Search for K + → π + νν̄ at the NA 62 experiment

Looking for phenomena beyond the Standard Model (SM) in rare decays is a complementary approach to direct searches for New Physics (NP) at colliders. One of the theoretically cleanest processes is the ultra rare decay K → πνν̄. The goal of the NA62 experiment at CERN SPS is to measure the branching ratio (BR) of this decay with 10% precision. The experiment has been launched in 2014. In 2015, the detector was commissioned at a low intensity beam. The experimental setup is described and performances achieved in 2015 are discussed in view of the final measurement. 1 K → πνν̄ decays The ultra rare decays K → πνν̄ and KL → π νν̄ proceed via flavor changing neutral currents (FCNC) and are highly suppressed in the SM due to the GIM mechanism. The Feynman diagrams of box and penguin contributions are shown in Fig. 1. e-mail: Viacheslav.Duk@cern.ch G. Aglieri Rinella, R. Aliberti, F. Ambrosino, R. Ammendola, B. Angelucci, A. Antonelli, G. Anzivino, R. Arcidiacono, I. Azhinenko, S. Balev, M. Barbanera, J. Bendotti, A. Biagioni, L. Bician, C. Biino, A. Bizzeti, T. Blazek, A. Blik, B. BlochDevaux, V. Bolotov, V. Bonaiuto, M. Boretto, M. Bragadireanu, D. Britton, G. Britvich, M.B. Brunetti, D. Bryman, F. Bucci, F. Butin, E. Capitolo, C. Capoccia, T. Capussela, A. Cassese, A. Catinaccio, A. Cecchetti, A. Ceccucci, P. Cenci, V. Cerny, C. Cerri, B. Checcucci, O. Chikilev, S. Chiozzi, R. Ciaranfi, G. Collazuol, A. Conovaloff, P. Cooke, P. Cooper, G. Corradi, E. Cortina Gil, F. Costantini, F. Cotorobai, A. Cotta Ramusino, D. Coward, G. D’Agostini, J. Dainton, P. Dalpiaz, H. Danielsson, J. Degrange, N. De Simone, D. Di Filippo, L. Di Lella, S. Di Lorenzo, N. Dixon, N. Doble, B. Dobrich, V. Duk, V. Elsha, J. Engelfried, T. Enik, N. Estrada, V. Falaleev, R. Fantechi, V. Fascianelli, L. Federici, S. Fedotov, M. Fiorini, J. Fry, J. Fu, A. Fucci, L. Fulton, S. Gallorini, S. Galeotti, E. Gamberini, L. Gatignon, G. Georgiev, A. Gianoli, M. Giorgi, S. Giudici, L. Glonti, A. Goncalves Martins, F. Gonnella, E. Goudzovski, R. Guida, E. Gushchin, F. Hahn, B. Hallgren, H. Heath, F. Herman, T. Husek, O. Hutanu, D. Hutchcroft, L. Iacobuzio, E. Iacopini, E. Imbergamo, O. Jamet, P. Jarron, E. Jones, T. Jones K. Kampf, J. Kaplon, V. Kekelidze, S. Kholodenko, G. Khoriauli, A. Khotyantsev, A. Khudyakov, Yu. Kiryushin, A. Kleimenova, K. Kleinknecht, A. Kluge, M. Koval, V. Kozhuharov, M. Krivda, Z. Kucerova, Yu. Kudenko, J. Kunze, G. Lamanna, G. Latino, C. Lazzeroni, G. Lehmann-Miotto, R. Lenci, M. Lenti, E. Leonardi, P. Lichard, R. Lietava, L. Litov, R. Lollini, D. Lomidze, A. Lonardo, M. Lupi, N. Lurkin, K. McCormick, D. Madigozhin, G. Maire, C. Mandeiro, I. Mannelli, G. Mannocchi, A. Mapelli, F. Marchetto, R. Marchevski, S. Martellotti, P. Massarotti, K. Massri, P. Matak, E. Maurice, A. Mefodev, E. Menichetti, E. Minucci, M. Mirra, M. Misheva, N. Molokanova, J. Morant, M. Morel, M. Moulson, S. Movchan, D. Munday, M. Napolitano, I. Neri, F. Newson, A. Norton, M. Noy, G. Nuessle, T. Numao, V. Obraztsov, A. Ostankov, S. Padolski, R. Page, V. Palladino, G. Paoluzzi, C. Parkinson, E. Pedreschi, M. Pepe, F. Perez Gomez, M. Perrin-Terrin, L. Peruzzo, P. Petrov, F. Petrucci, R. Piandani, M. Piccini, D. Pietreanu, J. Pinzino, I. Polenkevich, L. Pontisso, Yu. Potrebenikov, D. Protopopescu, F. Raffaelli, M. Raggi, P. Riedler, A. Romano, P. Rubin, G. Ruggiero, V. Russo, V. Ryjov, A. Salamon, G. Salina, V. Samsonov, C. Santoni, G. Saracino, F. Sargeni, V. Semenov, A. Sergi, M. Serra, A. Shaikhiev, S. Shkarovskiy, I. Skillicorn, D. Soldi, A. Sotnikov, V. Sugonyaev, M. Sozzi, T. Spadaro, F. Spinella, R. Staley, A. Sturgess, P. Sutcliffe, N. Szilasi, D. Tagnani, S. Trilov, M. Valdata-Nappi, P. Valente, M. Vasile, T. Vassilieva, B. Velghe, M. Veltri, S. Venditti, P. Vicini, R. Volpe, M. Vormstein, H. Wahl, R. Wanke, P. Wertelaers, A. Winhart, R. Winston, B. Wrona, O. Yushchenko, M. Zamkovsky, A. Zinchenko. DOI: 10.1051/ , 06001 (2016) 130 EPJ Web of Conferences 13006001 MESON 2016 epjconf/2016 © The Authors, published by EDP Sciences. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0 (http://creativecommons.org/licenses/by/4.0/). Figure 1. Box and penguin diagrams contributing to K → πνν̄ decays. Within the SM, BR theoretical values contain CKM matrix elements as external inputs. These elements give the largest uncertainty to the BR. Using tree-level observables, one obtains [1, 2]: BR(K → πνν̄) = (8.4 ± 1.0) × 10 , BR(KL → π νν̄) = (3.4 ± 0.6) × 10 . These values are not polluted by contributions from the NP and hence are suitable for a NP search as a deviation from the SM BR value. More precise calculation with the loop-level measurements of the CKM elements assumes the absence of NP and provides the best accuracy for further recalculation of CKM elements from the comparison of theoretical and experimental BR values: BR(K → πνν̄) = (9.1 ± 0.7) × 10 , BR(KL → π νν̄) = (3.0 ± 0.3) × 10 . K → πνν̄ decays are very sensitive to NP. The largest contribution to the BR could come from models with new sources of flavor violation [3, 4]. The 10% precision measurement of the BR will allow to probe NP at mass scales up to 100 TeV. The only experimental measurement in the charged mode based on the observation of 7 events and decay-at-rest technique has been performed by the E787 and E949 experiments [5], while for the neutral mode there is only an upper limit on the BR from the E391 collaboration [6]: BR(K → πνν̄) = (17.3 −10.5 ) × 10 , BR(KL → π νν̄) < 2.6 × 10 . 2 NA62 experiment The NA62 experiment plans to collect 10 kaon decays in the fiducial volume which corresponds to ∼100 SM events of K → πνν̄ with ∼10% acceptance. The detailed description of the NA62 apparatus can be found in [7]. The schematic view of the setup is shown in Fig. 2. A proton beam from CERN SPS (400 GeV/c) impinging on a Beryllium target produces a secondary hadron beam of 75 GeV/c with ∼6% of K. Kaons are identified by the CEDAR detector (KTAG), a differential Cherenkov counter which combines high identification efficiency (at least DOI: 10.1051/ , 06001 (2016) 130 EPJ Web of Conferences 13006001 MESON 2016 epjconf/2016

Within the SM, BR theoretical values contain CKM matrix elements as external inputs.These elements give the largest uncertainty to the BR.Using tree-level observables, one obtains [1,2]: These values are not polluted by contributions from the NP and hence are suitable for a NP search as a deviation from the SM BR value.More precise calculation with the loop-level measurements of the CKM elements assumes the absence of NP and provides the best accuracy for further recalculation of CKM elements from the comparison of theoretical and experimental BR values: K → πνν decays are very sensitive to NP.The largest contribution to the BR could come from models with new sources of flavor violation [3,4].The 10% precision measurement of the BR will allow to probe NP at mass scales up to 100 TeV.
The only experimental measurement in the charged mode based on the observation of 7 events and decay-at-rest technique has been performed by the E787 and E949 experiments [5], while for the neutral mode there is only an upper limit on the BR from the E391 collaboration [6]:

NA6experiment
The NA62 experiment plans to collect 10 13 kaon decays in the fiducial volume which corresponds to ∼100 SM events of K + → π + νν with ∼10% acceptance.The detailed description of the NA62 apparatus can be found in [7].The schematic view of the setup is shown in Fig. 2. A proton beam from CERN SPS (400 GeV/c) impinging on a Beryllium target produces a secondary hadron beam of 75 GeV/c with ∼6% of K + .Kaons are identified by the CEDAR detector (KTAG), a differential Cherenkov counter which combines high identification efficiency (at least 95%) and excellent time resolution (∼100 ps), as needed for the time matching of a primary and a secondary track.The momentum of beam particles is measured by the beam tracker (Gigatracker, or GTK).Each of the three stations of GTK is made of 200 μm thick silicon sensors (300×300 μm size, 0.5 X 0 thickness per station, time resolution better than 200 ps, momentum resolution 0.2%).The detector is operated at ∼750 MHz rate.The momentum of secondary particles is measured by the Straw chamber spectrometer (STRAW).Each of the four STRAW stations comprises 1792 straw tubes arranged in four views (X, Y, U, V) and is operated in vacuum.The achieved momentum resolution is better than 1%.
The RICH detector aims at separating secondary pions from muons at the level of 1% (muon contamination in the pion sample) in the momentum range 15-35 GeV/c.A good time resolution (below 100 ps) is necessary both for time matching of two tracks (primary and secondary) and for providing a signal for the L0/L1 trigger.The charged particle hodoscope (CHOD) consisting of 128 scintillator slabs arranged in 2 planes (horizontal and vertical) measures the track crossing time with ∼200 ps resolution and also contributes to the L0 trigger.
The photon veto system covers the angular range up to 50 mrad and includes 4 detectors: LAV (12 stations operated in vacuum, made of lead glass blocks and covering the angular region from 8.5 to 50 mrad), LKr (liquid Krypton calorimeter from the NA48/2 experiment, vetoing the region from 1 to 8.5 mrad), IRC and SAC (small angle calorimeters covering the angular region <1 mrad).The rejection power of the veto system should be ∼10 8 for decays with π 0 in the final state.The CHANTI detector placed after the GTK3 identifies upstream inelastic interactions and muon halo.The iron/scintillator calorimeters (MUV1, MUV2) provide further pion/muon separation.Together with the RICH, this separation (muon mis-ID) should be ∼10 −7 .A fast muon veto MUV3 (iron wall followed by 5 cm thick scintillator, the light is detected by PMs) identifies muons and provides L0 trigger signals.

Analysis strategy
The signal detection principles are driven by the low BR of the decay (and hence a huge background), weak experimental signature (one primary and one secondary track) and the required background level (∼10%).This leads to the following strategy: • high beam intensity and fast timing; • decays-in-flight technique; • kinematic reconstruction and efficient particle identification (PID); • efficient veto system.
The main kinematic variable used in the analysis is the missing mass squared:  pion 4-momenta, p K and p π are their 3-momenta and θ πK is the angle between two particles in the laboratory frame.
Fig. 3 shows the distribution of this variable for the signal and backgrounds.Two signal regions are constrained by the main background from the decays K + → μ + ν μ (on the left), K + → πππ (on the right) and K + → π + π 0 (a peak from this decay splits the signal region into 2 halves).Kinematic cuts on the m 2 miss allow to reject ∼90% of background events.Further suppression is done by the veto system and PID.

Preliminary results of the 2015 run
The main goal of the 2015 run was to check the detector performances and the data quality.The data were collected at low beam intensity (∼10% of the nominal), the single-track selection criteria were close to those of the final measurement of the BR(K + → π + νν): • one primary track in GTK; • one secondary track in STRAW; • reconstructed vertex within the decay region; • energy deposit in LKr; • hits in the RICH and CHOD; • spatial matching (STRAW-CHOD, STRAW-LKr); • time matching (GTK-RICH-CHOD).
The m 2 miss distributions for the 2015 data are shown in Fig. 4. The first distribution is done with a kaon-like signal in the KTAG and demonstrates main backgrounds, while the second one illustrates the main sources of downstream tracks not related to the kaon decays: beam elastic scattering in the material along the beamline (KTAG, GTK) and inelastic scattering in the last GTK station.
Time resolutions of the KTAG and GTK are found to match the design values (100 and 200 ps respectively).The resolution of the m 2 miss is measured using the peak from the K + → π + π 0 decay and   found to be 1.2×10 −3 close to the design value 10 −3 .The resolution as a function of momentum is shown in Fig. 5.
The identification of secondary charged particles is done in the RICH detector.Fig. 6 illustrates the dependence of the Cherenkov ring radius on the momentum for the single-track selection.The pion-muon separation is based on the difference in ring radii for π and μ.It was measured by selecting a pure muon sample from K + → μ + ν μ decay and a pion sample from K + → π + π 0 decay.In the momentum range between 15 and 35 GeV/c the required muon contamination of 1% was achieved at the 80% pion ID efficiency.
The photon veto system is designed to suppress decays with photons in the final state.For photons from π 0 decays the rejection power provided by LAV, LKr, IRC and SAC detectors should be at least  10 8 , or, in other words, the veto inefficiency for photons with energy above 10 GeV should not exceed 10 −5 .The π 0 veto inefficiency measured from the K + → π + π 0 decay is shown in Fig. 7.The amount of statistics recorded in 2015 limits this measurement to 10 −6 (90% CL).

Figure 1 .
Figure 1.Box and penguin diagrams contributing to K → πνν decays.

Figure 3 .
Figure 3. m 2 miss = (P K − P π ) 2 for the signal and the most frequent background decay modes.The signal is multiplied by 10 10 , background modes are normalized according to their BR.

Figure 4 .
Figure 4. m 2 miss = (P K − P π ) 2 vs momentum after single-track event selection for kaon decays (left) and without a kaon tag (right).

Figure 5 .
Figure 5. Resolution of the m 2 miss vs momentum.Empty squares correspond to the values obtained with the nominal kaon momentum, black points -with the kaon momentum measured by GTK.