Direct Measurement of Neutrino Mass -2 Flavio Gatti University and - PowerPoint PPT Presentation

Direct Measurement of Neutrino Mass -2 Flavio Gatti University and INFN of Genoa E.Fermi School, ISAPP 2011 Varenna August 2nd, 2011 1

Calorimetric spectroscopy Initial motivation: perform a model  independent measurement. External source spectrometers need  a precise model of the atom- ß i, molecule to calculate in particular the so called “final states effect” External source spectrometers need  also a model of the energy losses in the source material, scattered trajectories,… dN/A Models contains unknown  systematics (see problem of m 2 <0 of the years ’90  1996 PDG excluded all these determinations)  2

Principles of the calorimetric method In a calorimeter the energy E i = E( b i )+ D i is measured for  each event Then the spectrum become dN(E) = A S i w i (E i -E 0 ) 2 dE  3

Principles of the calorimetric method Advantages: no model dependent corrections for atomic and molecular  final states. no correction for nuclear recoil energy and for electron  energy losses. Drawbacks: Beta source inside the detector: whole spectrum must be  acquired and the interesting area is proportional only to (mc 2 /E 0 ) 3 Needed to work with Low Q Value Isotopes  187 Re : lowest Q ~ 2.5 keV.  187 Re: (mc 2 /E 0 ) 3 ~1/400 of H 3  4

Simple analytical estimation of the sensitivity  Counts in D E below the end-point E 0 at fixed m n  Counts at m n =0  Counts at m n ≠0  Pileup counts  Sensitivity 5

MC simulation tested on pilot experiments and extrapolated to a very high statistics experiments 6

Few historical notes  The first calorimetric experiment applied to the beta decay has been made by Ellis and Wooster in 1927  At that time it was established that “ a - ray” were emitted as mono-energetic lines by nuclei, as expected within the general framework of the quantum theory of the “disintegration of the bodies”  But the “ b - ray” behavior is in sharp contrast to this: the kinetic energy spectrum is widely distributed.

Further Notes  “ b -spectrum is continuum because of the slowing down in the material” (Lisa Meitner) or “in collision with atomic electron” (E.Rutherford)  The results was <E> calorimeter = 0.33 ± 0.03 MeV/atom against E max =1.05 MeV/atom   ”Not conservation of energy” ( N.Bohr)   E max - <E> “carried out by escaping particle” ( Heisemberg)  Pauli conjecture of the neutrino (1930)  First fully calorimetric detector for b -decay even if not able to detect single particle.

Single particle detection with thermal detector in1949: a technique incredibly similar to the present one

The beginning of the calorimetric beta spectroscopy 

The “Simpson” experiment Apparatus R(E) S(E)= ∫ R(E) O b (E)

The “Simpson” experiment: how to check that it was “calorimetric” Initial state: implanted 3 H moves under the effect of channeling and  places as neutral atoms in the tetragonal structure od Si  atomic 3 H in well defined site. Final state: 3 He+ in 1s or 2s (+ possible shake-off processes)   recombination processes: (a) second electron in 1s (1S 0 )(<10ns),  (b) in 2s (meta-stable if free) but it decays faster (radiatively) , due to  screening of Si electrons and/or via Stark mixing with 2p 1/2 to 1S in t<ns (58,4 nm emission). The main de-excitation processes involves emission of several tens  of eV (20 eV 1s2p-1s1s), while the energy gap of silicon is 1.1 eV , w=3.66 eV (F.Sholze, JAP(1998)), It can be considered an Energy Dispersive “charge calorimeter” for  beta decay.

First steps towards “pure” thermal calorimetric beta spectroscopy 1985 – Dan Mc Cammon (Univ.  of Wisconsin) proposed to adopt a fast thermal calorimeter to tritium beta decay spectroscopy (AIP Conf. Proc. 1985) 1985 - First conceptual proposal  of approach to the calorimetric spectroscopy method of determining neutrino mass by 185 Re 75 5/2 + using 187-Re (S.Vitale, Univ. INFN Genova) INFN Report /BE- 85/2) b- 187 Os 76 , 1/2 -

How a calorimeter works b V I  R

187-Re decay in a crystal: is it a true calorimeter? .  HCP lattice  T c =1.69 K   = 21 g/cm 3  T(Debye)= 460 K  M.P.=3000 K  Z=75  A=185(37%), 187(63%)  1/2 Re-187=4x10 10 y Initial and final states are in the crystal The spectrum end-point energy is lower than the one of isolated isotope.  E endpoint =(Q-m e c 2 )-(e f +E Fermi )- D B lattice  where E Fermi =11.2 eV,  Work function f =5.1 eV  Crystal binding energy B lattice = 16.9 eV  Change of binding from Re->Os D B lattice =2.7% B lattice  16

187-Re decay in a crystal: case of Rhenium Metal. 75 Re [Xe] 4f14 5d5 6s2 , Etot= -429402.3 eV  76 Os+ [Xe] 4f14 5d6 6s1, Etot= -443164.5 eV  76 Os [Xe] 4f14 5d6 6s2 Etot=-443172.8 eV   D B coul =13.7KeV greater than D E nucl   the bare 187-Re cannot undergo continuum b -decay. Bound state  decay of 187Re 75+ has been observed in storage ring having 32 y half life and 63KeV Q value During the decay the beta particle pass through the atom. The electron  may not have the time to rearrange the electrons, the atomic binding energy difference Re-Os+ is very close to the binding energy difference of the initial end final atomic state Being the energy of the final state Os + after the decay almost that one of  the ground state of Os + , high excited state of te final atom are very unlikely Further, due to the very similar atomic wave-function the probability of a  transition toward an excited state is very small being Os eigenstate orthogonal to the one of Re. A first evaluation of this probability is 7x10 -5 .

187-Re decay in a crystal: case of Rhenium Metal.  Recoil energy at level of few meV respect to several eV per dislocations; recoil contribute directly to the generation of phonon of elastic branch.  Recoil free beta decay (not yet observed) but extending Mossbauer and tacking into account the so small recoil effect, this should be negligible.  Shake off probability at 1% level only for N and O shells, that can emits photons of 50 eV (avg) or Auger electron. They are fully absorbed in hundreds of Ang.  Inner Bremsstrahlung: same Q value but larger penetration depth, however at um level  Esternal Bremsstrahlung: can be fully contained as before  Collective excitation based on long living quasi-particle states

Collective Excitations: Phonon and qp after the primary events: simulation at 6keV Energy trapped in qp. Recondense from us to 100 ms depending on T/T c Heat promtly read out by calorimeter What is the prompt thermalization efficiency?  Ecal/Eparticle

There are mechanism that speed up the thermalization of qp 20

First pilot experiment (Genoa)

Another solid state effect: Beta Environmental Fine Structure (BEFS) Residuals from spectrum fitting

BEFS E b >>E(Fermi)  beta electrons interacts with  atomic cores. k 2 =2m(E( b )-V)/h 2 ,,   l (100 eV) ~ 0.2A, l (1000) If V ~ -15 eV  ~0.04A a=2.76A, c=4.45 A, c/a=1.61A (1.63 A).  Self interference of outgoing and reflected waves  from atomic shells: Oscillation  (backscattering amplitude) x (self-  interference amplitude on Re nucleus from each atomic shell) x (number of atoms of shell) Thermal motion energy: T  0  ~ exp( 2k 2 /M Q D ) b wave attenuation (“range”):  ~ exp(- g R), g ~3-20 A First hypothesis: S.E.Koonin in 91(Nature  354,486), never observed.

Second Pilot expriment (Milan)

Microcalorimeter: some details C(T) G(T,T b ) T b P g R(T) P lin k . Thermal model contains non-  linear terms Linearized equations give simple  exponential thermal response

Sensor: case of superconducting Transition Edge Sensor (TES) Insert Sensor Model  a = (T/R) dR/dT Insert bias power for sensor  Sensor readout sensitivity Improve the model taking into  account of all C and G terms including their models Set of non- linear equation…       dT    - - -  n n n n 2 TES C K T T K T T R T I  2 1 TES Abs TES TES h x TES b dt     dT    - -  n n Abs C K T T P t  b Abs 2 Abs TES dt     dI q     -    b R I t I R T I L 2   -   -  st 0 b x TES b p T T T T dt C 0 0    -  -    2     R T ( ) R R 1 e H 1 e dq 0 s       I  b dt

Electro-thermal feedback Main concept of the whole dynamic response of the TES coupled calorimeter:  the bias power act as negative feedback reducing thermal swing and time response  more linear and fast response Parameter of ETF : L = a P bias /GT b   =(C/G) 1(1+L)  L range: 10-10 2         , , , ,    y0b S1 y1b S1 Ib S1 0.0839 0.0837 ETF effect 0.0835 T [K] 0.0833 0.0831 0.0829 2 . 10 5 4 . 10 5 6 . 10 5 8 . 10 5 1 . 10 4 0 t [s] TES w ETF Abs w ETF TES absorber