Discovering Inelastic Thermal Dark Matter Gordan Krnjaic + Eder - - PowerPoint PPT Presentation

Gordan Krnjaic

+ Eder Izaguirre, Brian Shuve

1508.03050

Cosmic Visions, UMD March 24, 2017

+ Eder Izaguirre, Yonatan Kahn, Matthew Moschella

1703.06881

Discovering Inelastic Thermal Dark Matter

Thermal Equilibrium Advantage #2: Narrows Mass Range

mDM

∼ 100M

∼ 10−20 eV

too hot too much < 10 keV > 100 TeV

GeV

mZ

MeV

nonthermal nonthermal

mP l ∼ 1019 GeV

“WIMPs”

Direct Detection (Alan Robinson) Indirect Detection (Alex Drlica-Wagner) Colliders (Yang Bai)

{

Light DM

{

18

< MeV

Thermal Contact Narrows Mass Range

Neff / BBN

2

CMB Bounds for light DM

1 10 100 1000 10000 mχ[GeV] 10−27 10−26 10−25 10−24 10−23 feff σv [cm3 s−1]

Thermal relic Planck TT,TE,EE+lowP WMAP9 CVL Possible interpretations for: AMS-02/Fermi/Pamela Fermi GC

Planck 1303.5076

Rules out s-wave annihilation < 10 GeV For viable models need:

(1) p-wave annihilation (2) annihilation shuts off before CMB OR

3

CMB Bounds for light DM

1 10 100 1000 10000 mχ[GeV] 10−27 10−26 10−25 10−24 10−23 feff σv [cm3 s−1]

Thermal relic Planck TT,TE,EE+lowP WMAP9 CVL Possible interpretations for: AMS-02/Fermi/Pamela Fermi GC

Planck 1303.5076

Rules out s-wave annihilation < 10 GeV For viable models need:

(1) p-wave annihilation (2) annihilation shuts off before CMB OR

4

nχ2 ∼ e−∆/T

χ1 SM χ2 SM

∆ ⌘ mχ2 mχ1 eV Heavier state disappears before z~1100 No (tree level) direct detection ∆ > 100 keV

Easy to build, large couplings, hard to test!

No indirect detection

Inelastic DM is CMB Safe

Direct Coannihilation into SM

iDM direct detection: Weiner, Tucker-Smith arXiv: 0101338

L ⊃ gDA0

µ ¯

ψγµψ + M ¯ ψψ + HD ¯ ψcψ

Dirac mass Vector current Charge 2 dark Higgs

Example Model

Four component fermion + familiar dark photon

L ⊃ gDA0

µ ¯

ψγµψ + M ¯ ψψ + HD ¯ ψcψ

Break dark U(1) with dark Higgs VEV

Dirac mass Vector current Charge 2 dark Higgs

Lmass = M ¯ ψψ + hHDi ¯ ψcψ

Dirac Majorana

Example Model

Four component fermion + familiar dark photon

L ⊃ gDA0

µ ¯

ψγµψ + M ¯ ψψ + HD ¯ ψcψ

Break dark U(1) with dark Higgs VEV

Dirac mass Vector current Charge 2 dark Higgs

Lmass = M ¯ ψψ + hHDi ¯ ψcψ

Diagonalizing to mass basis splits Dirac components (pseudo-Dirac)

Dirac Majorana

ψ ≡ (ξ, η†)

• int. eigenstates

(χ1, χ2) , ∆ ≡ m2 − m1

mass eigenstates

Example Model

Four component fermion + familiar dark photon

A0 γ χ χ e+ e

e

✏

χ1 χ2 gD

y ≡ ✏2↵D ✓ m1 mA0 ◆4

Vector current is now off-diagonal in mass basis

L ⊃ gDA0

µ ¯

χ2γµχ1 + h.c.

As before, define relic density variable

mA0 > m1 + m2

direct annihilation Different “y” for each ∆

Example Model

freeze out is subtle…

Coannihilation

χ1 χ2 A0 f f

Upscahttering &

e e A0 χ2 χ1

·

χ1 A0⇤ e e+ χ2

Excited State Decays

Γ(2 → 1 e+e−) = 4✏2↵↵D∆5 15⇡m4

A0

Downscattering

Inelastic Novelties

Y1(0) Y2(0) Y1 Y2 10 20 30 40 50 60 70 10-17 10-13 10-9 10-5 10-1 x = m2/T Y = n / s

iDM Thermal Freeze-Out

Heavier state feels Boltzmann suppression earlier Need larger rate to compensate!

number density t i m e

Coannihilation Relics

1 10 102 10-16 10-15 10-14 10-13 10-12 10-11 10-10 10-9 10-8 10-7 10-6

m1 [MeV] y = ϵ2 αD (m1/mA')4

Thermal Coannihilation, mA' = 3 m1

∆ = 0.1m1 ∆ = 0.4m1

Vary Mass Splitting

Beam e/p ! Target/Dump Detector

χi

A Z e/p e/p χ1 χ2 p Z γ A π0, η χ1 χ2

A χi χj T T and/or χ1 χ2 f f + A

Proton LSND MiniBooNE Electron

Beam Dump Signals

E137 BDX

Others possible (SeaQuest, T2K, DUNE…)

Kim Park Shin 1612.06867 BdNMC deNiverville, Chen, Pospelov, Ritz 1609.01770 Morrissey, Spray 1402.4817

can then decay promptly inside the detector to deposit a visible signal.

e− − → ECAL/HCAL Target Tracker e− χ1χ2 Invisible e− − → Active Target (ECAL/HCAL) e− χ1χ2 Invisible

A Z e e χ1 χ2

Heavier state decays outside veto region Signal looks like missing energy/momentum LDMX NA64

Missing Energy/Momentum

May also be sensitive to the decay!

Δ = 0.1 m1 Δ = 0.2 m1 Δ = 0.4 m1

10 102 103 10-2 10-1 100 101 102 103 104 105 106 107 108 109 m2 [MeV] c τ [yrelic/y] [cm]

Rest Frame Decay Length χ2 → χ1 f f , y = yrelic

Generically Macroscopic Decays

LDMX missing mom. LSND scatter (g-2)μ MiniBooNE scatter

R e l i c D e n s i t y

LEP BaBar mono γ BDX scatter Belle II (g-2)μ E137 scatter

N eff , (model dep.)

→ 1 10 102 103 10-16 10-15 10-14 10-13 10-12 10-11 10-10 10-9 10-8 10-7 10-6

m1 [MeV] y = ϵ2 αD (m1/mA')4

Thermal iDM, Δ = 0.01 m1 , mA' = 3 m1 , αD = 0.1

Tiny Splitting ~ 1%

Similar to plots from plenaries

LDMX missing mom. LSND decay LSND scatter MiniBooNE decay E137 decay BDX decay BDX scatter E137 scatter (g-2)μ

Relic Density

N eff , (model dep.)

BaBar mono γ

→

LEP

1 10 102 103 10-16 10-15 10-14 10-13 10-12 10-11 10-10 10-9 10-8 10-7 10-6

m1 [MeV] y = ϵ2 αD (m1/mA')4

Thermal iDM, Δ = 0.1 m1, mA' = 3 m1 , αD = 0.1 Thermal iDM, Δ = 0.3 m , m = 3 m , α = 0.1

Small Splitting ~ 10%

m1 [MeV]

LDMX missing mom. LSND decay (g-2)μ E137 scatter E137 decay LSND scatter MiniBooNE decay BDX decay BDX scatter BaBar mono γ LEP

Relic Density

→

N eff , (model dep.)

1 10 102 103 10-16 10-15 10-14 10-13 10-12 10-11 10-10 10-9 10-8 10-7 10-6

m1 [MeV] y = ϵ2 αD (m1/mA')4

Thermal iDM, Δ = 0.4 m1 , mA' = 3 m1 , αD = 0.1

Target moves up, bounds/projections move down

Large Splitting ~ 40%

Vary DM/Mediator Coupling

LDMX missing mom. LSND decay E137 decay BDX decay BDX scatter E137 scatter LSND scatter LEP BaBar mono γ (g-2)μ

Relic Density

N eff , (model dep.)

→

MiniBooNE decay

1 10 102 103 10-16 10-15 10-14 10-13 10-12 10-11 10-10 10-9 10-8 10-7 10-6

m1 [MeV] y = ϵ2 αD (m1/mA')4

Thermal iDM, Δ = 0.1 m1, mA' = 3 m1 , αD = α

LDMX missing mom. LSND decay LSND scatter MiniBooNE decay E137 decay BDX decay BDX scatter E137 scatter (g-2)μ

Relic Density

N eff , (model dep.)

BaBar mono γ

→

LEP

1 10 102 103 10-16 10-15 10-14 10-13 10-12 10-11 10-10 10-9 10-8 10-7 10-6

m1 [MeV] y = ϵ2 αD (m1/mA')4

Thermal iDM, Δ = 0.1 m1, mA' = 3 m1 , αD = 0.1 Thermal iDM, Δ = 0.3 m , m = 3 m , α = 0.1

LDMX missing mom. LSND decay LSND scatter MiniBooNE decay E137 decay BDX decay BDX scatter E137 scatter (g-2)μ

Relic Density

N eff , (model dep.)

BaBar mono γ

→

LEP

1 10 102 103 10-16 10-15 10-14 10-13 10-12 10-11 10-10 10-9 10-8 10-7 10-6

m1 [MeV] y = ϵ2 αD (m1/mA')4

Thermal iDM, Δ = 0.1 m1, mA' = 3 m1 , αD = 0.1 Thermal iDM, Δ = 0.3 m , m = 3 m , α = 0.1

Vary DM/Mediator Mass Ratio

m1 [MeV]

LDMX missing mom. LSND decay LSND scatter MiniBooNE decay BDX scatter E137 scatter (g-2)μ

R e l i c D e n s i t y

BaBar mono γ BDX decay

N eff , (model dep.)

E137 decay

→

LEP

1 10 102 103 10-16 10-15 10-14 10-13 10-12 10-11 10-10 10-9 10-8 10-7 10-6

m1 [MeV] y = ϵ2 αD (m1/mA')4

Thermal iDM, Δ = 0.1 m1, mA' = 10 m1 , αD = 0.1

p p j DM∗ DM DM , `+`− . . . ← − c⌧DM∗

Hadron Collider Lepton Collider

e+ e−

• DM∗

DM DM , `+`− . . . ← − c⌧DM∗

J + 6ET + `+`− + 6E + `+`−

Izaguirre, GK, Shuve 1508.03050

Above the GeV Scale?

LDMX missing mom. LSND decay LSND scatter MiniBooNE decay E137 decay BDX decay LHC displaced Belle II mono γ BDX scatter LEP BaBar displaced E137 scatter (g-2)μ

N eff , (model dep.)

→

R e l i c D e n s i t y

1 10 102 103 104 105 10-15 10-14 10-13 10-12 10-11 10-10 10-9 10-8 10-7 10-6

m1 [MeV] y = ϵ2 αD (m1/mA')4

Thermal iDM, Δ = 0.1 m1 , mA' = 3 m1 , αD = 0.1

Small Splitting ~ 10%

Collider Complementarity

m1 [MeV]

LDMX missing mom. LSND decay E137 scatter E137 decay LSND scatter

R e l i c D e n s i t y

MiniBooNE decay BDX decay LHC l+l-+MET BaBar displaced LHC displaced BDX scatter (g-2)μ

N eff , (model dep.)

→

LEP

1 10 102 103 104 105 10-15 10-14 10-13 10-12 10-11 10-10 10-9 10-8 10-7 10-6

m1 [MeV] y = ϵ2 αD (m1/mA')4

Thermal iDM, Δ = 0.4 m1 , mA' = 3 m1 , αD = 0.1

Large Splitting ~ 40%

Collider Complementarity

Conclusion

Coannihilation Freeze Out

• Mass difference changes freeze out
• Need larger couplings (increases with splitting!)

Fixed-Target, Neutrino, & B-Factory Experiments

• Still have scattering/missing energy searches
• Two level dark sector (pseudo-Dirac example)
• Also have powerful decay searches for excited state

Can test nearly all scenarios

• Increasing the splitting doesn’t decouple the bounds
• Covering splittings up to ~ 50% gets everything!
• Other experiments? SeaQuest, DUNE, NOvA
• Collider displaced vertex searches @ higher masses