HREPHAH 0H3HKA, 2004, moM 67, № 11, c. 1977-1982
= NEUTRINO PHYSICS =
SENSITIVITY AND SYSTEMATICS OF KATRIN EXPERIMENT
© 2004 N. A. Titov (for the KATRIN Collaboration^)
Institute for Nuclear Research, Russian Academy of Sciences, Moscow Received January 20, 2004
KArlsruhe TRItium Neutrino experiment (KATRIN) will measure a "kinematical" electron antineutrino mass upper limit up to 0.2 eV/c2. The experimental setup based on electrostatic spectrometer with adiabatic magnetic collimation and windowless gaseous tritium source is briefly described. This sensitivity to the neutrino mass could be reached with a 10-m diameter spectrometer after three years of data taking. Several major sources of the systematic errors are discussed.
A neutrino oscillation is a well-established fact that implies a nonzero neutrino mass [1]. Oscillation data provide us with a neutrino mass spectral pattern, but not the absolute mass values. It is only possible to deduce that at least one neutrino mass eigenstate is heavier than 0.03 eV/c2. A "kinematical" experiment based on analysis of kinematics of a weak decay is the only laboratory experiment suitable to provide an absolute neutrino mass value. Particularly, the proposed KATRIN setup will be able to set an electron antineutrino mass upper limit at the level of 0.2 eV/c2. Such a study makes sense because there are two neutrino mass schemes: hierarchical and quasidegenerate (Fig. 1). In the hierarchical scheme the mass eigenstates have a different scale determined by Am2tm and Am2ol and the heaviest mass is about 0.05 eV/c2. In the quasidegenerate scheme all neutrinos have about the same mass much larger than the mass splitting. The latter scheme is somewhat favored by the observed large mixing angles of different mass eigenstates sin dj k> 1 [1]. The neutrino mass in the quasidegenerate scheme has a chance to be detected in the tritium experiment (otherwise this scheme will be mostly excluded).
The tritium f3 decay is a superallowed transition. In the quasidegenerate regime neutrino mass splitting can be neglected and the electron spectrum is described by the well-known formula:
dN/dE = KF(Z, E)pEtot x (1)
x (Eo - E)2\/(Eq — E)2 — m2.
KATRIN Collaboration: U Bonn; LNP/JINR, Dubna; FH Fulda; FZ & U Karlsruhe; U Mainz; INP, Rez; RAL; UW, Seattle; UW Swansea; INR RAS, Troitsk. * E-mail: titov@al20.inr.troitsk.ru
Almost all of the spectrum data points have (E0 — — E)2 » ml and the neutrino mass signature is a negative constant shift of the parabolic spectrum with respect to the background level (Fig. 2):
dN/dE - (Eo — E)2 — m2v/2. (2)
The absolute value of statistical error bar is a linear function of the distance from the endpoint (E0 — E). The sensitivity to nonzero neutrino mass is steadily vanishing far from the spectrum endpoint as it is shown on the inserted graph in Fig. 2.
A real experimental parameter in the tritium-decay experiment is a neutrino mass square (see (1)). During the last 10—15 years the experimental sensitivity to neutrino mass square was improved by about two orders of magnitude (Fig. 3). This improvement was achieved by invention of an electrostatic spectrometer with an adiabatic magnetic collimation (AMC). The AMC allows high-resolution and high-luminosity requirements to be decoupled. This idea was independently developed by several researchers [2]. The AMC is based on a conservation (as an adiabatic invariant) of the ratio of transversal
Table 1. KATRIN parameters
WGTS Spectrometer Measurements
Tritium column density 5 x 1017 mol/cm2 Diameter 90 mm (effective 81 mm) Acceptance angle 51° Tritium purity 95 % Diameter 10 m (effective 9 m) Resolution AE = 0.93 eV Three years of data taking Optimized set of data points
1977
1978
TITOV
m, eV/c2
100
Am23
vatm
Am22
Vsol
LMA
10
1-2
10-4 , 10-4
: Mainz & Troitsk
F l KATRIN
m3 55 meV/c2) \ Quasidegenerate v masses
m2 (~ 8 meV/c 2)
m1 Hierarchical v masses
10
-2
100
m1; eV/c2
Fig. 1. Neutrino mass scheme. Hierarchical and quasidegenerate parts are shown. The current experimental mass limit [2] as that of KATRIN [3] are shown.
Fig. 2. The exact tritium ^-electron energy spectrum and its parabolic approximation. The neutrino mass signature is a shift of the parabolic part of the spectrum. The absolute data error bar increases for high-intensity points. On the inserted graph the signal-to-error bar ratio is presented for different background levels.
kinetic energy to magnetic field strength
V = Et/B (3)
for a charged particle moving through the magnetic field. For the conservation of the adiabatic invariant it is only required that along the trajectory of a
moving particle the magnetic field variation should be small on the time scale of one gyration. The idea of the AMS is that a tritium /3 decay takes place in a strong magnetic field within a large solid angle (Fig. 4). When electrons are transported to the low-magnetic-field region their moments become aligned
SENSITIVITY AND SYSTEMATICS OF KATRIN EXPERIMENT mV, eV2/c4
1979
-50
-100
-150
Beijing
Livermore
Los Alamos
Mainz
Tokyo
Troitsk
Troitsk
Zuerich
Electrostatic spectrometers
Magnetic spectrometers
1990
1992
1994
1996
1998
2000 Year
Fig. 3. The electron antineutrino mass-square measurements summary for previous decade.
0
along the magnetic field due to the conservation of the adiabatic invariant. The aligned electron energy can be analyzed with an electrostatic spectrometer. To define the tritium decay solid angle electrons should pass the region with magnetic field Bmax somewhat stronger than one in the decay region. Finally, one obtains an integral spectrometer — a high pass filter at an electrostatic mirror potential U0 with a full width resolution:
AE = \eUo\
B.
analys
B„
(4)
Thus, the spectrometer resolution is only determined by the magnetic fields ratio with an acceptance solid angle of the order of 1 sr and an arbitrary source diameter.
The second important invention is a windowless gaseous tritium source (WGTS) first used in the LANL experiment [4] and significantly modified later at Troitsk [5]. The WGTS provides an excellent intensity of P electrons, has high uniformity and well-controlled energy losses.
The KATRIN project united almost all experts in the field with the aim to get ultimate sensitivity to mv using the electrostatic spectrometer with AMC and the WGTS. The KATRlN setup is given in Fig. 5. The tritium P decay takes place inside WGTS surrounded by the differential pumping stations. The outgoing tritium is collected, purified, and reinjected in the
center of WGTS. The decay electrons are guided by magnetic field through WGTS and the cryotrapping section toward the pre- and main spectrometer. In order to have a low spectrometer background, the multiple differential pumping stations and cryotrapping sections have to reduce tritium partial pressure in the spectrometer compared with WGTS by a factor of 1016. Inside the pre- and main spectrometers the decay electron energies are analyzed by the electrostatic mirror formed by a set of electrodes. After being analyzed, the decay electrons are registered by a segmented semiconductor detector. The total setup length is up to 90 m long. Magnetic field values
Bs,
» B
analys
Uo
<D
Fig. 4. The operation principle of the electrostatic spectrometer with an adiabatic magnetic collimation. Electron moments are aligned along the guiding magnetic field after transition into the low-field region and are analyzed with electrostatic barrier.
ftOEPHAfl OH3HKA TOM 67 № 11 2004
1980
TITOV
Rear diff. pumping
WGTS
T2
in side
Prespectrometer
Front diff. pumping
Cryotrapping
Fig. 5. The KATRIN setup general view. The tritium ¡3 decay takes place inside the windowless gaseous tritium source surrounded by the differential pumping stations. Decay electrons are guided by magnetic field through the cryotrapping section toward the pre- and main spectrometer. After being analyzed by an electrostatic mirror the decay electrons are registered by a semiconductor detector.
are 3.6 (WGTS), 6 (maximal field), and 0.0003 T (analyzing plane). The WGTS diameter is 90 mm, main spectrometer diameter is 10 m, spectrometer vacuum is up to 10"12 mbar.
To evaluate the KATRIN setup statistical sensitivity one should take into account the fact that several parameters can be further optimized but others are interrelated. The spectrometer resolution and acceptance define the magnetic field ratios and geometry of the vessels. Technical limitations are a maximal diameter of the extrahigh-vacuum spectrometer vessel (10 m) and a maximal field in the superconducting magnets (6 T). For the ultimate accuracy only the electrons leaving WGTS without inelastic scattering are useful (see below). The source column density and acceptance angle are selected in such a way that the nonscattering fraction of outgoing decay electrons is near saturation. The spectrometer resolution improvement below 1 eV provides no sensitivity increase, because the effective resolution is limited by roto-vibrational excitation of a recoil molecular ion (3HeT)+. At last, a data point distribution was optimized to reach maximal sensitivity. A set of parameters for the sensitivity calculations is presented in Table 1. Assuming
asyst ^ CTstat and a background level of 10 mHz with parameters from Table 1 one
Fig. 6. Energy flow chart. First line — corrections to the released energy. Second line — the nuclear mass difference is a released energy main source. Other lines — different energy drains.
obtains the KATRIN neutrino mass upper limit to be 0.2 eV/c2 (90% C.L.).
Considering the KATRIN systematic uncertainties one should keep in mind that the neutrino mass signature is some drain of a part of /-electron kinetic energy to the neutrino rest mass. All other competing sources of the energy drain can mimicry the neutrino mass. The chart of the energy flow is presented in Fig. 6.
Для дальнейшего прочтения статьи необходимо приобрести полный текст. Статьи высылаются в формате PDF на указанную при оплате почту. Время доставки составляет менее 10 минут. Стоимость одной статьи — 150 рублей.