of-novel-reactor-spectra,-especially-Prompt Energy (MeV) U-fuel--U-fuel--U-fuel-- ◾ uncertainties: statistical + • Accurate 235U spectrum measurement can reference spectra T.J. LangfordT.J. Langford- Yale University - Yale University 4 4 Date/SeminarDate/Seminar T.J. Langford - Yale University 4 See-P.-Huber’s-talk:-“Reactor-anLneutrino-fluxes-X-status-See-P.-Huber’s-talk:-“Reactor-anLneutrino-fluxes-X-status-Date/Seminar ◾ significance of discrepancy See-P.-Huber’s-talk:-“Reactor-anLneutrino-fluxes-X-status- [5, 7] MeV: - Help find out the nature of the distortionand-challenges”-- and-challenges”-- and-challenges”-- ▸ only 235U: 4.2 σ 3- 3- ▸ no excess in HEU: 5.5 σ 3- - Constrain existing reactor models ◾ significance including energy resolution: ▸ 235 σ C.Buck et al., Phys.Lett.B765, 159 (2017) only U: 3.7 arXiv:1512:xxx ▸ no excess in HEU: 4.7 σ C. Buck, A.P. Collin, J. Haser, M. Lindner Reactor Antineutrino Anomaly Alessandro Minotti (CEA - IRFU) 17

2015/12/07 15 / 16 Signal and Background

• Signal: inverse beta decay (IBD) • Background: correlated

- From cosmic μ spallation (fast n, n- γ from showers) and reactor (fast n) n

- Online rejection: overburden, active p + μ veto νe + p → e + n - Offline rejection: topology, PSD, subtraction from reactor-off periods • Signature: two-fold coincidence • Background: accidental coincidences - Prompt: e⁺ ionisation+annihilation

(Ee+ = 1-8 MeV, Eν ≃ Ee+ + 0.8 MeV) - Reactor related (nth leakage, high-E γ’s from n-capture on metals, 16N, ecc.) - Delayed: n-Capture - On-site measurements ✦ n+Gd→Gd*+γ’s (8 MeV) - Rejection with shielding and n ✦ n+6Li→α+3H (4.78MeV) topological cuts, estimation with off-time windows

Reactor Antineutrino Anomaly Alessandro Minotti (CEA - IRFU) 18 A Worldwide Effort

Neutrino4 SoLiδ DANSS Prospect NuLat Stereo

Chandler NEOS

Chandler

Reactor Antineutrino Anomaly Alessandro Minotti (CEA - IRFU) 19 A Worldwide Effort

Core PTh Core Size Overburden Segmentation Baseline Material

Chandler 72 MW (235U) ⦰ = 50 cm ~10 mwe 6.2 cm (3D) 5.5 m PS + Li layer

h = 3.6 m DANSS 3 GW (LEU) ~50 mwe 5 cm (2D) 10.7-12.7 m Gd-doped PS ⦰ = 3.1 m h = 3.7 m NEOS 2.8 GW (LEU) ~20 mwe - 23.7 m Gd-doped LS ⦰ = 3.1 m 35x42x42 Neutrino4 90 MW (235U) few mwe 22.5 cm (2D) 6-12 m Gd-doped LS cm3 40/1790 MW NuLat few mwe 6.35 cm (3D) 4.7/24 m Li-doped PS (235U/LEU) h = 0.5 m Prospect 85 MW (235U) few mwe 15 cm (2D) 7 m Li-doped LS ⦰ = 0.2 m

SoLiδ 72 MW (235U) ⦰ = 0.5 m ~10 mwe 5 cm (3D) 5.5 m PS + Li layer

Stereo 58 MW (235U) ⦰ = 37 cm ~15 mwe 25 cm (1D) 8.9 m Gd-doped LS

Reactor Antineutrino Anomaly Alessandro Minotti (CEA - IRFU) 20 Comparing Different Technologies

• What segmentation?

- No segmentation - compare a unique antineutrino spectrum with prediction, requires high statistics and accurate prediction (e.g. NEOS)

- “Coarse” segmentation - compare spectra in different segments to look for a distortion induced by oscillation in sterile neutrino (e.g. Stereo)

- “Fine” segmentation - allows in addition for a rejection of background induced by cosmic muons thanks to topology (e.g. SoLid)

- Ultimate cell size limited by dead matter and harder inter-calibration 38th$Interna0onal$Conference$on$High$Energy$Physics$ IBD Candidates August$3$>$10,$2016,$Chicago$$$

6 Backgrounds - Correlated Detection Technology • IBD analysisLiF:ZnS(Ag) techniques 250 um developed → cross checked with simulation ν̄ e • EM event and neutron produced in same process. Studied using reactor off data, e.g: • Granularity of the detector allows detailed topological studies !SoLid:!A!compact!detector!for!very!! • λ-shifting 3x3 mm2 short5baseline!neutrino!experiments!Muon spallation in the detector - combat with muon ID (energy and channel topology) • Neutrinos seen via inverse beta decay • High energy neutron - combat with multiplicity selections (proton recoils) (IBD) events (unique topology): PVT + νe + p → e + n

Leonidas)N.)Kalousis)(VUB)) for$the$SoLid$collabora0on$ • + Prompt e scintillation signal: [email protected]) • Energy deposition in small cluster of IBD Correlated bkg cubes, away from annihilation ɣs 5 • Manageable containment of ɣs leakage/ 5 Background candidates. Neutrons in red, EM signals use colour scale. IBD candidates from SM1. Neutrons in red, EM signals use colour scale Left: muon spallation event (Data). Right: cosmic neutron event (Sim). Left: isolated candidate (waveforms above). Right: candidate with accidental gammas - can be used in analysis pileup - technological advantage Reactor Antineutrino Anomaly Alessandro Minotti (CEA - IRFU) SoLid: Recent Analysis Results 2114 [email protected] SoLid: Recent Analysis Results 12 [email protected]

• Delayed n signal from 6LiF:ZnS(Ag): • Spatially near the positron Advertisement • Distinguished from PVT via pulse Dedicated talk on SoLid technology: shape discrimination Leonidas Kalousis, 5th Aug 12:00 Detector R & D session

SoLid: Recent Analysis Results 6 [email protected] Comparing Different Technologies

• Which n-capturer?

- Gd is well-established - high released energy and Cross Section (e.g. Neutrino-4), stable mixing in the compound

- Li - Localised Edep: quenched but allows selection of delayed via PSD (e.g PROSPECT)

• Plastic or liquid scintillator? ✦ n+Gd→Gd*+γ’s (8 MeV) - PS - better for segmentation, less “edge effects” ✦ n+6Li→α+3H (4.78MeV) - LS - allows for larger volumes PROSPECT-50 Demonstrator

PROSPECT - A Precision Reactor Neutrino 252Cf ¥ In operation since March 2016, near continuousOscillation and Spectrum Experiment data-collection K.M. Heeger on behalf of the PROSPECT collaboration (n,Li) ¥ Measured light collection with 6LiLS: >550PE/MeVWright Laboratory, Department of Physics, Yale University See also: P03.053, P03.054, P03.056 Neutrino Anomalies & Sterile v Hypothesis Reactor Neutrino Anomalies ¥ 5% energy resolution at 1MeV MiniBoone Ga Anomaly Cosmology (WMAP) Flux Deficit Spectral Deviation

¥ Measured PSD Figure of Merit: 1.25 at (n,Li) 6m capture R=0.86±0.05

Anomalies in Neutrino Data If new oscillation signal, E(MeV) ¥ Losc =2.5 >99.9% background rejection implies Δm2 ~ O(1eV2) and sin22θ > 10-3 m2(eV 2) Extra neutrino oscillations or New feature in 4-6 MeV LSND (νe appearance) artifact of flux predictions? region of spectrum. ➔ very short baseline oscillation for reactor ν 2 10 m MiniBoone (νe appearance) ' Ga calibration source anomaly Systematics or experimental effects? Understanding reactor flux and spectrum anomalies requires new ¥ Double-ended readout “Light sterile neutrinos: Neff in cosmology ➔ test each experimental effect. reactor measurements A white paper”, Phys.Rev.Lett. 116 (2016) 6, 061801 Short-baseline reactor anomaly (νe disappearance) arXiv:1204.5379 Daya Bay collaboration ¥ Position reconstruction along cell length High Flux Isotope Reactor, ORNL PROSPECT Detector System as a Compact Antineutrino Source A Segmented, 6Li-Loaded Detector (b) AD-I Active Volume (d) AD-I Segments and Containment High Flux Isotope Reactor, ORNL HEU Reactor Spectrum (a) PROSPECTMeV Phase I

Reactor

Shielding 78.74 47.24 2000 1200 200.0 200.0 cm 120.0 120.0 cm 22

SectionSECTION A-A A-A AD-I LiLS (c) AD-I Unit Segment
DN Na z-scan Passive Segments 120
DN cm Shielding -JRVJE
4DJOUJMMBUPS

14.614.6#cm# cm 14.6#cm# 66.93 57.50 Separator 1700 1460.5 A A 146.5 146.5 cm Reactor Antineutrino Anomaly Alessandro MinottiHFIR (CEA - IRFU 170.0 cm ) 22 Core 14.4#cm#14.4#cm# 14.6#cm# 14.6#cm#

14.4#cm#

14.4#cm#

Corner 14.6 cm 175.369.00 cm Rod 1752.6 235 a# a# 90.55 HEU core provides static spectrum of mainly U. b#b# 230.0 2300 cm

194#cm#194#cm# power: 85 MW (research) c# c# 119#cm#119#cm# Compact reactor core 6 Antineutrino Event Identification fuel: highly enriched - 3000L of Li liquid scintillator - 100 scintillator loaded cells, ~15x15x120cm with 6Li Doped Scintillator uranium (235U) - double ended PMT readout, light guides, core shape: cylindrical <4-5%/√E resolutions size: h=0.5m r=0.2m - thin optical separators, minimal dead (compact) material - containment vessel, filled in place duty-cycle: 41% 40μs delayed n capture Nucl. Instrum. Meth. A806 (2016) 401Ð419, Compact core (< 1m) avoids arXiv:1506.03547, signal inverse beta decay (IBD) Pulse Shape Discrimination oscillation washout PROSPECT collaboration γ-like prompt, n-like delay backgrounds fast neutron simulation extrapolation to Phase I n-like prompt, n-like delay Background Rejection ViaIBD-like Segmentation n capture accidental gamma 9 neutron-coincident events 2.5 8 γ-like prompt, γ-like delay 5 7 10 before cuts segment z 2.0 Background Reduction Steps n+H 6 9 (1), (2), (3) 5 1.5 2.5 4 rate [mHz/segment] 8 10 (4), (5) 4 Background reduction is key challenge ¥ Efficient PSD and neutron 3 1.0 (6) 7 3 PSD 2 segment z 2.0 10 0.5 tagging 1 6 12C inelastic 0 2 0 2 4 6 8 10 0.0 5 1.5 10 segment x

¥ Identification of multiple rate [mHz/segment] shower veto active veto requirements: 4 particle interactions Event rate [mHz/MeV] 10 3 1.0 neutrinos! neutron capture 2 PROSPECT Physics 1 topology • 0.5 ¥ Fiducialization • recoil PSD 1 10−1 fiducialization gamma/electron energy 0 • 0 2 4 6 8 10 0.0 Precision Oscillation Experiment Precision Spectrum Experiment Active suppression by >3 10−2 = same properties as segment x 0 2 4 6 8 10 Representative 500 MeV primary 235 orders of magnitude prompt ionization [MeV] detector bulk; use same 4σ test of best fit after 1 year Measurement of U spectrum technology and fiducialize. Ref: PROSPECT collaboration 20 >3σ test of favored region after 3 years 5σ test of allowed region after 3+3 years PROSPECT Detector and Shielding Development IBD signal backgrounds after analysis cuts PROSPECT-0.1 5cm length (a) PROSPECT PhasePROSPECT I ! (b) Antineutrino AD-I Detector (AD) (d) AD Cells and Containment Characterize LS 0.1 liters 6 10x10 segments between 2-6 MeV:

LS, LiLS

Aug 2014-Spring 2015 RAMF Reactor1.2m length average stat. Shielding! 78.74 47.24 2000 1200 200.0 cm PROSPECT-2 Building the shielding ~3 tons 120.0 cm precision < 1.5%, 6 Background studies 12.5 cm length LiLS systematics < 2% Dec 2014 - Aug 2015 1.7 liters PROSPECT (c) AD Unit Cell SectionSECTION A-A A-A 6 
DN LiLS RAMF AD! 
DN Passive PROSPECT -JRVJE
4DJOUJMMBUPSPhase I AD-I

PROSPECT-20 Shielding! Comparison of different reactor models 14.614.6#cm# cm 1m length 14.6#cm# 66.93 57.50 Segment characterization Separator 1700 1460.5 A A 146.5 cm HFIR 170.0 cm phase I

23 liters Scintillator studies 14.414.4#cm# cm T.J. Langford - Yale University 4 December Workshop - ORNL reactor 14.4#cm# Core LS, 6LiLS ! 14.6#cm# 14.6#cm# core 14.4#cm# Background studies 14.4#cm#

Corner! 14.6 cm 69.00 175.3 1752.6 cm T.J. Langford - Yale University 7/30/16Rod 14.4 cm - Neutrinos in Nuclear Physics a# a# phase90.55 II Spring/Summer 2015 22 b#b# 230.0 2300 cm

194#cm#194#cm# 1x2 segments c# 119#cm#119#cm# PROSPECT-50 c# Daya Bay 1.2m length Baseline design 50 liters prototype 6LiLS Winter 2015 PROSPECT Publications arXiv: 1506.03547, 1508.06575, 1512.02202

prospect.yale.edu P3.055 Some Results

• NEOS recently released a revisited paper with an exclusion plot arXiv:1610.05134

• Very high statistics, systematics driven by comparison with Daya Bay spectrum

• Neutrino-4 also recently released new results arXiv:1702.00941

Reactor Antineutrino Anomaly Alessandro Minotti (CEA - IRFU) 23 Outlook

• Neutrino physics and oscillation

• Reactor neutrinos and the Reactor Antineutrino Anomaly

• The STEREO experiment: principle, configuration, and timeline of the installation and commissioning phase

• The STEREO experiment: status of the analysis of first collected data

STEREO: Principle and Installation Alessandro Minotti (CEA - IRFU) 24 Chasing the Light Sterile Neutrino with the Stereo Experiment David Lhuillier* and Alessandro Minotti*, IRFU, CEA-Saclay, France STEREOon behalf of the STEREO collaboration Collaboration

Supported by

STEREO: Principle and Installation Alessandro Minotti (CEA - IRFU) 25 STEREO @ the ILL Research Reactor

• The STEREO detector is located the ILL research reactor facility, Grenoble

- [8.9–11.1] m from the reactor core, which is..

- ..compact (⦰ = 37 cm), with a 93% 235U fuel, and 58.3 MWth nominal power

- Under a water channel (~15 mwe overburden from cosmics)

STEREO: Principle and Installation Alessandro Minotti (CEA - IRFU) 26 The Detector of STEREO

• The STEREO inner detector consists of:

- 6 Target cells, filled with Gd-doped LS - ν target

- 4-sided Outer Crown, filled with un-doped LS - reduce edge effects, tag external bkg Stereo detector - 12 upper acrylic Buffers, filled with mineral oil, housing 48 PMTs - optical coupling Two sub-volumes : Target (for• IBD)Walls/separators: segmented in 6 VM2000 identical cells+ air-gap + acrylic sandwiches - transport light to PMTs Gamma-catcher to collect escaping gamma, improve eciency and energy resolution

Acrylic buffer Buffer oil

Gamma-catcher outer crown filled with liquid scintillator (no Gd) 1.5m Target 6 cells filled with Gd-loaded liquid scintillator

2 mm thick acrylic plate

3m arXiv:1604.08877 Reflective foil Bridal veil VM2000

48 PMTsSTEREO: : 4 PMTs Principle per and Target Installation cells and 4 or 8Alessandro PMTs per Minotti Gamma-catcher (CEA - IRFU) cells 27 Cell in acrylic with multi-layer wall of VM2000 and air for total reflection improved light collection ∆ Detector assembled in spring 2016 and tested without liquid !

T.Salagnac Stereo - AAP 2016 1st December 2016 4 / 17 Background and shielding Neutron and “ background measurement : Fast n from reactor and n-guide Thermal n from neighboring experiments n-capture causing “ emission Background Mitigation - Reactor∆ External shielding :

• Full on-site measurements of background (µ, n, γ) Core

• Reactor background from core and neighbouring experiments reduced with shielding

- μ-metal (magnetic fields)

- B-loaded polyethylene + B4C (neutrons) Beam stop

Concrete - Lead ( -rays) γ Lead

B4C - Soft iron (magnetic fields) Polyethylene

Magnetic field from IN20 gain variation of PMTs over time ∆ PE µ rate and distribution measurement μ Veto Simulation of the “, n, µ and magnetic field backgrounds Soft Iron μ -metal design andB4C validation of the detector shielding Detector ∆ Lead T.Salagnac Stereo - AAP 2016 1st December 2016 7 / 17

STEREO: Principle and Installation Alessandro Minotti (CEA - IRFU) 28 Background Mitigation - Cosmics

• Cosmic background rejected with μ veto and pulse shape discrimination (PSD), and subtracted statistically from reactor-off periods Muon veto • Muon Veto: water Cherenkov tank to detect muons

Both the light output and the decay time of organic - 20 PMTs scintillators depend on the ionization density. Muon veto : Water cerenkov tank to detect muons - Tyvek reflective sheets to enhance light collection with 20 PMTs and Tyvek sheets for reflectivity

Studies : Several prototypes tested before final instrument

Decay time in stilbene for various particles Geometrical eect vs µ eciency (from Bollinger and Thomas)

Lower “ sensitivity with 4 PMTs charge trigger • PSD: disentangle particles with large dE/dx (protons, α’s) that interact in the scintillator

- Intrinsic property of organic scintillators

- Use Qtail/Qtot ratio (output of the acquisition)

Introduction to Radiation Detectors and Electronics Copyright  1998 by Helmuth Spieler STEREO: Principle and Installation Alessandro Minotti (CEA - IRFU) III. Scintillation Detectors 29

T.Salagnac Stereo - AAP 2016 1st December 2016 9 / 17 Calibration

• Different calibration techniques to calibrate/monitor detector response and stability - Radioactive sources deployed inside the cells @ different heights, and along and underneath the detector via dedicated systems - LED light injected in the detector through fibres - Natural events (spallation n’s, cosmogenic 12B)

• Each technique covers different aspects - LEDs: calibration of PMTs, linearity, study liquid properties and geometrical effects - n’s: energy response stability, background topology, IBD detection efficiency, PSD - γ sources/12B decays: energy calibration, energy scale (quenching curve)

STEREO: Principle and Installation Alessandro Minotti (CEA - IRFU) 30 Physics Goals

• Looking for oscillation patterns: relative shape in the 6 identical cells (different L)

• STEREO aims to cover the RAA region in 300 days of data @ nominal reactor power

• Proposal conditions:

- 400 ν/day in 300 days - S/B = 1.5

- Ee+ > 2 MeV, En > 5 MeV - εdet = 60%

- δEscale = 2%, δΦν = 4%

— No Oscillation — Closest Cell

— Furthest Cell Number of Events of Number

STEREO: Principle and Installation Alessandro Minotti (CEA - IRFU) 31 STEREO Inner Detector

• Steel tank delivered in March 2016

• VM2000-acrylic sandwiches prepared in Saclay, final vessel delivered in February 2016

• Buffers, each equipped with four 8” PMT’s, assembled on March 2016

STEREO: Principle and Installation Alessandro Minotti (CEA - IRFU) 32 Detector Integration and Moving

• All components were integrated by April 2016 @ LPSC, Grenoble

• DAQ tests and LED runs taken with the empty detector

STEREO: Principle and Installation Alessandro Minotti (CEA - IRFU) 33 Detector Integration and Moving

• STEREO detector and its lead shielding were transported from LPSC to ILL and moved in the 11th of May Background and shielding • May-September 2016 - works inside ILL on shielding (Stereo layers, external walls) Detector shielding : 6tonsofboratedpolyethylene 65 tons of lead

B4C sheet all around the detector structure µ-Veto Magnetic shielding (soft iron + µMetal)

Lead

Soft iron

B4C

STEREO: Principle and Installation Alessandro Minotti (CEA - IRFU) 34

August 2016 : Assembly of the shielding and the detector complete September 2026 : Detector moved to its data-taking position !

T.Salagnac Stereo - AAP 2016 1st December 2016 8 / 17 Detector Integration and Moving

• October 2016 - STEREO moves on air cushions to its final position

• November 2016 - we receive ASN approval → start filling and commissioning

- ~2 days for the filling procedure

- ~2 weeks for the commissioning

STEREO: Principle and Installation Alessandro Minotti (CEA - IRFU) 35 Outlook

• Neutrino physics and oscillation

• Reactor neutrinos and the Reactor Antineutrino Anomaly

• The STEREO experiment: principle, configuration, and timeline of the installation and commissioning phase

• The STEREO experiment: status of the analysis of first collected data

STEREO: First Results Alessandro Minotti (CEA - IRFU) 36 Commissioning Phase Electronics • The commissioning of STEREO started the 10th November 2016 Dedicated electronics hosted in a single • Main tasks: µ-TCA crate Two programmable levels of trigger - Define trigger thresholds, acquisition conditions(FPGA) ✓ : I 1st level per front-end board - Find a cell-by-cell preliminary pe-MeV conversion ✓ trigger on single PMT, sum 4 or 8 PMTs∆ - Define a Qtot and Qtail window in the acquisition for the PSD ✓ I 2nd level on the trigger board trigger between the dierent ∆ • Acquisition has 0 deadtime up to few kHz → setsub-detectors trigger threshold (target, gamma-catcher,as low as ∼250 keV veto) hQ_TG_full

• 1kHz without deadtime TG has trig

Take regular runs saving waveforms ) 10 -1 Debug mode : save pulse on disk but with a large deadtime && no muon trig && QTG > 0.5*QDet

1 Counts (s Counts Counts [1/(s.bin)]

10−1 STEREO Preliminary March 2017 10−2

10−3

3 10−4 ×10 0 20 40 60 80 100 120 140 160 180 QDet [ADC] Qdet (ADC)

STEREO: T.SalagnacFirst Results AlessandroStereo - AAP Minotti 2016 (CEA - IRFU) 1st December 2016 10 / 17 37 Commissioning Phase

• A few problems were encountered

- The buffers above one of the 6 Target (4th) and GC cells (front one) leaked oil in the LS

̣ No effect on light propagation (attenuation length)..

̣ ..but light collection efficiency (pe/MeV) for involved cell has ~halved

- Separation walls showed a gradual degradation of their reflective power

̣ No sizeable effect on global light collection (absorption is ~0)..

̣ ..but these Light Leaks (LL) affect cells inter-calibration and energy reconstruction

STEREO: First Results Alessandro Minotti (CEA - IRFU) 38 Calibration Sources

• STEREO Target and GC cells are calibrated on a regular basis to follow their evolution

- Energy response monitoring each ~2 days (week): 54Mn internal (external) calibration

- Full calibration campaign each ~3 weeks: full scan - internal and external - with all sources

- PSD/n-detection tuning weekly: AmBe scan - internal, external, underneath

Source Output 68Ge 2x0.51 MeV γ 137Cs 0.66 MeV γ 54Mn 0.83 MeV γ 65Zn 1.12 MeV γ 124Sb 0.60 & 1.69 MeV γ 60Co 1.17 & 1.33 MeV γ 24Na 1.37 & 2.75 MeV γ AmBe 4.4 MeV γ + n 252Cf fission γ’s & n’s

STEREO: First Results Alessandro Minotti (CEA - IRFU) 39 Cell 1 Cell 2 Cell 3 Internal Calibration χ2 / ndf 16.68 / 7 χ2 / ndf 3.149 / 7 χ2 / ndf 14.92 / 7 Const 6785 ± 40.4 Const 202.7 ± 7.2 Const 701.3 ± 13.1 8000 Mean 196.3 ± 0.2 Mean 209.8 ± 0.9 Mean 196.2 ± 0.5 Mn-54 @ 10 cm 250 900 Sigma 21.72 ± 0.26 Sigma 21.7 ± 1.5 Sigma 20.73 ± 0.71 Mn-54 @ 45 cm • 7000 800 Internal calibration: 3 tubes in the Target (Cells 1,4,6) - γ can travelMn-54 @to 80 cm neighbouringχ2 / ndf cells 4.627 / 7 χ2 / ndf 3.815 / 7 χ2 / ndf 6.029 / 7 Const 7733 ± 44.1 200 Const 257.1 ± 7.9 Const 945.5 ± 15.3 6000 Mean 203.3 ± 0.1 Mean 213.3 ± 1.0 700 Mean 201.1 ± 0.3 Sigma 20.35 ± 0.20 Sigma 23.19 ± 1.57 Sigma 18.42 ± 0.45 600 5000 • Relative z-dependence measured by deploying at different height 150 χ2 / ndf 18.17 / 7 χ2 / ndf 6.323 / 7 χ2 / ndf 15.57 / 7 Const 6134 ± 38.0 Const 191.8 ± 6.8 500 Const 667.5 ± 12.8 4000 Mean 205.6 ± 0.2 Mean 215 ± 1.1 Mean 201.9 ± 0.5 Sigma 22.81 ± 0.31 Sigma 21.37 ± 1.51 400 Sigma 21.51 ± 0.81 54 100 • pe/MeV has a factor 0.4 for Cell 4 wrt other cells @ Mn γ energy3000 (0.835 MeV) 300 2000 50 200 • ~300 pe/MeV when summing up all cells (stable transparent1000 liquid, λatt above 5 m) 100

100 120 140 160 180 200 220 240 260 280 300 0 100 200 300 400 500 0 100 200 300 400 500 Cell Qtot (pe) Cell Qtot (pe) Cell Qtot (pe)

Cell 1 2 CellCell 2 4 22 CellCell 3 5 2 Cell 6 χ / ndf 16.68 / 7 χχ / ndf/ ndf 3.149 111.7 / 7 / 7 χ / χndf2 / ndf 14.92 6.803 / 7 / 7 χ2 / ndf 24.45 / 7 Const 6785 ± 40.4 ConstConst 1.429e+04 202.7 ± 6.336e+01± 7.2 ConstConst 701.3 1050 ± 13.1 ± 15.8 Const 4986 ± 35.6 8000 Mean 196.3 ± 0.2 250 MeanMean 209.8 80.14 ± 0.9± 0.06 MeanMean 196.2 198.6 ± 0.5 ± 0.5 Mean 183.8 ± 0.1 Mn-54 @ 10 cm 900 6000 — 54Mn @ cell bottom Sigma 21.72 ± 0.26 14000 SigmaSigma 21.7 12.85 ± 1.5± 0.05 1200 SigmaSigma 20.73 23.94 ± 0.71 ± 0.85 Sigma 19.12 ± 0.21 Mn-54 @ 45 cm 7000 — 54Mn @ cell middle 800 Mn-54 @ 80 cm 2 / ndf 4.627 / 7 2 2/ ndf/ ndf 3.815 61.71 / 7 / 7 2 / ndf2 6.029 / 7 2 — 54Mn @ cell top χ χχ χ χ / ndf 7.239 / 7 5000 χ / ndf 32.78 / 7 Const 7733 ± 44.1 12000200 ConstConst 1.477e+04 257.1 ± 6.525e+01 ± 7.9 1000 ConstConst 945.5 1299 ± 15.3 ± 17.5 Const 6525 ± 40.2 700 6000 Mean 203.3 ± 0.1 MeanMean 213.3 83.54 ± 1.0± 0.05 MeanMean 201.1 203.8 ± 0.3 ± 0.4 Mean 191.8 ± 0.1 Sigma 20.35 ± 0.20 SigmaSigma 23.19 12.83 ± 1.57 ± 0.05 SigmaSigma 18.42 ± 200.45 ± 0.5 Sigma 18.19 ± 0.17 10000 600 5000 800 4000 2 150 22 2 χ / ndf 18.17 / 7 χχ / ndf/ ndf 6.323 77.48 / 7 / 7 χ / ndf2 15.57 /9.5 7 / 7 2 13.57 / 7 500 χ / ndf χ / ndf Const 6134 ± 38.0 8000 ConstConst 1.207e+04 191.8 ± ±5.625e+01 6.8 ConstConst 667.5 958.2 ± 12.8 ± 15.3 Const 5844 ± 37.9 4000 Mean 205.6 ± 0.2 MeanMean 215 84.95 ± 1.1± 0.08 600 MeanMean 201.9 206.6 ± 0.5 ± 0.5 3000 Mean 197.5 ± 0.2 Sigma 22.81 ± 0.31 SigmaSigma 21.37 14.92± 1.51 ± 0.08 400 SigmaSigma 21.51 23.94 ± 0.81 ± 0.88 Sigma 21.62 ± 0.27 3000 1006000 1 2 3 4 5 6 300400 2000 2000 4000 50 200 STEREOarch Preliminary 2017 200 1000 1000 M 2000 100 γ 100 120 140 160 180 200 220 240 260 280 300 020 40100 60 20080 300100 120400 140 500160 0 0 100100 200200 300300 400400 500500 0 100 200 300 400 500 Cell Qtot (pe) CellCell Qtot Qtot (pe) (pe) CellCell Qtot Qtot (pe) (pe) Cell Qtot (pe)

Cell 4 Cell 5 Cell 6 χ2 / ndf 111.7 / 7 χ2 / ndf 6.803 / 7 χ2 / ndf 24.45 / 7 STEREO: First Results Alessandro Minotti (CEAConst 1.429e+04 - IRFU ± 6.336e+01) Const 1050 ± 15.840 Const 4986 ± 35.6 Mean 80.14 ± 0.06 Mean 198.6 ± 0.5 Mean 183.8 ± 0.1 6000 14000 Sigma 12.85 ± 0.05 1200 Sigma 23.94 ± 0.85 Sigma 19.12 ± 0.21

2 / ndf 61.71 / 7 2 2 χ χ / ndf 7.239 / 7 5000 χ / ndf 32.78 / 7 12000 Const 1.477e+04 ± 6.525e+01 1000 Const 1299 ± 17.5 Const 6525 ± 40.2 Mean 83.54 ± 0.05 Mean 203.8 ± 0.4 Mean 191.8 ± 0.1 Sigma 12.83 0.05 10000 ± Sigma 20 ± 0.5 Sigma 18.19 ± 0.17 800 4000 χ 2 / ndf 77.48 / 7 χ2 / ndf 9.5 / 7 χ2 / ndf 13.57 / 7 8000 Const 1.207e+04 ± 5.625e+01 Const 958.2 ± 15.3 Const 5844 ± 37.9 Mean 84.95 ± 0.08 600 Mean 206.6 ± 0.5 3000 Mean 197.5 ± 0.2 Sigma 14.92 ± 0.08 Sigma 23.94 ± 0.88 Sigma 21.62 ± 0.27 6000 400 2000 4000

200 1000 2000

20 40 60 80 100 120 140 160 0 100 200 300 400 500 0 100 200 300 400 500 Cell Qtot (pe) Cell Qtot (pe) Cell Qtot (pe) External Calibration

• External calibration: Motorised systems that moves the source around the detector

• γ peak selected in closest 2 PMTs → complete mapping

• pe/MeV has a factor 0.5 for GC Front (H) wrt GC Back (K) @ 54Mn γ energy (0.835 MeV)

• Some difference between GC IN20 (L) / D19 (M), symmetry along same GC segment (I,J) Data: selection by PMs to obtain Qtot for each point

Point H Point I Point J

STEREO Preliminary March 2017 GC back and front difference

Reactor Point K Point L Point M

V. Sergeyeva STEREO meeting 08/02/2017 33

STEREO: First Results Alessandro Minotti (CEA - IRFU) 41

06/01/2017 06/01/2017

QBACK / QFRONT = 2 and this ratio is maintained in time

V. Sergeyeva STEREO meeting 08/02/2017 10 Energy Response vs Time and Light Leaks

• Evolution of light (Npe) collected in Target cells is due to increasing Light Leaks (LL)

• Origin still unknown and under investigation - may be due to liquid leak inside the walls

• Does not affect the global light collection: LL depends on z position, but z relative peak is stable in time when considering the whole detector

• We developed 2 methods to track variation of LL regularly during the data taking

E peak for the 4 PMTs of Cell 1 220

215 Mn-54 in Cell 1 Bottom peak (pe)

γ Mn-54 in Cell 1 Middle 210

Mn Mn-54 in Cell 1 Top 54 205 Full E peak ratio 100cm/800cm - Cell 6 200 1 195 0,98

190 0,96 Q_Cell 0,94 185 Q_Target 0,92 STEREO Preliminary Q_FullDet 180 March 2017 0,9 Ratio of full E peaks 0 10 20 30 40 50 175 STEREO Preliminary 01/12 15/12 29/12 12/01 26/01 09/02 23/02 March 2017Day after Nov. 10, 2016 Date

STEREO: First Results Alessandro Minotti (CEA - IRFU) 42 Measuring the Light Leaks

• To calculate LL: search for full energy deposit in cell i and measure charge in cell j

- Using physics runs: fit Qj in lower band of Qi vs Qj distribution (hourly, non invasive)

54 - Using Mn calibration runs: select events with full-energy photoelectron in i (Qi > μQ+σQ) and fit Qj → μQj = LLij (allows to measure the z-dependence of LL)

GC Front to Cell 1 hLeaks_GCFronttoTgCell1_Pos10hLeaks_GCFronttoTgCell1_Pos45hLeaks_GCFronttoTgCell1_Pos80 hLeaks_GCFronttoTgCell2_Pos10hLeaks_GCFronttoTgCell2_Pos45hLeaks_GCFronttoTgCell2_Pos80 hLeaks_GCFronttoTgCell3_Pos10hLeaks_GCFronttoTgCell3_Pos45hLeaks_GCFronttoTgCell3_Pos80 hLeaks_GCFronttoTgCell4_Pos10hLeaks_GCFronttoTgCell4_Pos45hLeaks_GCFronttoTgCell4_Pos80 hLeaks_GCFronttoTgCell5_Pos10hLeaks_GCFronttoTgCell5_Pos45hLeaks_GCFronttoTgCell5_Pos80 hLeaks_GCFronttoTgCell6_Pos10hLeaks_GCFronttoTgCell6_Pos45hLeaks_GCFronttoTgCell6_Pos80 Entries 150116451284 Entries 15011645 1284 Entries 150116451284 Entries 16451284 1501 Entries 15011645 1284 Entries 15011645 1284 Gc Front 0.12 Gc Back 0.12 0.12 0.12 0.12 0.12 χ2 / ndf 14.94 / 9 Mean 0.086680.079710.07401 χ2 / ndf 15.37 / 9 Mean 0.014580.01414 0.01323 Mean 0.009974 0.010440.01053 Mean 0.0042710.004523 0.004283 Mean 0.0099220.009765 0.009661 Mean 0.009781 0.01126 0.01022 RMS 0.048580.048240.04523 RMS 0.011960.01053 0.01016 RMS 0.0080360.007043 0.00763 RMS 0.0033440.003798 0.003422 RMS 0.0077240.007959 0.007383 RMS 0.007569 0.00848 0.007841 Const 3353 ± 26.0 Underflow 0.035890.041140.04902 Const 799.5 ± 11.7 Underflow 0.52030.5701 0.5018 Underflow 0.79760.63360.7074 Underflow 0.89520.7398 0.8106 Underflow 1.1661.104 1.147 Underflow 0.8257 0.696 0.5433 0.1 0.1 0.1 0.1 0.1 0.1 Mean 295.5 ± 0.3 Overflow 0 Mean 315.2 ± 1.5 Overflow 00 Overflow 0 Overflow 00 Overflow 00 Overflow 00 1000 Integral 1 Integral 11 Integral 1 Integral 11 Integral 11 Integral 11

(pe) 22 2 22 22 22 54 Mn-54 @ 10 cm Sigma 32.5 ± 0.7 χ 2 / ndf 36.3337.2835.86 / 323129 Sigma 38.71 ± 2.47 χ / /ndf ndf 15.3715.99 8.228 / / 111010 χ / ndf 8.6256.6926.655 / 867 χ / /ndf ndf 9.563 1.291.811 / / 321 χ / /ndf ndf 6.8084.229 2.3 / / 576 χ / /ndf ndf 3.7987.593 9.435 / / 677 — Mn @ cell bottom 0.08 0.08 0.08 0.08 0.08 0.08 tot 3500 Const 0.052260.058760.05421 ± 0.001950.001900.00239 Const 0.2103 0.205 0.2408 ± ± 0.0094± 0.0090.0193 Const 0.28870.27990.2945 ± 0.01680.01330.0136 Const 1.0091.181 0.7558 ± ± 0.3090.7550.1066 Const 0.3466 0.388 0.3274 ± ± 0.0449± 0.0610.0326 Const 0.3387 0.3143 0.29 ± 0.0357± 0.0293 0.03 — 54MnMn-54 @ cell@ 45 middlecm Mean 0.069950.064180.07691 ± 0.001060.000980.00102 Mean 0.0055770.004731 0.0001415 ± ±0.0018280.001620 0.0030418 Mean 0.0025480.0034030.004666 ± 0.0017030.0014730.001181 Mean − 0.0035880.005673 − 0.0006048 ± ±0.0030260.006408 0.0015967 Mean − 0.0024880.005003 − 8.813e −05 ± ± 0.0033480.003812 2.521e−03 Mean − 0.0013440.001498 − 0.0005922 ± ±0.0028900.002766 0.0024182 2 / ndf 19.11 / 9 Sigma 0.035960.03215 0.0349 ± ± 0.001030.00102 0.0009 2 / ndf 7.151 / 9 Sigma 0.014760.01496 0.0163 ± 0.000970.00082 ± 0.0014 Sigma 0.009801 0.011650.01147 ± ± 0.000687 0.000840.00079 Sigma 0.0067420.007537 0.005765 ± 0.0009730.001928± 0.000616 Sigma 0.01416 0.0134 0.01222 ± ± 0.00140± 0.0013 0.00105 Sigma 0.01288 0.0148 0.01302 ± ± 0.00112± 0.0012 0.00095 χ 0.06 0.06χ 0.06 0.06 0.06 0.06 3000 54 — MnMn-54 @ cell@ 80 topcm Const 3961 ± 28.3 800 Const 1039 ± 14.5

Cell 3 Q Mean 302.1 0.3 Mean 321.2 0.7 ± 0.04 0.04 ± 0.04 0.04 0.04 0.04 STEREO Preliminary Sigma 28.21 0.46 Sigma 33.34 1.42 March 2017 2500 ± ± 6000.02 0.02 0.02 0.02 0.02 0.02 χ2 / ndf 20.7 / 9 χ2 / ndf 1.886 / 9

2000 Const 2852 ± 23.7 0 Const0 810.1 ± 12.7 0 0 0 0 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Mean 302.9 ± 0.5 Mean 321.3 ± 1.0 Sigma 35.79 1.03 Sigma 38.68 2.42 1500 ± 400 ± hLeaks_GCFronttoGCBack_Pos10hLeaks_GCFronttoGCBack_Pos45hLeaks_GCFronttoGCBack_Pos80 GC Front to Side hLeaks_GCFronttoGCD19_Pos10hLeaks_GCFronttoGCD19_Pos45hLeaks_GCFronttoGCD19_Pos80 hLeaks_GCFronttoGCIN20_Pos10hLeaks_GCFronttoGCIN20_Pos45hLeaks_GCFronttoGCIN20_Pos80 Entries 1284 15011645 EntrieshLeaks_GCBacktoTgCell1_Pos10hLeaks_GCBacktoTgCell1_Pos45hLeaks_GCBacktoTgCell1_Pos80 164512841501 hLeaks_GCBacktoTgCell2_Pos10hLeaks_GCBacktoTgCell2_Pos45hLeaks_GCBacktoTgCell2_Pos80Entries 164512841501 hLeaks_GCBacktoTgCell3_Pos10hLeaks_GCBacktoTgCell3_Pos45hLeaks_GCBacktoTgCell3_Pos80 hLeaks_GCBacktoTgCell4_Pos10hLeaks_GCBacktoTgCell4_Pos45hLeaks_GCBacktoTgCell4_Pos80 hLeaks_GCBacktoTgCell5_Pos10hLeaks_GCBacktoTgCell5_Pos45hLeaks_GCBacktoTgCell5_Pos80 hLeaks_GCBacktoTgCell6_Pos10hLeaks_GCBacktoTgCell6_Pos45hLeaks_GCBacktoTgCell6_Pos80 0.12 0.12 0.12 Mean 0.012890.01191 0.01178 MeanEntries 0.098540.09576 0.1014 309241 229 MeanEntries 0.23790.23750.2281 229309241 Entries 229 309241 Entries 229309241 Entries 229309241 Entries 229309241 RMS 0.0089030.008751 0.009049 0.12 RMS 0.03279 0.03390.0328 0.12 RMS 0.054970.048890.04861 0.12 0.12 0.12 0.12 Underflow 0.55540.6504 0.682 UnderflowMean 0.0013340.0024380.0031250.0050850.005311 0.004612 UnderflowMean 0.0048890.0046140.003914 0 Mean 0.005168 0.004824 0.0051 Mean 0.0025170.002469 0.00262 Mean 0.011670.013460.01361 Mean 0.09639 0.088410.0995 0.1 Overflow 0 0.1 0.1 1000 Overflow 0 OverflowRMS 0.003864 0.00419 0.003463 0 OverflowRMS 0.0036380.0032350.003019 0 RMS 0.004217 0.003675 0.00382 RMS 0.0021520.002105 0.00211 RMS 0.0077590.008269 0.00859 RMS 0.023640.024520.02319 Integral 1 Integral 1 Integral 1 χ22 // ndf ndf 16.265.092 3.864 // 88 χ 2Underflow / ndf 45.0743.4159.82 0.8828 1.006 / /1.16 424041 χUnderflow2 / ndf 91.0172.9552.32 0.86180.82580.8841 / 625857 Underflow 0.9244 0.9434 0.596 Underflow 0.77520.81760.7852 Underflow 0.18650.19770.1531 Underflow 0 0.08 Const 0.2541 0.224 0.2645 ± ± ±0.0162 0.0090.0191 2000.08 0.1 ConstOverflow 0.063650.06547 0.06411 ± 0.002090.002000.00234 00 0.080.1 ConstOverflow 0.036730.041220.04203 ± 0.001260.001330.00150 0 0.1 Overflow 00 0.1 Overflow 0 0.1 Overflow 0 0.1 Overflow 0 Mean 0.0067840.002423 0.0005608 ± ± 0.0011560.002029 0.0022801 Mean 0.097770.09477 0.1001 ± ± 0.000780.00089 0.0008 Mean 0.23560.2264 0.235 ± ± 0.00120.0013 0.001 STEREO Preliminary Integral 11 Integral 1 Integral 11 Integral 1 Integral 1 Integral 1 archSigma 0.012330.01367 0.01458 2017 ± ±0.000720.00094 0.00098 Sigma 0.030050.029530.03044 ±± 0.000570.000660.00061 Sigma 0.051220.046370.04569 ± 0.001140.000950.00102 M 22 2 22 2 2 500 0.06 0.06 χ / /ndf ndf 1.1731.339 0.1633 / / / 2 1 0.06 χ 2 / ndf 1.82e 0.03038 0.0565−10 // 101 χ / /ndf ndf 3.4510.05775 2.403 / // 2 11 χ / ndf 1.402e − 06 1 // 00 χ 2 / ndf 4.6494.157 2.24 / 64 χ 2 / ndf 11.8516.7227.32 / 212623 0.08 0.08 0.08 0.08 0.08 0.08 Const 0.5261 0.744 0.741 ± ± 0.0618 ± 0.393 0.300 Const 0.59390.6815 1.82 ± 0.06790.1276± 4.29 Const 1.089 0.70450.5938 ± ± 1.3400.29360.1152 Const 0.97010.9711 9.774 ± 0.76040.716966.925 Const 0.24240.22130.2223 ± 0.02130.01720.0194 Const 0.081150.082320.07602 ± 0.006990.006330.00767 0.04 0.04 Mean − 0.002206 −0.00261 0.0006321 ± ± ±0.006824 0.001900.0047901 0.04 Mean 0.001249− 0.013470.00212 ± ± 0.002520 0.026930.00165 Mean − 0.008251 − 0.0006902 0.0006631 ± ± ±0.013910 0.0055522 0.0027675 Mean 0.0032130.003336 − 0.01123 ± ± 0.0034490.002756 0.03292 Mean 0.008127 0.009420.01022 ± ± 0.001961 0.00141 Mean 0.099750.096870.09039 ± 0.001730.001370.00178

Sigma 0.0070940.005442 0.005897 ± 0.0025710.001065± 0.001880 Sigma 0.0050050.004835 0.01082 ± ± 0.0008410.001198 0.00665 Sigma 0.009448 0.006146 0.006181 ± 0.003785 ± 0.0022940.001204 Sigma 0.0019440.0019780.006261 ± 0.0018430.0016150.005607 Sigma 0.010230.010660.01101 ± 0.001420.001100.00113 Sigma 0.023820.022970.02386 ± 0.001430.001170.00202 50 100 150 200 2500.02 300 350 400 450 500 0.02500.06 100 150 200 250 300 3500.020.06 400 450 500 0.06 0.06 0.06 0.06

00 0.1 0.2 0.3Cell0.4 0.5Qtot0.6 (pe)0.7 0.8 0.9 1 00 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Cell00 0.1Qtot0.2 0.3(pe)0.4 0.5 0.6 0.7 0.8 0.9 1 Cell 2 Qtot (pe) 0.04 0.04 0.04 0.04 0.04 0.04

hLeaks_GCBacktoGCFront_Pos10hLeaks_GCBacktoGCFront_Pos45hLeaks_GCBacktoGCFront_Pos80 hLeaks_GCBacktoGCD19_Pos10hLeaks_GCBacktoGCD19_Pos45hLeaks_GCBacktoGCD19_Pos80 hLeaks_GCBacktoGCIN20_Pos10hLeaks_GCBacktoGCIN20_Pos45hLeaks_GCBacktoGCIN20_Pos80 Entries 309 241229 0.02 Entries 229309241 0.02 Entries 229309241 0.02 0.02 0.02 0.02 0.12 Mean 0.0036790.003612 0.003771 0.12 Mean 0.037420.03598 0.0398 0.12 Mean 0.045070.04077 0.0397 RMS Gc D190.0026450.002539 0.003545 2 Gc IN20RMS 0.014620.016860.01859 2 RMS 0.021440.017360.01661 Underflow 0.38570.3169 0.4049 χ / ndf 17.5 / 9 Underflow 0.0043860.004167 0.01311 χ / ndf 2.178 / 9 Underflow 0.0043860.009804 0 0.1 Overflow 0 0.1 0.1 STEREO: First Results Alessandro MinottiOverflow (CEA 0 - IRFU) 0 Overflow 43 0 0 Overflow 0 0 0 0 0 Integral 1 Const 29.31 ± 2.35 0 0.1 0.2 0.3 0.4 0.5 0.6 Integral0.7 0.8 0.9 1 1 Const0 0.1 0.2 0.3 43.30.4 ±0.5 2.90.6 0.7Integral0.8 0.9 1 1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 χ22 // ndf ndf 1.002e6.985e− 08 1 / / 01 χ 2 / ndf 10.8616.0810.68 / 141516 χ 2 / ndf 14.4611.34 27.2 / 1817 0.08 Const 0.7618 0.8035 3.65 ± ± ±0.1063 0.187418.20 Mean 222.5 ± 5.0 0.08 Const 0.13540.13720.1351 ± 0.01140.01040.0115 0.08 Mean 236.8 ± 4.0 Const 0.11850.12010.1284 ±± 0.01030.00900.0111 Mean 0.001426 0.0008061 −0.011 ± ± 0.001649 0.0024447 ± 0.035 100 Mean 0.039120.036570.03458 ± 0.001000.000850.00097 Mean 0.041740.039730.03833 ± 0.001140.000940.00102 Sigma 0.0040440.007656 0.004255 ± ±0.0007590.007294 0.001034 Sigma 30.94 ± 7.37 Sigma 0.014130.013840.01422 ± 0.000750.000690.00082 Sigma 36.29 ± 8.65 Sigma 0.015420.014880.01571 ± 0.000880.000750.00086 0.06 300 0.06 0.06 hLeaks_GCD19toTgCell1_Pos10hLeaks_GCD19toTgCell1_Pos45hLeaks_GCD19toTgCell1_Pos80 hLeaks_GCD19toTgCell2_Pos10hLeaks_GCD19toTgCell2_Pos45hLeaks_GCD19toTgCell2_Pos80 hLeaks_GCD19toTgCell3_Pos10hLeaks_GCD19toTgCell3_Pos45hLeaks_GCD19toTgCell3_Pos80 hLeaks_GCD19toTgCell4_Pos10hLeaks_GCD19toTgCell4_Pos45hLeaks_GCD19toTgCell4_Pos80 hLeaks_GCD19toTgCell5_Pos10hLeaks_GCD19toTgCell5_Pos45hLeaks_GCD19toTgCell5_Pos80 hLeaks_GCD19toTgCell6_Pos10hLeaks_GCD19toTgCell6_Pos45hLeaks_GCD19toTgCell6_Pos80 0.04 0.04 Entries 318300365 0.04 Entries 318300365 Entries 318300365 Entries 318300365 Entries 318300365 Entries 318300365 2 0.12 0.12 2 0.12 0.12 0.12 0.12 χ / ndf 4.924 / 9 Mean 0.021630.022780.02383 χ / ndf 13.57 / 9 Mean 0.018140.016130.01833 Mean 0.014240.014470.01426 Mean 0.0074680.0069470.006691 Mean 0.018780.018650.01647 Mean 0.030250.02722 0.0268 0.02 Const 35.09 ± 2.69 800.02 RMS 0.01679 0.01990.0191 0.02 Const 52.49 ± 3.43 RMS 0.011080.012830.01311 RMS 0.010290.010120.01143 RMS 0.0071250.0054890.006467 RMS 0.014530.012950.01169 RMS 0.022080.02036 0.0192 250 Underflow 0.25430.2471 0.25 Underflow 0.16730.1928 0.2 Underflow 0.14390.1774 0.145 Underflow 0.23260.21460.2046 Underflow 0.20480.20460.2091 Underflow 0.1439 0.229 0.2 0.1 0.1 0.1 0.1 0.1 0.1 00 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Mean 220.8 ± 4.2 00 0.1 0.2 0.3 0.4 0.5 0.6 0.7Overflow0.8 0.9 1 0 00Mean0.1 0.2 0.3 244.60.4 0.5± 2.20.6 0.7Overflow0.8 0.9 1 0 Overflow 0 Overflow 0 Overflow 0 Overflow 0 Sigma 36.86 10.37 Integral 1 Sigma 28.41 4.35 Integral 1 Integral 1 Integral 1 Integral 1 Integral 1 ± ± 2 2 2 2 2 χ 2 / ndf 10.2913.6927.98 / 131517 χ 2 / ndf 12.345.761 28.2 / /1110 9 χ 2 / ndf 8.00711.5611.66 / 89 χ / ndf 5.5266.17112.63 / 537 χ 2 / ndf 27.43 12.668.119 / /13 9 χ 2 / ndf 15.3918.1823.02 / 1614 0.08 Const 0.40160.18970.1651 0.04600.82000.0863 0.08 Const 0.16370.15940.1701 0.01270.01470.0123 0.08 Const 0.21530.2151 0.204 0.01580.0156 0.018 0.08 Const 0.56870.48660.4703 0.22150.12870.0341 0.08 Const 0.16590.17750.1521 0.01290.01300.0142 0.08 Const 0.09927 0.11180.1428 0.012110.01800.0542 200 hLeaks_GCD19toGCFront_Pos10hLeaks_GCD19toGCFront_Pos45hLeaks_GCD19toGCFront_Pos80 hLeaks_GCD19toGCBack_Pos10hLeaks_GCD19toGCBack_Pos45hLeaks_GCD19toGCBack_Pos80 ± hLeaks_GCD19toGCIN20_Pos10hLeaks_GCD19toGCIN20_Pos45hLeaks_GCD19toGCIN20_Pos80± ± ± ±± ± ±± Entries 318300365 Entries 318300365 60 Mean − 0.010690.062770.02397 ± 0.017150.109940.02413 EntriesMean 0.0089830.007163 0.01015 ± ± 0.0032180.003954 0.00186 318300365 Mean 0.0081610.009006 0.01031 ± ± 0.0025180.001491 0.00127 Mean − − 0.0010420.0037380.004441 ± 0.0066070.0053390.001116 Mean 0.0073080.006278 0.01217 ± ± 0.0040190.004100 0.00199 Mean − 4.39e0.002981 − 0.02154 −06 ± ± 0.0106361.31e 0.02581−02 0.12 Mean 0.014640.014420.01427 0.12 2 Mean 0.028640.027340.03071 0.12 2 Mean 0.011870.011610.01041 / ndf 5.309 / 9 Sigma 0.031750.03684 0.0485 ± ± 0.007150.00782 0.0287 / ndf 16.53 / 9 Sigma 0.016440.013480.01912 ± 0.002180.001640.00240 Sigma 0.010830.012580.01136 ± 0.000930.002170.00119 Sigma 0.0088380.005285 0.01034 ± ± 0.0023230.000820 0.00232 Sigma 0.017820.014180.01724 ± 0.002770.001450.00248 Sigma 0.035930.033460.04255 ±± 0.005380.006300.00911 RMS 0.021850.020550.01716 χ RMS 0.026340.031660.02921 χ RMS 0.014750.010090.01003 Underflow 0.17340.19520.1624 Const 29.54Underflow 2.27 0.11190.11520.1889 0.06 0.06Const 38.7 2.9 Underflow 0.69490.7381 0.59 0.06 0.06 0.06 0.06 0.1 Overflow 0 0.1 Overflow ± 0 0.1 ± Overflow 0 Integral 1 Integral 1 Integral 1 2 Mean 2 227 4.4 Mean 245.1 10.5 2 150 χ / ndf 25.4216.7520.58 / 161413 χ / ndf ± 38.6523.2739.26 / 222325 ± χ / ndf 10.1914.86 5.95 // 798 0.08 Const 0.42430.3213 0.595 ± ± 0.21300.0942 0.700 0.08 Const 0.10130.11270.1504 ± 0.00980.01330.0390 0.04 0.080.04 Const 0.28050.28190.2966 ±± 0.02620.06550.0261 0.04 0.04 0.04 0.04 Mean −0.004077 − − 0.02919 0.01245 ± ±± 0.008697 0.033470.01211 Sigma 29.28Mean − 0.003967 0.008894 0.005991 ± 5.73 ±± 0.0110540.0147650.008641 Sigma 34.53 ± 8.80 Mean 0.0044280.004602 − 0.00101 ± ± 0.0026440.002334 0.00838 Sigma 0.017740.014820.02531 ± 0.003740.003570.00887 Sigma 0.029350.030130.02941 ± 0.005480.006420.00580 40 Sigma 0.009497 0.013780.01004 ± ±± 0.001461 0.004130.00161 0.06 0.06 0.06 0.02 0.02 0.02 0.02 0.02 0.02 0.04 100 0.04 0.04

0.02 0.02 0 0.02 0 0 0 0 0 20 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 00 0.1 0.2 0.3500.4 0.5 0.6 0.7 0.8 0.9 1 00 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 00 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

hLeaks_GCIN20toGCFront_Pos10hLeaks_GCIN20toGCFront_Pos45hLeaks_GCIN20toGCFront_Pos80 hLeaks_GCIN20toGCBack_Pos10hLeaks_GCIN20toGCBack_Pos45hLeaks_GCIN20toGCBack_Pos80 hLeaks_GCIN20toGCD19_Pos10hLeaks_GCIN20toGCD19_Pos45hLeaks_GCIN20toGCD19_Pos80hLeaks_GCIN20toTgCell1_Pos10hLeaks_GCIN20toTgCell1_Pos45hLeaks_GCIN20toTgCell1_Pos80 hLeaks_GCIN20toTgCell2_Pos10hLeaks_GCIN20toTgCell2_Pos45hLeaks_GCIN20toTgCell2_Pos80 hLeaks_GCIN20toTgCell3_Pos10hLeaks_GCIN20toTgCell3_Pos45hLeaks_GCIN20toTgCell3_Pos80 hLeaks_GCIN20toTgCell4_Pos10hLeaks_GCIN20toTgCell4_Pos45hLeaks_GCIN20toTgCell4_Pos80 hLeaks_GCIN20toTgCell5_Pos10hLeaks_GCIN20toTgCell5_Pos45hLeaks_GCIN20toTgCell5_Pos80 hLeaks_GCIN20toTgCell6_Pos10hLeaks_GCIN20toTgCell6_Pos45hLeaks_GCIN20toTgCell6_Pos80 Entries 380515741 Entries 380515741 EntriesEntries 380515380515741 741 Entries 380515741 Entries 515741 380 Entries 380515741 Entries 380741 515 Entries 380515741 0.12 Mean 0.023460.02261 0.0246 0.12 Mean 0.025350.025520.02541 0.120.12 Mean 0.008451 0.01070.0097 0.12 0.12 0.12 0.12 0.12 RMS 0.025070.029930.02397 RMS 0.020230.023120.02277 RMSMean 0.0072060.0080540.005951 0.01072 0.011610.009968 Mean 0.0064160.0071620.006358 Mean 0.0078040.007665 0.006435 Mean 0.0035780.0044470.003785 Mean 0.0082850.007644 0.009184 Mean 0.012130.01028 0.0131 50 100Underflow 150 2000.008163 0.018770.02386 250 300 350 400 450Underflow 500 0.070420.066250.05556 50 100 150 200UnderflowRMS 250 300 0.01085 0.01340.4507 0.39810.010160.502 350 400 450 500 RMS 0.0054420.008627 0.01026 RMS 0.011580.01242 0.004879 RMS 0.0062340.0075980.005777 RMS 0.009790.01011 0.01298 RMS 0.015060.012820.01121 0.1 Overflow 0 0.1 Overflow 0 0.1 Overflow 0 Underflow 0.33330.3881 0.3046 Underflow 0.56530.5405 0.645 Underflow 0.4507 0.328 0.4902 Underflow 0.46150.54190.4195 Underflow 0.59660.3825 0.5015 Underflow 0.35880.26450.3571 Integral 1 Integral 1 Integral 1 2 2 0.1 2 0.1 0.1 0.1 0.1 0.1 χ / ndf 88.73152.7 87.9 / 272620 Cell Qtot (pe)χ / ndf 51.0256.9631.93 // 182421 χ Overflow / ndf 2.8583.7943.921 / 6500 Cell Qtot (pe) Overflow 0 Overflow 00 Overflow 0 Overflow 00 Overflow 0 0.08 Const 0.1783 0.214134.6 ± ± 0.0165 0.219219.6 0.08 Const 0.11250.13280.1131 ± 0.01070.01870.0060 0.08 Const 0.30480.3496 0.283 ± ±± 0.02370.0211 0.022 Integral 11 Integral 1 Integral 11 Integral 1 Integral 11 Integral 1 Mean − 0.011920.03998 −0.3595 ± ± 0.058170.00059 0.0913 Mean − 0.0038190.0086320.005574 ± 0.0084100.0096040.003414 Mean22 0.0078160.0032360.003287 ± 0.0010360.0021050.001423 2 22 2 22 2 Sigma 0.042070.09822 0.01001 ±± 0.018220.012790.00121 Sigma 0.027340.032610.02567 ±± 0.005190.004230.00221 Sigmaχ / /ndf 0.0084310.008731ndf 0.01041 ± ± 15.080.0007680.000785 9.9640.00115 20.24 / / /1012 8 χ / ndf 13.86 2.0887.685 / /11 38 χ / /ndf ndf 23.26 9.683 0.6115 / /13/ 83 χ / ndf 2.99516.6816.39 / 387 χ / /ndf ndf 17.44 12.84 19.9 / /10 119 χ / ndf 15.2617.5220.38 / 121411 0.06 0.06 0.06 0.08 Const 0.29330.3108 ±0.36 0.03350.0290 ± 0.05 0.08 Const 0.61180.73390.5163 ±± 0.18690.23020.0961 0.08 Const 0.45950.5856 0.4874 ± 0.04240.1319± 0.0804 0.08 Const 1.7542.5818.086 ± 2.1705.1813.202 0.08 Const 0.43690.6099 0.4576 ± 0.05040.1986± 0.1005 0.08 Const 0.33270.28660.3351 ±± 0.11150.02770.0261

Mean 0.0019570.002054 − 0.0009369 ± ±0.0040640.002366 0.0032980 Mean −0.0001570.0031520.003015 ± 0.0031610.0046610.003992 Mean − 0.0015370.001777 0.0001297 ± ±0.0017250.003631 0.0030422 Mean −0.0043370.006552 − 0.01238 ± ± 0.0080290.011977 0.00202 Mean − 0.001884 0.004711 − 0.0006317 ± ±0.0024800.005368 0.0046428 Mean − 0.0051720.0024510.001531 ± 0.0099830.0029680.002127 0.04 0.04 0.04 Sigma 0.011630.01067 0.01144 ±± 0.00229±0.00111 0.00150 Sigma 0.0077920.0088450.007665 ± 0.0013480.0017000.001403 Sigma 0.0069630.007744 0.008061 ± 0.0008640.001490 ± 0.001337 Sigma 0.0054730.0058770.006952 ± 0.0021590.0028370.000432 Sigma 0.0071210.009976 0.008672 ± 0.0013030.001909± 0.002085 Sigma 0.015360.011320.01024 ±± 0.003990.001550.00101 0.06 0.06 0.06 0.06 0.06 0.06 Figure 3: Method 2. On right Qtot cell 60.02 versus Qtot cell 2. The0.02 events without Qtot 0.02

0 0 00.04 0.04 0.04 0.04 0.04 0.04 in cell 6, whatever Qtot we have in cell 2, draw0 0.1 0.2 a0.3 horizontal0.4 0.5 0.6 0.7 0.8 0.9 1 line showing0 0.1 0.2 0.3 0.4 no0.5 0.6 correlation0.7 0.8 0.9 1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 00.10.20.30.40.50.60.70.80.910.010.020.030.040.050.060.070.080.090.110.120.130.140.150.160.170.180.190.210.220.230.240.250.260.270.280.290.310.320.330.340.350.360.370.380.390.410.420.430.440.450.460.470.480.490.510.520.530.540.550.560.570.580.590.610.620.630.640.650.660.670.680.690.710.720.730.740.750.760.770.780.790.810.820.830.840.850.860.870.880.890.910.920.930.940.950.960.970.980.99 between the amount of light in cell 2 andhLeaks_GCFronttoGCFront_Pos10Entries MeanRMSUnderflowOverflowIntegral2ConstSigmahLeaks_GCFronttoGCFront_Pos45hLeaks_GCFronttoGCFront_Pos80hLeaks_GCBacktoGCBack_Pos10hLeaks_GCBacktoGCBack_Pos45hLeaks_GCBacktoGCBack_Pos80hLeaks_GCD19toGCD19_Pos10hLeaks_GCD19toGCD19_Pos45hLeaks_GCD19toGCD19_Pos80hLeaks_GCIN20toGCIN20_Pos10hLeaks_GCIN20toGCIN20_Pos45hLeaks_GCIN20toGCIN20_Pos80 light1501 /0.01.421050.0111216451284229309241318300365380515741 collected ndfnannaninf in0 cell 6. On / the opposite, 199 0.02 0.02 0.02 0.02 0.02 0.02 χ± 0 0 0 0 0 0 graphs on the left show adjacent cell, Qtot of cell 3 versus Qtot of cell 2. We clearly 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 see a linear correlation between the minimum of light detected in cell 3 and the produce light in cell 2. Bottom graphs show the correlation and the linear fit which the slope give the light leaks.

5 Evolution of Light Leaks in Time

• Two methods compatible, Mn runs show clear z-dependence cell6 /Q

cell8 0.1 Q

0.08

0.06 Mn-54 in Cell 6 Bottom

Mn-54 in Cell 6 Middle

0.04 Mn-54 in Cell 6 Top

STEREOarch Preliminary 2017 0.02StatusM of leak simulations 17/11 01/12 15/12 29/12 12/01 26/01 09/02 23/02 Date

■ We still do not know what creates leaks (except the expected ones). • LL variation makes global analysis more complex - need■ Possibility a MC of liquidthat in thereproduces acrylic sandwich data? in terms of LL for energy reconstruction

• We developed an optical model with VM2000 soaked in scintillator (liquid in air gap) acrylic Air 1 Air 2 - Mean LL data values are well reproduced Air 3 acrylic - z-dependence under investigation

2 STEREO: First Results Alessandro Minotti (CEA - IRFU) 44 A Possible Approach: Reconstructed Energy

• The charge collected in a cell i is proportional to the light produced and remaining in the cell i, plus the contribution from other cells j due to LL

• Calibration coefficients Cj and LL values Lji for a given date can be injected in the equation → M(t)

• By inverting M(t), Edep(t) is calculated 54Mn in Cell6/45cm : Reconstructed Energy Evolution

0.022 November, 15th 54 - Stabilise detector response (e.g. Mn) 0.02 Simulation December, 14th 0.018 January, 13th 0.016 - Working to reach ~% level to compare 0.014 data-MC and calculate systematics 0.012 Density Probability [/bin width] STEREO Preliminary 0.01 March 2017 0.008

- Allows to work on topological cuts 0.006 0.004 - Under study - need to test on other 0.002 0 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2 events (diffused, different E) like n-H Reconstructed Energy (Full Detector) [MeV]

STEREO: First Results Alessandro Minotti (CEA - IRFU) 45 Signal: Preliminary IBD Selection

+ • Signal: IBD interactions νe + p → e + n

• IBD selection: prompt-delayed time coincidence + energy window for each of the 2 signals

dep dep - Prompt (e+ ionisation-annihilation) : 2 MeV < E det < 8 MeV , E GC < 1.1 MeV

dep dep - Delayed (Gd-n capture) : 5 MeV < E det < 10 MeV , E Tg > 1 MeV

- Coincidence time window: 0-70 μs

ν̄e

ΔT n p • Ee+ = 1-8 MeV Gd • En-Gd ∼ 8 MeV + e • clean between - Δt ∼ 18 μs prompt and delayed e t prompt delayed

STEREO: First Results Alessandro Minotti (CEA - IRFU) 46 Background: Accidentals Coincidences

• Accidental coincidences arise predominately from gammas and are

- Produced by the reactor

- Influenced by the activity of neighbouring experiments

• Remaining accidental component is measured online and accurately subtracted statistically using multiple off-time windows

• Prompt-delayed vertices strongly correlated for IBD → a simple cut on prompt-delayed spatial distance can reduces the accidentals (by a factor of ~2) without removing IBDs ) -1 600 Accidental rate

Accidental rate (prompt-delayed distance < 1.5 cells) Rate (day 500 Accidental 400 ΔT ' ON OFF ON (active IN20) (inactive IN20) 300

STEREO Preliminary 200 March 2017 t prompt virtual prompt Shift Gate 100

Event shifted NbShift times 0 15/12 22/12 29/12 05/01 12/01 19/01 26/01 02/02 09/02 16/02 Time

STEREO: First Results Alessandro Minotti (CEA - IRFU) 47 Background: Cosmogenic

• Cosmic muons induce multiple events that can mimic the IBD prompt-delayed coincidence

• To reduce contribution from cosmogenic correlated events, preliminary time isolation cuts are applied (under study)

- No event before prompt or after delayed in a 100 μs window

- 100 μs veto after a muon, identified via

̣ Active Muon Veto with preliminary software cut - 6.5% deadtime, Veto efficiency 98.95% ± 0.09% (evaluated with vertical muons)

̣ Released energy (>20 MeV) or saturation of PMT in the Detector

ΔT clean before clean after

clean between prompt and delayed t electronic gate prompt delayed

STEREO: First Results Alessandro Minotti (CEA - IRFU) 48 Background: Stopping Muons

• Decaying μ - Michel e coincidence can be mistaken as IBD

• μ have to stop at cell top to be in the prompt E window

- Inhomogeneous charge among cell PMTs max → discrimination using QPMT /Qcell μ e - Low energy μ have large dE/dx (Qtail/Qtot) → discrimination using PSD τμ = 2.2 μs

35 1 No specific cut 3000 - Prompt 0.9 30 Cut on asymetry of collection cell

/ Q 2500 0.8

MAX 25 Q

0.7 2000 20

0.6 1500 15 0.5 10 1000 0.4

5 500 0.3 Preliminary Cut 0.2 0 0 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 10 20 30 40 50 60 70 Interval between prompt and delayed event [us] QMAX / Qcell - Delayed

STEREO: First Results Alessandro Minotti (CEA - IRFU) 49 Background: Fast Neutrons

• Fast neutrons can produce recoil proton(s) before being captured

• Recoil protons have large dE/dx (Qtail/Qtot) → discrimination using PSD n - Difference for reactor on and off = electronic component Gd p - Correspondence between on and off = recoil protons p n → no strong indication of Fast Neutrons from reactor

22 ) )

tot 0.5 -1 -1 80 PSD ON ON-OFF / Q 20 tail 0.45 120 PSD OFF PSD Prompt ON Accidental Q 70

18 Rate (day Rate (day 0.4 STEREO Preliminary 60 March 2017 16 100 14 50 0.35 STEREO Preliminary 80 March 2017 12 40 0.3 60 10 30 0.25 8 40 20 0.2 6 10 4 20 0.15 0 2 0 0.1 0 1 2 3 4 5 6 7 8 9 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6 Q / Q Q / Q Event energy [MeV] tail tot tail tot

STEREO: First Results Alessandro Minotti (CEA - IRFU) 50 Correlated Rate Variation with Atmospheric Pressure

• Residual cosmic background is measured during reactor off

• Subtraction must take into account differences in μ rates between on and off periods

• Correction for μ rate variation with atmospheric pressure patm is made by extracting the Pressure correction to be applied dependency (on-off combined fit) and re-normalising to a pref

2 1400 χ / ndf 115.8 / 112 ON Prob 0.3839 1300 p0 6094 ± 616.9

Rate [/day] p1 −4.936 ± 0.6037 1200

1100

1000

900

800

700

600

500 1005 1010 1015 1020 1025 1030 1035 1040 1045 Pressure [hPa]

2 / ndf 115.8 / 112 950 STEREO Preliminary χ March 2017 Prob 0.3839 900 OFF p0 5750 ± 621.1 p1 4.936 0.6037

Rate [/day] − ± 850

800

750

700

650

600

550

500

450 1010 1015 1020 1025 1030 1035 1040 Pressure [hPa]

STEREO: First Results Alessandro Minotti (CEA - IRFU) 51 CEA Saclay (Stereo Analysis Meeting) Updates on pairs search and cell-by-cell energy dependent PSD 09 march 2017 11 / 11 Preliminary Rates and Perspectives

• Although having to face the challenge of working at ground level and in a reactor facility, we have a good control of our background

- Accidentals are estimated online and subtracted

- μ-induced background is rejected with Veto, isolation cuts (multi-n) charge deposit homogeneity (stopping muons), PSD (stopping muons, fast neutrons)

• PSD alone cannot deal with fast neutrons - work in progress on topological cuts to reach our nominal S/B = 1.5 (Edep in different cells, Gd multi-γ, Neural Networks)

Evolution of the correlated rate in the Stereo detector ) -1 1400

Rate (day 1200

1000

800

600

400 STEREO Preliminary March 2017 200

0 22/12 29/12 05/01 12/01 19/01 26/01 02/02 09/02 Time

STEREO: First Results Alessandro Minotti (CEA - IRFU) 52 To Wind Up

• STEREO has collected data from 1.5 reactor cycles (~70 days), and 25 days of reactor off, that we are currently analysing

• Monitoring tools of response have been developed to cope with detector evolution

• Mitigation of reactor-induced background was a challenge but accidentals coincidence are well controlled

• Now focusing on dealing with cosmic background, one of the major concerns for a ground level measurement, in order to reach our sensitivity goals

• Competition is tough, with many experiments working to resolve the RAA in the next few years and shed light in the complex matter of reactor flux estimation

• STEREO, NEOS, Neutrino-4 are opening an exciting season of the sterile neutrino search - RAA stands on the edge between New Physics and neutrino oscillation

STEREO: First Results Alessandro Minotti (CEA - IRFU) 53 …thank you for your attention