Double Chooz: Searching for theta13 with reactor neutrinos
aa r X i v : . [ h e p - e x ] M a y DOUBLE CHOOZ: SEARCHING FOR θ WITH REACTOR NEUTRINOS
P. NOVELLA
CIEMAT, Av. Complutense 22,Madrid 28040, Spain
The discovery of neutrino oscillations is a direct indication of physics beyond the StandardModel. The so-called atmospheric and solar sectors have been explored by several experi-ments, meanwhile the mixing angle θ connecting both sectors remains unknown. In contrastto accelerator experiments, reactor neutrinos arise as a clean probe to search for this angle.The Double Chooz experiment is meant to search for θ taking advantage of the neutrinosgenerated at the nuclear power plant of Chooz. Double Chooz relies on neutrino flux measure-ments at two different locations, the so-called far an near detectors, although the first phaseruns only with the far detector. The relative comparison of the fluxes at both sites will reducethe systematic uncertainties down to 0.6%. The commissioning of the far detector took placebetween January 2011 and March 2011, when physics data taking started. First results areexpected by the summer 2011. These results will improve the current limit to θ in case theoscillation signal is not observed. The final sensitivity to sin (2 θ ) is exepcted to be 0.03 at90% C.L. after 5 years of data taking. Neutrino oscillation data can be described within a three neutrino mixing scheme, in whichthe flavor states ν α ( e, ν, µ ) are connected to the mass states ν i ( i =1,2,3) through the PMNSmixing matrix U P MNS
1. This matrix can be expressed as the product of three matrices wherethe mixing parameters remain decoupled: U P MNS = U atm · U inter · U sol . The terms U atm and U sol describe the mixing in the so-called atmospheric and solar sectors, which are driven by themixing angles θ and θ , respectively. The U inter matrix stands for the interference sector whichconnects the previous two, according to the mixing angle θ and the phase δ responsible for the CP violation in the leptonic sector. Finally, the oscillation probability between two neutrinospecies becomes a function of the above oscillation parameters and the two independent masssquared differences ∆ m ij = m i − m j .The KamLAND experiment 2 has explored the oscillation in the solar sector and providedallowed and best fit values for θ and ∆ m , showing consistency with solar experiments data.In the same way, the MINOS experiment 3 has published results for atmospheric sector ( θ and | ∆ m | ), being consistent with atmospheric neutrino data. However, the subdominantoscillation corresponding to the interference sector has not been observed yet. Results fromCHOOZ experiment 4 show at 90% C.L. that sin (2 θ ) < .
15 for | ∆ m | = 2 . × − eV .Provided that δ appears in U inter only in combination with sin (2 θ ), the CP-violating phasealso remains unknown. As a direct consequence, the search for the third mixing angle stands asone of the major open issues in neutrino oscillation physics. Reactor neutrinos in the quest for θ Nuclear reactors produce nearly pure ¯ ν e fluxes coming from β decay of fission fragments. Atypical core delivers about 2 × ¯ ν e per second and GW th of thermal power. Such high isotropicfluxes compensate for the small neutrino cross-section and allow for an arbitrary location ofneutrino detectors, scaling the flux with 1 /L where L is the distance between the core and thedetector. Any oscillation effect in the ¯ ν e survival is governed by the following equation: P (¯ ν e → ¯ ν e ) ∼ = 1 − sin θ sin ( ∆ m L E ν ) − cos θ sin θ sin ( ∆ m L E ν ) (1)where E ν is the neutrino energy. The second and third terms of Eq. 1 describe the oscillationdriven by θ and θ (solar regime), respectively. The value of θ can be derived directlyby measuring P (¯ ν e → ¯ ν e ). Notice that in contrast to accelerator neutrino experiments, thismeasurement does not suffer from the δ − θ degeneracy. The most common way of detecting reactor neutrinos is via the inverse beta decay (IBD) ¯ ν e + p → n + e + . When this reaction takes place in liquid scintillator doped with Gadolinium, itproduces two signals separated by about ∼ µ s: the first one due to the e + and its annihilation(prompt signal), and the second one due to the n capture in a Gd nucleus (delayed signal). Thischaracteristic signature yields a very efficient background rejection. The e + energy spectrumpeaks at ∼ E ν . The mean energy of the ¯ ν e spectrum in a detectorfilled by such a scintillator is around 4 MeV, as shown in left panel of Fig. 1. According to Eq. 1,for this energy the oscillation effect due to θ starts to show up at L ∼ . θ is still negligible. Therefore, neutrino reactor experiments with short baselines become aclean laboratory to search for θ . Energy (MeV) a r b i t r a r y un i t s Neutrino visible energyReactor neutrino fluxIBD cross−sectionNeutrino visible energyReactor neutrino fluxIBD cross−sectionNeutrino visible energyReactor neutrino fluxIBD cross−section
L (m) ) e ν → e ν P ( N ea r D e t ec t o r F a r D e t ec t o r )=0.15 θ (2 sin eV −5 =8 10 m ∆ eV −3 =2.5 10 m ∆ )=0.87 θ (2 sin Figure 1: Left: ¯ ν e visible spectrum as a result of the flux shape and IBD cross-section. Right: ¯ ν e survivalprobability for E ν = 3 MeV, as a function of the distance L . In spite of its characteristic signature, the IBD signal can be mimicked by the so-calledaccidental and correlated backgrounds. The accidental background is defined as the coincidenceof a positron-like signal coming from natural radioactivity, and the capture in the detector of aneutron created by cosmic muon spallation in the surrounding rock. The correlated backgroundconsists of events which may mimic both the prompt and the delayed signals of the IBD. Fastneutrons and cosmogenic isotopes, both generated in muon interactions, are the main sourcesof this background. Fast neutrons are produced by muons in the surrounding rock and enterhe detector leading to proton recoils, thus faking a prompt signal, before being captured by aGd nucleus. Muons also produce inside the detector long-lived β -n decay isotopes, like Li and He. As the half-life of such cosmogenic isotopes is ∼
100 ms, their decay cannot be related tothe muon interaction.
The sensitivity to the θ -driven oscillation is optimized by detecting a deficit in the expectedneutrino events around 1 km away from the nuclear power plant, as shown in right panel of Fig. 1.However, some of the largest systematics in reactor experiments arise from the uncertaintiesin the original ¯ ν e fluxes. In order to reduce them, a relative comparison between two or moreidentical detectors located at different distances from the reactors becomes critical. In particular,a detector placed a few hundred meters away can measure the fluxes before any oscillation takesplace, as demonstrated in right panel of Fig. 1. The comparison between the so-called farand near detectors leads to a breakthrough in the sensitivity to θ , as all the fully correlatedsystematics cancel out. Further steps in the sensitivity optimization relay on reducing therelative normalization and the relative energy scale uncertainties of the detectors, as well as onminimizing the backgrounds. The Double Chooz experiment 5, located at the nuclear power plant of Chooz (France), aims atimproving the CHOOZ experience by means of a long-term stability multi-detector setup. Thecomparison between un-oscillated reactor neutrino flux at a near site and the oscillated flux at afar site allows for the cancellation of the reactor-related correlated errors. The detector-relatedsystematics are kept under control by constructing two identical detectors providing accurateenergy reconstruction and high signal-to-noise ratios. The Double Chooz collaboration involvesinstitutes from Brazil, France, Germany, Japan, Russia, Spain, UK and USA.The Chooz nuclear plant consists of two cores yielding a total thermal power of 8.54 GW th .The Double Chooz far detector is placed 1050 m away from the cores, in the same undergroundlaboratory used by the CHOOZ experiment. The laboratory is located close to the maximaloscillation distance and provides enough shielding (300 m.w.e.) against cosmic rays. A secondidentical detector (near detector) will be installed 400 m away from the reactor cores, in a newlaboratory (115 m.w.e) whose construction started in April 2011. The Double Chooz detectors design is optimized to reduce backgrounds. The detectors, shownin Fig. 2, consist of a set of concentric cylinders and an outer plastic scintillator muon veto ( outerveto ) on the top. The innermost volume ( target ) contains about 10 tons of Gd-loaded (0.1%)liquid scintillator inside a transparent acrylic vessel, where the neutrinos interact via the IBDprocess. This volume is surrounded by another acrylic vessel filled with unloaded scintillator( gamma-catcher ). This second volume is meant to fully contain the energy deposition of gammarays from the neutron capture on Gd and the positron annihilation in the target region. Thegamma-catcher is in turn contained within a third volume ( buffer tank ) made of stainless steeland filled with mineral oil. As the wall and the lids of the buffer are covered with an array of 39010” photomultiplier tubes (PMTs), meant to detect the scintillation light (13% photocathodecoverage), the oil shields the target and the gamma-catcher against the radioactivity of the PMTcomponents. The target, gamma-catcher and buffer tank define the inner detector . Finally, theouter volume containing the inner detector is a stainless steel vessel covered with 78 8” PMTsnd filled with scintillator. This volume plays the role of the inner muon veto . To protect theDouble Chooz detector from the radioactivity of the surrounding rock, a 15 cm layer of iron isused.
Figure 2: The Double Chooz detector design.
The detector performance is analyzed by means of a redundant set of calibration systems.Apart from the natural calibration sources (neutron captures in H, Gd and C), radioactivesources can be introduced in the different volumes of the detector, via a glove box. The goalis to achieve a relative error on the neutrino detection efficiency of 0.5% with both detectors,and an energy scale uncertainty of 0.5%. In addition, a set of LEDs embedded in the PMTsstructure is used to measure the PMTs gains and timing, as well as to monitor the stability ofthe detector.
The Double Chooz experiment aims at improving the CHOOZ result by means of an increase ofthe exposure and a reduction of the systematics. In order to reduce the statistical error downto 0.5% (was 2.8% in CHOOZ), a long-term stability scintillator has been developed, which willallow for a total data taking time of 5 years. Besides, a larger target volume (10.3 m m ) isused. The relative comparison of the fluxes between the far and the near detector will allow forthe reduction of the systematics error down to 0.6% (was 2.7% in CHOOZ), as shown in Tab.1. Backgrounds are also expected to be reduced with respect to CHOOZ due to the selectionof radiopure materials used in the detector, the two independent muon vetoes, and the buffervolume which isolates the PMTs from the active part of the detector. Table 1: Main systematic uncertainties in CHOOZ and Double Chooz reactor experiments.
CHOOZ Double Chooz
Reactor fuel cross section 1.9% –Reactor power 0.7% –Energy per fission 0.6% –Number of protons 0.8% 0.2%Detection efficiency 1.5% 0.5%
TOTAL
The newborn detector
The integration phase of the far detector of Double Chooz started in May 2008 with the inte-gration of the external shield, followed by the assembly of the inner veto tank. The buffer vesselwas completed in summer 2009, and the 330 10” PMTs were successfully mounted by fall 2009.Finally, the acrylic gamma-catcher and target vessels were installed inside the buffer, as shownin Fig. 3, and the detector was closed. First signals from the PMTs were observed in summer2010 as the DAQ and electronics systems became ready. The filling of the detector started inOctober 2010 and was completed by the end of 2010.
Figure 3: View of the acrylic vessels and the PMTs covering the the buffer tank walls.
The commissioning period took place between January 2011 and March 2011. First analysisof the detector response was carried out, assuring the good performance of both the innerdetector and the inner veto. First events in the filled detector were observed in January 2011.As an example, the display of a muon crossing both detectors is shown in Fig. 4.
Figure 4: Display of a crossing muon event in both inner detector (left) and inner veto (right). Colors show thecharge (digital units) collected at each PMT.
Getting the most from Double Chooz
The Double Chooz experiment is developed in two phases. Phase I started in March 2011once the commissioning of the far detector was completed. Even operating only one detector,this phase will be able to improve the current θ limit in a few months of data taking. Asensitivity of sin (2 θ ) ∼ . (2 θ ) of 0.03 (90% C.L.) will be achieved. A 3 σ measurement will be feasibleif sin (2 θ ) > .
05. A summary of the sensitivity of both phases is shown in Fig. 5.
Figure 5: Double Chooz expected sensitivity limit (90% C.L.) to sin (2 θ ) as a function of time for ∆m = 2.5 × − eV . Near detector is assumed to be ready 1.5 years after the start of the far detector operation. References
1. M. C. Gonzalez-Garcia and M. Maltoni, Phys. Rept. (2008)2. S. Abe et al. [KamLAND Collaboration], Phys. Rev. Lett. (2008)3. P. Adamson et al. [MINOS Collaboration], Phys. Rev. Lett. (2008)4. M. Apollonio et al. [CHOOZ Collaboration], Eur. Phys. J. C (2003)5. F. Ardellier et al.et al.