Improved Search for a Light Sterile Neutrino with the Full Configuration
of the Daya Bay Experiment
F. P. An,1A. B. Balantekin,2H. R. Band,3 M. Bishai,4S. Blyth,5,6D. Cao,7G. F. Cao,8J. Cao,8W. R. Cen,8Y. L. Chan,9 J. F. Chang,8L. C. Chang,10Y. Chang,6H. S. Chen,8Q. Y. Chen,11S. M. Chen,12Y. X. Chen,13Y. Chen,14J.-H. Cheng,10 J. Cheng,11Y. P. Cheng,8 Z. K. Cheng,15J. J. Cherwinka,2M. C. Chu,9A. Chukanov,16J. P. Cummings,17J. de Arcos,18 Z. Y. Deng,8X. F. Ding,8Y. Y. Ding,8M. V. Diwan,4M. Dolgareva,16J. Dove,19D. A. Dwyer,20W. R. Edwards,20R. Gill,4 M. Gonchar,16G. H. Gong,12H. Gong,12M. Grassi,8W. Q. Gu,21M. Y. Guan,8L. Guo,12R. P. Guo,8X. H. Guo,22Z. Guo,12 R. W. Hackenburg,4R. Han,13S. Hans,4,*M. He,8K. M. Heeger,3Y. K. Heng,8A. Higuera,23Y. K. Hor,24Y. B. Hsiung,5 B. Z. Hu,5T. Hu,8W. Hu,8E. C. Huang,19H. X. Huang,25X. T. Huang,11P. Huber,24W. Huo,26G. Hussain,12D. E. Jaffe,4 P. Jaffke,24K. L. Jen,10S. Jetter,8X. P. Ji,27,12X. L. Ji,8J. B. Jiao,11R. A. Johnson,28J. Joshi,4 L. Kang,29S. H. Kettell,4 S. Kohn,30M. Kramer,20,30K. K. Kwan,9M. W. Kwok,9T. Kwok,31T. J. Langford,3K. Lau,23L. Lebanowski,12J. Lee,20 J. H. C. Lee,31R. T. Lei,29R. Leitner,32J. K. C. Leung,31C. Li,11D. J. Li,26F. Li,8G. S. Li,21Q. J. Li,8S. Li,29S. C. Li,31,24 W. D. Li,8X. N. Li,8Y. F. Li,8Z. B. Li,15H. Liang,26C. J. Lin,20G. L. Lin,10S. Lin,29S. K. Lin,23Y.-C. Lin,5J. J. Ling,15 J. M. Link,24L. Littenberg,4B. R. Littlejohn,18D. W. Liu,23J. L. Liu,21J. C. Liu,8C. W. Loh,7C. Lu,33H. Q. Lu,8J. S. Lu,8
K. B. Luk,30,20 Z. Lv,34Q. M. Ma,8X. Y. Ma,8 X. B. Ma,13Y. Q. Ma,8 Y. Malyshkin,35D. A. Martinez Caicedo,18 K. T. McDonald,33 R. D. McKeown,36,37 I. Mitchell,23M. Mooney,4 Y. Nakajima,20J. Napolitano,38D. Naumov,16 E. Naumova,16H. Y. Ngai,31Z. Ning,8J. P. Ochoa-Ricoux,35A. Olshevskiy,16H.-R. Pan,5J. Park,24S. Patton,20V. Pec,32
J. C. Peng,19L. Pinsky,23C. S. J. Pun,31 F. Z. Qi,8M. Qi,7 X. Qian,4 N. Raper,39J. Ren,25R. Rosero,4 B. Roskovec,32 X. C. Ruan,25H. Steiner,30,20G. X. Sun,8 J. L. Sun,40W. Tang,4 D. Taychenachev,16K. Treskov,16K. V. Tsang,20 C. E. Tull,20N. Viaux,35B. Viren,4 V. Vorobel,32C. H. Wang,6M. Wang,11N. Y. Wang,22R. G. Wang,8 W. Wang,37,15
X. Wang,41Y. F. Wang,8 Z. Wang,12Z. Wang,8Z. M. Wang,8 H. Y. Wei,12L. J. Wen,8 K. Whisnant,42C. G. White,18 L. Whitehead,23T. Wise,2H. L. H. Wong,30,20S. C. F. Wong,15E. Worcester,4C.-H. Wu,10Q. Wu,11W. J. Wu,8D. M. Xia,43 J. K. Xia,8Z. Z. Xing,8J. Y. Xu,9J. L. Xu,8Y. Xu,15T. Xue,12C. G. Yang,8H. Yang,7L. Yang,29M. S. Yang,8M. T. Yang,11
M. Ye,8 Z. Ye,23M. Yeh,4 B. L. Young,42 Z. Y. Yu,8 S. Zeng,8 L. Zhan,8 C. Zhang,4 H. H. Zhang,15J. W. Zhang,8 Q. M. Zhang,34X. T. Zhang,8Y. M. Zhang,12Y. X. Zhang,40Y. M. Zhang,15Z. J. Zhang,29Z. Y. Zhang,8 Z. P. Zhang,26
J. Zhao,8 Q. W. Zhao,8Y. B. Zhao,8 W. L. Zhong,8 L. Zhou,8N. Zhou,26H. L. Zhuang,8 and J. H. Zou8
(Daya Bay Collaboration)
1
Institute of Modern Physics, East China University of Science and Technology, Shanghai 2University of Wisconsin, Madison, Wisconsin USA
3
Department of Physics, Yale University, New Haven, Connecticut USA 4Brookhaven National Laboratory, Upton, New York USA 5
Department of Physics, National Taiwan University, Taipei 6National United University, Miao-Li
7
Nanjing University, Nanjing 8Institute of High Energy Physics, Beijing 9
Chinese University of Hong Kong, Hong Kong 10Institute of Physics, National Chiao-Tung University, Hsinchu
11
Shandong University, Jinan
12Department of Engineering Physics, Tsinghua University, Beijing 13
North China Electric Power University, Beijing 14Shenzhen University, Shenzhen 15
Sun Yat-Sen (Zhongshan) University, Guangzhou 16Joint Institute for Nuclear Research, Dubna, Moscow Region
17
Siena College, Loudonville, New York USA
18Department of Physics, Illinois Institute of Technology, Chicago, Illinois USA 19
Department of Physics, University of Illinois at Urbana-Champaign, Urbana, Illinois USA 20Lawrence Berkeley National Laboratory, Berkeley, California USA
21
Department of Physics and Astronomy, Shanghai Jiao Tong University, Shanghai Laboratory for Particle Physics and Cosmology, Shanghai
22
Beijing Normal University, Beijing
24Center for Neutrino Physics, Virginia Tech, Blacksburg, Virginia USA 25
China Institute of Atomic Energy, Beijing 26University of Science and Technology of China, Hefei
27
School of Physics, Nankai University, Tianjin
28Department of Physics, University of Cincinnati, Cincinnati, Ohio USA 29
Dongguan University of Technology, Dongguan
30Department of Physics, University of California, Berkeley, California USA 31
Department of Physics, The University of Hong Kong, Pokfulam, Hong Kong 32Charles University, Faculty of Mathematics and Physics, Prague, Czech Republic
33
Joseph Henry Laboratories, Princeton University, Princeton, New Jersey USA 34Xi’an Jiaotong University, Xi’an
35
Instituto de Física, Pontificia Universidad Católica de Chile, Santiago, Chile 36California Institute of Technology, Pasadena, California USA
37
College of William and Mary, Williamsburg, Virginia USA
38Department of Physics, College of Science and Technology, Temple University, Philadelphia, Pennsylvania USA 39
Department of Physics, Applied Physics, and Astronomy, Rensselaer Polytechnic Institute, Troy, New York USA 40China General Nuclear Power Group, Shenzhen
41
College of Electronic Science and Engineering, National University of Defense Technology, Changsha 42Iowa State University, Ames, Iowa USA
43
Chongqing University, Chongqing (Received 6 July 2016; published 7 October 2016)
This Letter reports an improved search for light sterile neutrino mixing in the electron antineutrino disappearance channel with the full configuration of the Daya Bay Reactor Neutrino Experiment. With an additional 404 days of data collected in eight antineutrino detectors, this search benefits from 3.6 times the statistics available to the previous publication, as well as from improvements in energy calibration and background reduction. A relative comparison of the rate and energy spectrum of reactor antineutrinos in the three experimental halls yields no evidence of sterile neutrino mixing in the2 × 10−4≲ jΔm241j ≲ 0.3 eV2 mass range. The resulting limits on sin22θ14are improved by approx imately a factor of 2 over previous results and constitute the most stringent constraints to date in thejΔm241j ≲ 0.2 eV2 region.
DOI:10.1103/PhysRevLett.117.151802
The three-neutrino mixing framework, in which the flavor eigenstates (νe,νμ,ντ) mix with the mass eigenstates (ν1, ν2, ν3) via the Pontecorvo-Maki-Nakagawa-Sakata matrix [1–3], has been extremely successful in explaining the results observed in most solar, atmospheric, reactor, and long-baseline accelerator neutrino oscillation experiments
[4]. Despite this success, the exact mechanism by which neutrinos acquire their mass remains unknown, and the possible existence of additional neutrinos is under active consideration. To be consistent with precision electroweak measurements [5], these additional neutrinos are called “sterile” [2], that is, noninteracting within the standard model and thus with no known mechanism for direct detection. Nonetheless, an unambiguous signal of their existence can be sought in neutrino oscillation experiments, where they could affect the way in which the three active neutrinos oscillate if they mix with the latter.
In the simplest extension of the standard model, where only one sterile neutrino is considered in addition to the three active ones, the mixing can be expressed as
να¼
X4 i¼1
Uαiνi; ð1Þ
where U is a unitary 4 × 4 mixing matrix and Uαi is the neutrino mixing matrix element for the flavor eigenstateναand the mass eigenstateνi. The survival probability for a relativistic ναwith an energyE and a traveling distance L is given by
Pνα→να ¼ 1 − 4 X3 i¼1 X4 j>i jUαij2jUαjj2sin2Δji; ð2Þ
where Δji ¼ 1.267Δm2jiðeV2Þ½LðmÞ=EðMeVÞ and Δm2
ji¼ m2j− m2i is the mass-squared difference between
the mass eigenstatesνj andνi. As indicated in Ref.[6], in the case of electron antineutrino disappearance the neutrino mixing matrix elementsUeican be parametrized in terms of theθ14,θ13, andθ12mixing angles. Compared with standard three-neutrino mixing, the neutrino oscillation probability includes three additional oscillation frequencies associated withΔm241; Δm242, andΔm243. WhenjΔm241j ≫ jΔm231j these three parameters are virtually indistinguishable, and for the Daya Bay baselines Eq.(2)approximates to
P¯νe→¯νe≈ 1 − 4ð1 − jUe4jÞ2jUe4j2sin2Δ41 − 4ð1 − jUe3j2− jUe4j2ÞjUe3j2sin2Δ31
≈ 1 − sin22θ
Thus, to first order, evidence for light sterile neutrino mixing consists of an additional spectral distortion with a frequency different from standard three-neutrino oscillations.
No conclusive evidence for the existence of sterile neutrinos has been obtained. A few anomalies in short baseline neutrino oscillation experiments [7–13] can be explained with additional sterile neutrinos, but these results are in tension with the limits obtained from other experi-ments[14–17]. The majority of experimental searches have centered on mass-squared differences around 1 eV2 and higher, whereas the Daya Bay and other medium baseline reactor antineutrino experiments can make unique contri-butions in the sub-eV scale[6,18–25]. In 2014, the Daya Bay Collaboration reported on a search for light sterile neutrino mixing based on the first 217 days of data acquired with a partial configuration of six functionally identical antineutrino detectors (ADs) deployed at three experimen-tal halls (EHs), the results of which excluded a large, previously unexplored region of parameter space in the3 × 10−4≲ jΔm2
41j ≲ 0.1 eV2 range [26]. In this partial
con-figuration, three ADs were installed in two near halls (two in EH1 and one in EH2) and another three in a far hall (EH3). This Letter reports on an improved search made with the full eight-detector configuration shown in Fig. 1
that resulted from the installation of two additional ADs, one in EH2 and another in EH3, in the summer of 2012. The additional 404 days of eight-detector data collected from October 2012 to November 2013 amount to a 3.6 times increase in statistics.
Each AD is a three-zone cylindrical detector composed of two nested acrylic vessels within a concentric stainless steel vessel. The central vessel is filled with 20 tons of gadolinium-doped liquid scintillator (Gd-LS) that serves as
the primary target for antineutrino detection. A 22-ton pure LS volume encloses the central target and enables the detection ofγ rays that escape from the Gd LS volume. The outermost cylinder contains 40 tons of mineral oil that provide shielding againstγ-ray radiation from the detector components. A total of 192 photomultiplier tubes are installed on the vertical surfaces, and the top and bottom surfaces are covered with optical reflectors. Three auto-mated calibration units[27]that store and deploy calibra-tion sources and Light Emitting Diodes sit on top of the stainless steel vessel. The ADs are housed inside a muon veto system consisting of two optically separated inner and outer water pools[28]that provide shielding from ambient radiation and serve as active water Cherenkov muon detectors. Four layers of resistive plate chambers are installed on top of each water pool. More information on the Daya Bay detectors and their performance can be found in Refs.[29,30].
Reactor antineutrinos are detected via the inverse beta decay (IBD) reaction ¯νeþ p → eþþ n. The positron deposits its energy in the scintillator and then annihilates with an electron. This generates a prompt signal that can be measured with a resolution ofσE=E ∼ 8% at 1 MeV and which preserves most of the incident antineutrino’s energy. The neutron is primarily captured by the gadolinium inside the central target, yielding an ∼8 MeV delayed signal. Requiring coincidence of the prompt and delayed signal pair effectively suppresses backgrounds.
A summary of the IBD candidates for the six-AD and eight-AD periods, together with the estimated background levels and the baselines of the three experimental halls to each pair of reactor cores, is shown in TableI. In the eight-AD period the backgrounds amount to only 2% of the total candidate samples in the near and far halls[31]. Two out of three Am-C calibration sources in the automated calibration units on the top of each far AD were removed during the installation of the two additional ADs in the summer of 2012, which reduced the far hall’s Am-C background by a factor of 4 compared to that in the previous publication. This data set also incorporates a reduction in the AD-uncorrelated energy scale uncertainty from 0.35% to 0.2% due to the implementation of better vertex- and time-dependent corrections [31]. This is one of the dominant systematic uncertainties, and is quantified by studying the differences in detector response using various calibration and natural radioactive sources.
The search for sterile neutrino mixing at the Daya Bay Reactor Neutrino Experiment is carried out through a relative comparison of the antineutrino rates and energy spectra at the three experimental halls. The unique con-figuration of multiple baselines to three pairs of nuclear reactors allows exploration ofΔm241spanning more than 3 orders of magnitude. Figure 2 shows the ratios of the observed prompt energy spectra at EH2 and EH3 to the best fit prediction from EH1 in the three-neutrino case. In this FIG. 1. Layout of the Daya Bay Reactor Neutrino Experiment.
The dots represent reactor cores, labeled as D1, D2, L1, L2, L3, and L4. The Daya Bay experiment started data taking with six antineutrino detectors (AD1–AD6) installed in three experimen-tal halls (EH1–EH3). From August to October 2012, two addi-tional detectors (AD7 and AD8) were installed in EH3 and EH2, respectively.
figure, the data are compared with the four-neutrino mixing scenario assuming sin22θ14¼ 0.05 for two representative Δm2
41 values, illustrating that the sensitivity at Δm241¼
4 × 10−2ð4 × 10−3Þ eV2originates primarily from the
rel-ative spectral shape comparison between EH1 and EH2 (EH3). The physical size of the Daya Bay reactor cores and detectors as well as the nonuniform distribution of the fission isotopes inside the cores have a negligible impact on the sensitivity.
The two different analysis methods used in the previous search [26] were updated to include the eight-AD data sample. Both methods, referred to as method A and method B, use the full expression in Eq.(2)to predict the neutrino oscillation signatures. The oscillation parameters sin22θ14, sin22θ13, and jΔm241j are set as free variables, while the others are constrained through external measurements: sin22θ12¼0.8460.021, Δm221¼ð7.530.18Þ×10−5eV2, andjΔm232j ¼ ð2.44 0.06Þ × 10−3 eV2[32]. The normal mass ordering is assumed for both Δm231 and Δm241, although this choice has only a marginal impact on the results.
Method A explicitly minimizes the dependence on the reactor antineutrino flux modeling [31]by predicting the prompt energy spectrum at the far hall from the measured spectra at the near halls. This process is done independently for each prompt energy bin i, by applying a weighting factor wiðΔm241; sin22θ14; sin22θ13Þ calculated from the known baselines and the reactor power profiles. The oscillation hypothesis is tested by evaluating a χ2 defined as
χ2¼X i;j
ðNfj− wjNnjÞðV−1ÞijðNfi − wiNniÞ; ð4Þ
whereNfðnÞi is the observed number of events after back-ground subtraction in theith bin at a far (near) detector, and V is a covariance matrix including both systematic and statistical uncertainties. The sensitivity to a spectral dis-tortion between the two near sites is retained by treating their data separately and by having indices i, j run over both the EH3-EH1 and EH3-EH2 combinations. A χ2 constructed with an alternative combination of the near and far detectors, such as EH2-EH1 and EH3-EH1, yields an equivalent sensitivity. All the sources of systematic error included in the most recent oscillation analysis of Ref.[31]
are considered, in addition to the uncertainty in the estimation ofΔm232.
Method B simultaneously fits the spectra from all ADs using the predicted reactor antineutrino flux. A binned log-likelihood function is constructed with nuisance parameters for the various systematic terms, including the detector response and the backgrounds. The reactor antineutrino flux is constrained based on the Huber [33]and Mueller
[34]fissile antineutrino models. The spectral uncertainties in the models are enlarged as motivated by the observed discrepancy between the predicted reactor antineutrino spectrum and the data [35–38], as well as by the recent reexamination of the systematic uncertainties in Ref.[39]. Specifically, the uncorrelated spectral uncertainties for
235U, 239Pu, and 241Pu are conservatively increased to
above 4%, while that of 238U is kept above 10%. The uncertainty of the predicted reactor¯νerate is also increased to 5%.
The two complementary analysis methods produce practically identical sensitivities for jΔm241j ≲ 0.3 eV2. Method A is more robust against uncertainties in the TABLE I. Summary of total number of IBD candidates and backgrounds, and baselines of the three experimental halls to the reactor pairs. Statistical and systematic errors are included.
Site IBD candidates Backgrounds Mean distance to reactor cores (m)
(Six ADs) (Eight ADs) (Six ADs) (Eight ADs) Daya Bay Ling Ao Ling Ao–II
EH1 205 135 408 678 4076.6 462.4 7547.9 908.0 365 860 1310
EH2 93 742 383 402 1580.3 147.8 5791.2 586.5 1348 481 529
EH3 41 348 108 907 1878.9 94.6 2105.2 208.1 1909 1537 1542
(Measured) / (Expected from EH1)
0.9 1
1.1 Data Unc. of 3ν prediction
2 eV -3 = 4x10 41 2 m Δ 2 eV -2 = 4x10 41 2 m Δ EH2 = 0.05 assumed 14 θ 2 2 sin
Prompt Energy (MeV)
1 2 3 4 5 6 7 8
0.9 1 1.1 EH3
FIG. 2. Prompt energy spectra observed at EH2 (top) and EH3 (bottom), divided by the prediction from EH1 with the three-neutrino best fit oscillation parameters from the most recent Daya Bay analysis [31]. The gray band represents the one-standard-deviation uncertainty of the three-neutrino oscillation prediction, which includes the statistical uncertainty of the EH1 data, as well as all the systematic uncertainties. Predictions with sin22θ14¼ 0.05 and two representative Δm2
41 values are also shown as the dotted and dashed curves.
predicted reactor antineutrino flux, while method B has a slightly higher reach in sensitivity forjΔm241j ≳ 0.3 eV2as a result of its incorporation of absolute reactor antineutrino flux constraints. The different treatments of systematic uncertainties provide a thorough cross-check of the results. For method A, the minimumχ2value obtained with a free-floating Δm241, sin22θ14, and sin22θ13 is χ24ν=NDF ¼ 129.1=145, where NDF stands for the number of degrees of freedom. The corresponding value in the three-neutrino scenario, in which sin22θ13 is the only free parameter, is χ2
3ν=NDF ¼ 134.7=147. The p-value of observing Δχ2¼
χ2
3ν− χ24ν¼ 5.6 without sterile neutrino mixing is
deter-mined to be 0.41 using a large sample of Monte Carlo pseudo-experiments. Similarly, the minimumχ2values for method B are χ24ν=NDF ¼ 179.74=205 and χ23ν=NDF ¼ 183.87=207, with a corresponding p-value of 0.42. As indicated by these p-values, no apparent signature for sterile neutrino mixing is observed.
The limits in theðjΔm241j; sin22θ14Þ plane are also set by two independent approaches, the first of which follows the Feldman-Cousins method [40]. For each point η ≡ ðjΔm2
41j; sin22θ14Þ, the value of Δχ2ðηÞ ¼ χ2ðηÞ−
χ2ðη
bestÞ is evaluated, where χ2ðηÞ is the smallest χ2value
with a free-floating sin22θ13. This Δχ2ðηÞ is then com-pared with the critical value Δχ2cðηÞ encompassing a fractionα of the events, estimated by fitting a large number of pseudo-experiments that include statistical and system-atic fluctuations. The pointη is then declared to be inside the α confidence level (CL) acceptance region if Δχ2dataðηÞ < Δχ2cðηÞ.
The second approach to set the limits is the CLs statistical method [41,42]. For each point in the (sin22θ14,jΔm241j) parameter space, a two-hypothesis test is performed in which the null hypothesisH0is the three-neutrino model and the alternative hypothesis H1 is the four-neutrino model with fixed sin22θ14 andjΔm241j. The CLs value is defined as
CLs¼1 − p1
1 − p0; ð5Þ
wherep0andp1are thep-values for the three-neutrino and four-neutrino hypotheses, respectively. Thesep-values are calculated from theχ2difference of those two hypotheses. The value of sin22θ13 is independently set for each hypothesis based on a fit to the data. The condition of CLs≤ 1 − α is required to set the CLs exclusion region at [α] confidence level.
When used with the same analysis method (method A or method B), the difference in sensitivity between the Feldman-Cousins and CLs approaches is found to be smaller than 10%. The Feldman-Cousins approach pro-vides a unified method to define confidence intervals, but has the drawback that it involves fitting a large amount of
simulated data sets. Hence, it is used only for method A, which eliminates all of the nuisance parameters by utilizing a covariance matrix. In contrast, the CLsimplementation is significantly less computationally intensive, and also pro-vides an alternative for combining the results between multiple experiments [41,42]. Accordingly, both the Feldman-Cousins limit from method A and the CLs limit from method B are presented in this work.
Figure3shows the 95% confidence level contour from the Feldman-Cousins approach and the 95% CLsexclusion contour. Both contours are centered around the 95% CL expectation and are mostly contained within the1σ band constructed from simulated data sets with statistical and systematic fluctuations. The high-precision data at multiple baselines allow exclusion of a large section of (sin22θ14, jΔm2
41j) parameter space. The sensitivity in the 0.01 ≲
jΔm2
41j ≲ 0.3 eV2 region originates predominantly from
the relative spectral comparison between the two near halls, and in thejΔm241j ≲ 0.01 eV2region from the comparison between the near and far halls. The dip structure at jΔm2
41j ≈ jΔm232j ≈ 2.4 × 10−3 eV2 is due to the
degen-eracy between sin22θ14and sin22θ13. The fine structure of the data contours compared to the expectation originates from statistical fluctuations in the data.
In Fig. 3, there is a slight difference between the CL contour from method A and the CLscontour from method B for jΔm241j ≲ 2 × 10−3 eV2. In this region, most of the
14 θ 2 2 sin -3 10 10-2 10-1 1 ] 2 | [eV 41 2 mΔ| -4 10 -3 10 -2 10 -1 10 Daya Bay 95% C.L. s Daya Bay 95% CL ) σ 1 ± Daya Bay 95% expected ( Bugey 90% C.L.
FIG. 3. Exclusion contours in the (sin22θ14, jΔm241j) plane, under the assumption ofΔm232> 0 and Δm241> 0. The red long-dashed curve represents the 95% CL exclusion contour with the Feldman-Cousins method[40] from method A. The black solid curve represents the 95% CLs exclusion contour [41] from method B. The expected 95% CL1σ band in yellow is centered around the sensitivity curve, shown as a thin blue line. The region of parameter space to the right side of the contours is excluded. For comparison, Bugey’s[43]90% CL limit on¯νedisappearance is also shown as the green dashed curve.
oscillation effects appear in the far hall at prompt energies ≲2 MeV, where the statistics are more limited. A study based on a large sample of Monte Carlo pseudo-experiments determined that the two methods react differ-ently to statistical fluctuations and produce slightly different limits in this region. The difference observed in Fig.3is found to be consistent with the expectation from this study at the ∼1σ level.
The resulting limits on sin22θ14 are improved by roughly a factor of 2 compared to the previous publication
[26]. The increased statistics are the largest contributor to this improvement, although the reductions in background and in the AD-uncorrelated energy scale uncertainty also play a role. The uncertainty in jΔm232j is the dominant systematic uncertainty in the jΔm241j ≲ jΔm232j region, while for higher values of jΔm241j the AD-uncorrelated energy scale and detector efficiency uncertainties are dominant. The total uncertainty is dominated by the statistics; another factor of 2 improvement in sensitivity is expected by 2017. This result can be combined withð−Þνμ disappearance searches [44] in order to constrain ð−Þνμ→
ν
ð−Þ
e transitions [45], since the oscillation probability of
ν
ð−Þ μ→ ν
ð−Þ
e in the four-neutrino scenario is approximately
proportional to jUe4j2jUμ4j2, and the individual sizes of jUe4j2 and jUμ4j2 can be constrained with ð−Þνe and ð−Þνμ
disappearance searches, respectively.
In summary, we report an improved search for light sterile neutrino mixing with the full configuration of the Daya Bay Reactor Neutrino Experiment in the electron antineutrino disappearance channel. No evidence of a light sterile neutrino is found through a relative comparison of the observed antineutrino energy spectra at the three experimental halls. With 3.6 times the statistics of the previous publication, these results set the most stringent limits to date on sin22θ14 in the 2 × 10−4≲ jΔm241j ≲ 0.2 eV2 region.
The Daya Bay Reactor Neutrino Experiment is supported in part by the Ministry of Science and Technology of China, the U.S. Department of Energy, the Chinese Academy of Sciences (CAS), the CAS Center for Excellence in Particle Physics, the National Natural Science Foundation of China, the Guangdong provincial government, the Shenzhen munici-pal government, the China General Nuclear Power Group, the Laboratory Directed Research and Development Program of the Institute of High Energy Physics, the Shanghai Laboratory for Particle Physics and Cosmology, the Research Grants Council of the Hong Kong Special Administrative Region of China, the University Development Fund of the University of Hong Kong, the MOE program for Research of Excellence at National Taiwan University, National Chiao-Tung University, MOST funding support from Taiwan, the U.S. National
Science Foundation, the Alfred P. Sloan Foundation, the Laboratory Directed Research and Development Program of Berkeley National Laboratory and Brookhaven National Laboratory, the Ministry of Education, Youth, and Sports of the Czech Republic, Charles University in Prague, the Joint Institute of Nuclear Research in Dubna, Russia, the NSFC-RFBR joint research program, and the National Commission for Scientific and Technology Research of Chile. We acknowledge the Yellow River Engineering Consulting Co., Ltd., and the China Railway 15th Bureau Group Co., Ltd., for building the underground laboratory. We are grateful for the ongoing cooperation from the China General Nuclear Power Group and China Light and Power Company.
*Present address: Department of Chemistry and Chemical
Technology, Bronx Community College, Bronx, New York 10453, USA.
[1] B. Pontecorvo, Sov. Phys. JETP 6, 429 (1957). [2] B. Pontecorvo, Sov. Phys. JETP 26, 984 (1968).
[3] Z. Maki, M. Nakagawa, and S. Sakata,Prog. Theor. Phys.
28, 870 (1962).
[4] K. A. Olive et al. (Particle Data Group),Chin. Phys. C 38,
090001 (2014).
[5] S. Schael et al. (ALEPH Collaboration, DELPHI Collabo-ration, L3 CollaboCollabo-ration, OPAL CollaboCollabo-ration, SLD Collaboration, LEP Electroweak Working Group, SLD Electroweak Group, SLD Heavy Flavour Group), Phys.
Rep. 427, 257 (2006).
[6] A. Palazzo,J. High Energy Phys. 10 (2013) 172.
[7] A. Aguilar et al. (LSND Collaboration),Phys. Rev. D 64,
112007 (2001).
[8] A. Aguilar-Arevalo et al. (MiniBooNE Collaboration),
Phys. Rev. Lett. 110, 161801 (2013).
[9] A. A. Aguilar-Arevalo et al. (MiniBooNE Collaboration),
Phys. Rev. Lett. 98, 231801 (2007).
[10] G. Mention, M. Fechner, T. Lasserre, T. A. Mueller, D. Lhuillier, M. Cribier, and A. Letourneau,Phys. Rev. D 83,
073006 (2011).
[11] W. Hampel et al. (GALLEX Collaboration),Phys. Lett. B
420, 114 (1998).
[12] J. N. Abdurashitov, V. N. Gavrin, V. V. Gorbachev, P. P. Gurkina, T. V. Ibragimova, A. V. Kalikhov, N. G. Khairnasov, T. V. Knodel, I. N. Mirmov, A. A. Shikhin, E. P. Veretenkin, V. E. Yants, G. T. Zatsepin, T. J. Bowles, S. R. Elliott, W. A. Teasdale, J. S. Nico, B. T. Cleveland, and J. F. Wilkerson (SAGE Collaboration), Phys. Rev. C 80,
015807 (2009).
[13] C. Giunti and M. Laveder,Phys. Rev. C 83, 065504 (2011). [14] G. Karagiorgi, Z. Djurcic, J. M. Conrad, M. H. Shaevitz, and
M. Sorel, Phys. Rev. D 80, 073001 (2009).
[15] J. Kopp, M. Maltoni, and T. Schwetz,Phys. Rev. Lett. 107,
091801 (2011).
[16] M. Maltoni and T. Schwetz, Phys. Rev. D 76, 093005
(2007).
[17] S. Gariazzo, C. Giunti, M. Laveder, Y. F. Li, and E. M. Zavanin, J. Phys. G 43, 033001 (2016).
[19] A. de Gouvea and T. Wytock, Phys. Rev. D 79, 073005
(2009).
[20] K. Bora, D. Dutta, and P. Ghoshal,J. High Energy Phys. 12 (2012) 025.
[21] S. K. Kang, Y.-D. Kim, Y.-J. Ko, and K. Siyeon,Adv. High
Energy Phys. 2013, 138109 (2013).
[22] P. Bakhti and Y. Farzan, J. High Energy Phys. 10 (2013) 200.
[23] A. Esmaili, E. Kemp, O. L. G. Peres, and Z. Tabrizi,Phys.
Rev. D 88, 073012 (2013).
[24] M. Bergevin, C. Grant, and R. Svoboda,arXiv:1303.0310.
[25] I. Girardi, D. Meloni, T. Ohlsson, H. Zhang, and S. Zhou,
J. High Energy Phys. 08 (2014) 057.
[26] F. An et al. (Daya Bay Collaboration),Phys. Rev. Lett. 113,
141802 (2014).
[27] J. L. Liu, B. Cai, R. Carr, D. Dwyer, W. Q. Gu, G. S. Li, X. Qian, R. D. McKeown, R. H. Tsang, W. Wang, F. F. Wu, and C. Zhang,Nucl. Instrum. Methods Phys. Res., Sect. A
750, 19 (2014).
[28] F. P. An et al. (Daya Bay Collaboration), Nucl. Instrum.
Methods Phys. Res., Sect. A A773, 8 (2015).
[29] F. An et al. (Daya Bay Collaboration), Nucl. Instrum.
Methods Phys. Res., Sect. A 685, 78 (2012).
[30] F. An et al. (Daya Bay Collaboration), Nucl. Instrum.
Methods Phys. Res., Sect. A 811, 133 (2016).
[31] F. An et al. (Daya Bay Collaboration),Phys. Rev. Lett. 115,
111802 (2015).
[32] K. A. Olive et al. (Particle Data Group),Chin. Phys. C 38,
090001 (2014).
[33] P. Huber,Phys. Rev. C 84, 024617 (2011).
[34] T. A. Mueller, D. Lhuillier, M. Fallot, A. Letourneau, S. Cormon, L. Fechner, M. Giot, T. Lasserre, J. Martino, G. Mention, A. Porta, and F. Yermia,Phys. Rev. C 83, 054615
(2011).
[35] F. P. An et al. (Daya Bay),Phys. Rev. Lett. 116, 061801
(2016).
[36] F. P. An et al. (Daya Bay Collaboration),arXiv:1607.05378.
[37] J. H. Choi et al. (RENO Collaboration),Phys. Rev. Lett.
116, 211801 (2016).
[38] Y. Abe et al. (Double Chooz Collaboration),J. High Energy
Phys. 10 (2014) 086 (2014);02 (2015) 74(E).
[39] A. C. Hayes, J. L. Friar, G. T. Garvey, G. Jungman, and G. Jonkmans,Phys. Rev. Lett. 112, 202501 (2014).
[40] G. J. Feldman and R. D. Cousins, Phys. Rev. D 57, 3873
(1998).
[41] A. L. Read,J. Phys. G 28, 2693 (2002).
[42] X. Qian, A. Tan, J. J. Ling, Y. Nakajima, and C. Zhang,
Nucl. Instrum. Methods Phys. Res., Sect. A 827, 63
(2016).
[43] Y. Declais et al. (Bugey Collaboration),Nucl. Phys. B434,
503 (1995).
[44] P. Adamson et al. (MINOS Collaboration), following Letter,
Phys. Rev. Lett. 117, 151803 (2016).
[45] P. Adamson et al. (DayaBay and MINOS Collaboration), preceding Letter,Phys. Rev. Lett. 117, 151801 (2016). [46] See Supplemental Material at http://link.aps.org/
supplemental/10.1103/PhysRevLett.117.151802, for χ2
