Prospects for dark matter detection with inelastic transitions of xenon Christopher M c Cabe preliminary results —work in progress— TeVPA, Tokyo, Japan - 27th October 2015
An old idea… • The original direct detection paper: Christopher M c Cabe GRAPPA - University of Amsterdam
An old idea… Inelastic scattering • What is it? • Why is it interesting? • Why consider it now? Can it ever be detected? Christopher M c Cabe GRAPPA - University of Amsterdam
What is it? DM elastic scattering: measure: DM N’s recoil energy N N r e c o i l DM inelastic scattering: measure: DM γ N’s recoil energy N N* r e c + photon energy o i l N Christopher M c Cabe GRAPPA - University of Amsterdam
What is a good target? XENON Christopher M c Cabe GRAPPA - University of Amsterdam
Why Xenon? Inelastic scattering is not A 2 enhanced ★ Only accessible for spin-dependent interactions ➡ Elastic and inelastic scattering rates comparable Vietze et al arXiv:1412.6091 ★ Ideal target should have i. good spin-dependent sensitivity ( . E DM − kinetic ≈ 100 keV) ii. a low lying excitation Christopher M c Cabe GRAPPA - University of Amsterdam
Why Xenon? • 47.6% of xenon sensitive to spin-dependent interactions: + 5/2 129 Xe 131 Xe 900 129 Xe 800 Natural abundance: 26.4% Excitation energy (keV) + + 3/2 700 5/2 + - Lowest excitation: 39.6 keV 1/2 7/2 + 7/2 600 + Lifetime: 0.97 ns (1/2,3/2) + (5/2) + 500 7/2 + 7/2 + (5/2) + + 1/2 3/2 400 131 Xe + + 5/2 3/2 + 5/2 - + 9/2 3/2 300 Natural abundance: 21.2% + - 5/2 (9/2) - 11/2 - 11/2 200 - Lowest excitation: 80.2 keV 9/2 - 11/2 + 3/2 Lifetime: 0.48 ns 100 + 11/2 1/2 + 3/2 + + + 1/2 1/2 3/2 0 Exp Exp Theory Christopher M c Cabe GRAPPA - University of Amsterdam
Previous studies • Previous searches with single phase-detectors • No limits or studies for two-phase detectors (LUX, XENON) Christopher M c Cabe GRAPPA - University of Amsterdam
Why is it interesting? Inferring properties of dark matter is difficult! We should search for all signals that provide information A detection should: • give independent evidence for dark matter scattering - point strongly to a spin-dependent interaction - help with mass reconstruction (because of different kinematics) - Christopher M c Cabe GRAPPA - University of Amsterdam
Why now? We can accurately quantify the signal and background - Structure functions known (needed for cross-section) - Backgrounds are more-or-less known - Future detector properties are more-or-less known Christopher M c Cabe GRAPPA - University of Amsterdam
An old idea… Inelastic scattering Can it ever be detected? Christopher M c Cabe GRAPPA - University of Amsterdam
Scattering rate • Rate depends on the DM velocity distribution: d 3 v f ( v ) dR Z ∝ g ( v min ) = dE R v v min Baudis et al 1309.0825 1 Standard Halo Model Double Power Law • v min is higher for inelastic Tsallis Model 0.1 (DM kinetic energy must g(v min )/g(0) also excite the nucleus) 0.01 • This suppresses 129 Xe 131 Xe the inelastic rate 0.001 Inelastic Inelastic Elastic by factor ~10 -4 10 0 100 200 300 400 500 600 700 800 v min (km/s) Christopher M c Cabe GRAPPA - University of Amsterdam
Structure functions • Known for axial-vector interaction: • Rate depends on the structure functions dR / d σ � 2 � h Xe ∗ | ¯ / S n ψ q γ µ γ 5 ψ q | Xe i � � A = dE R dE R ��� �� ��������� ��������� �� - � • Smaller for inelastic � ������� ( �� + �� ) � � � ��������� ( �� + �� ) � � (Small E R most relevant) n ( E R ) • This suppresses �� - � S A the inelastic rate by factor ~10 �� - � � �� �� �� �� ��� E R [ keV ] Baudis et al 1309.0825 Christopher M c Cabe GRAPPA - University of Amsterdam
The rate • Rate as a function recoil energy (not directly measured) � = �� - �� �� � � �� = ���� ���� σ � ��� �� ������� ��� �� ��������� �� � � / � � � [ ������ / � / �� / ��� ] ��� �� ������� ��� �� ��������� ����� ������� ����� ��������� � �� - � �� - � � �� �� �� �� ��� � � �� ��� �� ������� [ ��� ] • Inelastic rate smaller by factor ~100 ➡ Always see an elastic signal first Christopher M c Cabe GRAPPA - University of Amsterdam
Two-phase xenon detectors • Express the signal in terms of measured quantities: S 1 S 2 S2 = g 2 n e S1 = g 1 n γ γ E e- field Particle 52 phe 4540 phe g 1 , g 2 and drift field are the crucial parameters Christopher M c Cabe GRAPPA - University of Amsterdam
Mock detectors • I’ll consider two benchmark scenarios: XenonA200 XenonB1000 γ γ g 1 =0.07 PE/ g 1 =0.12 PE/ g 2 =12.5 PE/e g 2 =50 PE/e (50% extraction efficiency) (100% extraction efficiency) drift field=200 V/cm drift field=1000 V/cm • Number of photons & electrons modelled with NEST Szydagis et al 1106.1613 Christopher M c Cabe GRAPPA - University of Amsterdam
Mock signals • Include detector and recombination fluctuations � ⨯ �� � ��������� ���������� ��� ⨯ �� � γ ���� ��� �� γ ���� ��� �� �������� �� � [ �� ] �������� �� � [ �� ] + �� ��� �� � ⨯ �� � + �� ��� �� � ⨯ �� � � ⨯ �� � γ ���� ��� �� + �� ��� �� γ ���� ��� �� � ⨯ �� � + �� ��� �� � ⨯ �� � � ⨯ �� � �� ��� �� �� ��� �� � � � ��� ��� ��� ��� � ��� ��� ��� ��� ��� ��� �������� �� [ �� ] �������� �� [ �� ] • For same energy, electronic recoils produce a much larger S1 and S2 Christopher M c Cabe GRAPPA - University of Amsterdam
Mock signals 😄 • Looks like real data… 3 10 ��������� 60000 ��� ⨯ �� � γ ���� ��� �� �������� �� � [ �� ] + �� ��� �� 50000 80 keV +NR ee 2 10 � ⨯ �� � 40000 S2 [PE] γ ���� ��� �� + �� ��� �� 30000 � ⨯ �� � 40 keV +NR 10 20000 ee �� ��� �� NR 10000 � � ��� ��� ��� ��� 0 1 0 100 200 300 400 500 600 �������� �� [ �� ] S1 [PE] Data from PandaX-I arXiv:1505.00771 Christopher M c Cabe GRAPPA - University of Amsterdam
Background • Background spectra expected in LZ/XENONnT: LZ Design: 1509.02910 129 Xe 131 Xe �� � ���������� ����� � � / � � [ ������ / � / �� / ��� ] � νββ (± � %) �� 136 Xe ����� �� (± ��� %) � �� (± �� %) �� (± �� %) ����� � �� (± � %) �� - � ��������� (± �� %) � �� ��� ��� ��� ������ � [ ��� ] • 2-neutrino — 2-beta decay of 136 Xe dominates above 20 keV Christopher M c Cabe GRAPPA - University of Amsterdam
Reminder: Usual signal plane LUX arXiv:1310.8214 electronic recoil band nuclear recoil band signal region S1 < 30 PE Christopher M c Cabe GRAPPA - University of Amsterdam
Background versus signal • Signal region at higher values of S1 ����������� � ����� - ���� ���������� � ����� - ���� ��� � = �� - �� �� � � = �� - �� �� � � �� = ���� ���� σ � � �� = ���� ���� σ � ��� �� ��������� ��� �� ��������� ��� �� ��������� ��� �� ��������� �� ���� �� ���� ��� �� ���� �� ���� ��� ��� �� ( �� � / �� ) ��� �� ( �� � / �� ) ��� ��� ��� ��� ��� � ��� ��� ��� ��� ��� ��� ��� � ��� ��� ��� ��� ��� ��� ��� �������� �� [ �� ] �������� �� [ �� ] Large backgrounds…but some signal-to-background discrimination • Better discrimination for higher drift fields • Christopher M c Cabe GRAPPA - University of Amsterdam
Discovery limit • Quantify the sensitivity of future experiments with a ‘discovery limit’ Billard et al 1110.6079 The smallest cross-section at which 90% of experiments can make a 3 σ detection of the signal • Profile likelihood ratio: ˆ ˆ ~ � (0) = L ( � 0 n = 0 , A BG ) n , ˆ L ( ˆ ~ � 0 A BG ) - Include background uncertainties Christopher M c Cabe GRAPPA - University of Amsterdam
Recommend
More recommend