Energy-Injection Constraints from CMB and How to Evade Them Sam - - PowerPoint PPT Presentation

Energy-Injection Constraints from CMB and How to Evade Them

Sam McDermott New Directions in the Search for Light Dark Matter Particles

What’s the Issue?

10 100 1000 10000 DM mass (GeV) 10-27 10-26 10-25 10-24 10-23 10-22 <σ v> (cm3/s) 10-6 10-5 10-4 10-3 10-2 10-1 100 DM mass (GeV) 10-33 10-32 10-31 10-30 10-29 10-28 10-27 10-26 <σ v> (cm3/s)

Annihilation channels:

Slatyer, 1506.03811 & PRD

What’s the Issue?

10 100 1000 10000 DM mass (GeV) 10-27 10-26 10-25 10-24 10-23 10-22 <σ v> (cm3/s) 10-6 10-5 10-4 10-3 10-2 10-1 100 DM mass (GeV) 10-33 10-32 10-31 10-30 10-29 10-28 10-27 10-26 <σ v> (cm3/s)

Annihilation channels:

Slatyer, 1506.03811 & PRD

Outline

• Energy Injection During CMB
• P-wave annihilation
• Hidden Sector
• Asymmetric Dark Matter
• Freeze-in

Outline

• Energy Injection During CMB
• P-wave annihilation
• Hidden Sector
• Asymmetric Dark Matter
• Freeze-in

Caveats: Many other good ideas to avoid CMB constraints! Coannihilation (annihilation with partner, for inelastic states Griest & Seckel 1991, D’Agnolo & Ruderman 2018), forbidden/ impeded dark matter (Griest & Seckel 1991, D’Agnolo & Ruderman 2015, Kopp et al.

2016,), coscattering/co-decaying/“elastic

decoupling”(D’Agnolo & Ruderman 2017, Dror et al. 2016, Kuflik et al. 2017), … (many more)

Outline

• Energy Injection During CMB
• P-wave annihilation
• Hidden Sector
• Asymmetric Dark Matter
• Freeze-in

Energy Injection During CMB

Why is it plausible that DM annihilation is strongly constrained by CMB?

(shamelessly stolen from a talk I remember seeing by Tracy)

Energy Injection During CMB

Why is it plausible that DM annihilation is strongly constrained by CMB?

• binding energy of hydrogen = 13.6 eV

(shamelessly stolen from a talk I remember seeing by Tracy)

Energy Injection During CMB

Why is it plausible that DM annihilation is strongly constrained by CMB?

• binding energy of hydrogen = 13.6 eV
• single annihilation of mDM ~ O(MeV) can

ionize 105 H atoms

(shamelessly stolen from a talk I remember seeing by Tracy)

Why is it plausible that DM annihilation is strongly constrained by CMB?

• binding energy of hydrogen = 13.6 eV
• single annihilation of mDM ~ O(MeV) can

ionize 105 H atoms

Energy Injection During CMB

(shamelessly stolen from a talk I remember seeing by Tracy) Poulin, Serpico, Lesgourgues 1508.01370 & JCAP

Why is it plausible that DM annihilation is strongly constrained by CMB?

• binding energy of hydrogen = 13.6 eV
• single annihilation of mDM ~ O(MeV) can

ionize 105 H atoms

Energy Injection During CMB

(shamelessly stolen from a talk I remember seeing by Tracy) Poulin, Serpico, Lesgourgues 1508.01370 & JCAP

Energy Injection Rate

Energy injection rate ~ E × nDM2 × ⟨σv⟩ann

• E = 2 mDM, nDM = ρDM/mDM
• define pann, which scales out density and includes

efficiency factor: pann ≡ feff ⟨σv⟩ann/mDM

Energy Injection Rate

Energy injection rate ~ E × nDM2 × ⟨σv⟩ann

• E = 2 mDM, nDM = ρDM/mDM
• define pann, which scales out density and includes

efficiency factor: pann ≡ feff ⟨σv⟩ann/mDM

Green, Meerburg, Meyers 1804.01055 & JCAP

Energy Injection Rate

Energy injection rate ~ E × nDM2 × ⟨σv⟩ann

• E = 2 mDM, nDM = ρDM/mDM
• define pann, which scales out density and includes

efficiency factor: pann ≡ feff ⟨σv⟩ann/mDM

Green, Meerburg, Meyers 1804.01055 & JCAP

Energy Injection Rate

Energy injection rate ~ E × nDM2 × ⟨σv⟩ann

• E = 2 mDM, nDM = ρDM/mDM
• define pann, which scales out density and includes

efficiency factor: pann ≡ feff ⟨σv⟩ann/mDM

Green, Meerburg, Meyers 1804.01055 & JCAP

Note: similar lines of evidence from 21cm signal (Liu & Slatyer 1803.09739 & PRD;…), from behavior of star forming halos (Schön, et al. 1411.3783 & MNRAS), and other probes

Constraints

Planck 2015: pann < 4.1 × 10-28cm3/s/GeV Planck 2018: pann < 3.2 × 10-28cm3/s/GeV

Madhavacheril, Sehgal, Slatyer 1310.3815 & PRD

Constraints

Planck 2015: pann < 4.1 × 10-28cm3/s/GeV Planck 2018: pann < 3.2 × 10-28cm3/s/GeV

Madhavacheril, Sehgal, Slatyer 1310.3815 & PRD

10 100 1000 10000 DM mass (GeV) 10-27 10-26 10-25 10-24 10-23 10-22 <σ v> (cm3/s) 10-6 10-5 10-4 10-3 10-2 10-1 100 DM mass (GeV) 10-33 10-32 10-31 10-30 10-29 10-28 10-27 10-26 <σ v> (cm3/s)

Annihilation channels:

Slatyer, 1506.03811 & PRD

Constraints

Outline

• Energy Injection During CMB
• P-wave annihilation
• Hidden Sector
• Asymmetric Dark Matter
• Freeze-in

What is p-wave annihilation?

• In generality ⟨σv⟩ann = ∑l cl × (σv)ann,l × vDM2l

What is p-wave annihilation?

• In generality ⟨σv⟩ann = ∑l cl × (σv)ann,l × vDM2l
• If there is a nonzero l = 0 term (e.g., c0≠0), the

annihilation is called s-wave

What is p-wave annihilation?

• In generality ⟨σv⟩ann = ∑l cl × (σv)ann,l × vDM2l
• If there is a nonzero l = 0 term (e.g., c0≠0), the

annihilation is called s-wave

• If the first nonzero term is l =1 (e.g., c0=0 & c1≠0),

the annihilation is called p-wave

What is p-wave annihilation?

• In generality ⟨σv⟩ann = ∑l cl × (σv)ann,l × vDM2l
• If there is a nonzero l = 0 term (e.g., c0≠0), the

annihilation is called s-wave

• If the first nonzero term is l =1 (e.g., c0=0 & c1≠0),

the annihilation is called p-wave (σv)ann,p(vb)/(σv)ann,p(va) = (vb/va)2

Why is this of interest?

• vCMB ~ (TCMB/mDM)1/2 ~ 10-3 (mDM/MeV)-1/2 is much

smaller than vfo ~ 1/5

Why is this of interest?

• vCMB ~ (TCMB/mDM)1/2 ~ 10-3 (mDM/MeV)-1/2 is much

smaller than vfo ~ 1/5

• thus, annihilations during (and after) CMB are

suppressed by (vCMB/vfo)2 ~ 10-5/(mDM/MeV)

Why is this of interest?

• vCMB ~ (TCMB/mDM)1/2 ~ 10-3 (mDM/MeV)-1/2 is much

smaller than vfo ~ 1/5

• thus, annihilations during (and after) CMB are

suppressed by (vCMB/vfo)2 ~ 10-5/(mDM/MeV) ⟹ CMB bounds potentially lifted

How unusual is p-wave annihilation?

How unusual is p-wave annihilation?

L = gχ ¯ χχφ + gf ¯ ffφ + 1 2m2

φφ2

How unusual is p-wave annihilation?

L = gχ ¯ χχφ + gf ¯ ffφ + 1 2m2

φφ2

χ χ _ f f _ φ

How unusual is p-wave annihilation?

for s=4mχ2(1+v2/4+…) and mφ ≫ mχ ≫ mf σv = g2

χg2 fm2 χv2

8πm4

φ

1 − m2

f

m2

χ

!3/2

L = gχ ¯ χχφ + gf ¯ ffφ + 1 2m2

φφ2

χ χ _ f f _ φ

What about hidden sector?

L = gχ ¯ χχφ + 1 2m2

φφ2 + 1

2λφHφ2|H|2

What about hidden sector?

χ χ _ φ φ

L = gχ ¯ χχφ + 1 2m2

φφ2 + 1

2λφHφ2|H|2

φ e+ e- ) (…

What about hidden sector?

for s=4mχ2(1+v2/4+…) and mχ ≫ mφ

χ χ _ φ φ

σv = 3g4

χv2

128πm2

χ

L = gχ ¯ χχφ + 1 2m2

φφ2 + 1

2λφHφ2|H|2

φ e+ e- ) (…

φ e+ e- (… )

What about hidden sector?

for s=4mχ2(1+v2/4+…) and mχ ≫ mφ

χ χ _ φ φ

σv = 3g4

χv2

128πm2

χ

P-wave annihilation is not (just) a theorist’s trick!

L = gχ ¯ χχφ + 1 2m2

φφ2 + 1

2λφHφ2|H|2

Implications for Indirect Bounds

• cm s
• )

s-wave p-wave Decay

106 107 108 109 1010 1011 1012 10-4 10-3 10-2 10-1 1 Dark Matter Mass mχ [eV] Max xe Before Reionisation 0.025% 0.1% 1% 10% 90% 0.025% 0.1% 1% 10% 90%

Liu, Slatyer, Zavala, 1604.02457 & PRD

Implications for Indirect Bounds

s-wave p-wave Decay

104 105 106 107 108 109 1010 1011 1012 10-4 10-3 10-2 10-1 1 Dark Matter Mass mχ [eV] Max xe Before Reionisation 0.025% 0.1% 1% 10% 90% 0.025% 0.1% 1% 10% 90%

Liu, Slatyer, Zavala, 1604.02457 & PRD

Outline

• Energy Injection During CMB
• P-wave annihilation
• Hidden Sector
• Asymmetric Dark Matter
• Freeze-in

Hidden Sector

• We generically find constraints from overproducing

photons at sensitive points in cosmic history

Hidden Sector

• We generically find constraints from overproducing

photons at sensitive points in cosmic history

• What if we annihilate to a hidden sector only?

Hidden Sector

• We generically find constraints from overproducing

photons at sensitive points in cosmic history

• What if we annihilate to a hidden sector only?
• Assume some coupling to SM (after all, this is a

direct detection workshop!)

Hidden Sector

• We generically find constraints from overproducing

photons at sensitive points in cosmic history

• What if we annihilate to a hidden sector only?
• Assume some coupling to SM (after all, this is a

direct detection workshop!) for what couplings are CMB bounds lifted?

Higgs Portal Model

χ χ _ φ φ

L = gχ ¯ χχφ + 1 2m2

φφ2 + 1

2λφHφ2|H|2

φ e+ e- ) (…

τ −1

φ

= sin2θ y2

emφ

8π ' " 1025s ✓ sin θ 10−14 ◆2 mφ MeV #−1

Constraints on Mediator

L = gχ ¯ χχφ + 1 2m2

φφ2 + 1

2λφHφ2|H|2

Liu, Slatyer, Zavala, 1604.02457 & PRD

τ −1

φ

= sin2θ y2

emφ

8π ' " 1025s ✓ sin θ 10−14 ◆2 mφ MeV #−1

Constraints on Mediator

L = gχ ¯ χχφ + 1 2m2

φφ2 + 1

2λφHφ2|H|2

Liu, Slatyer, Zavala, 1604.02457 & PRD See also Slatyer and Wu, 1610.06933 & PRD

τ −1

φ

= sin2θ y2

emφ

8π ' " 1025s ✓ sin θ 10−14 ◆2 mφ MeV #−1

Outline

• Energy Injection During CMB
• P-wave annihilation
• Hidden Sector
• Asymmetric Dark Matter
• Freeze-in

Asymmetric Dark Matter

Nussinov 1985; Barr et al 1990; Barr 1991; D.B. Kaplan 1992; Dodelson et al 1992; Fujii and Yanagida 2002; Kitano and Low 2004; Farrar and Zaharijas 2005; Gudnason et al 2006; Kitano et al 2008; D. E. Kaplan, Luty, Zurek 2009; …

• What if dark matter doesn’t have any antiparticles?

Asymmetric Dark Matter

• What if dark matter doesn’t have any antiparticles?
• Freezeout in the presence of an asymmetry

(Graesser, Shoemaker, Vecchi 1103.2771 & JHEP; Imminiyaz, Drees, and Chen 1104.5548 & JCAP)

n(¯ χ) n(χ)

• t→∞

' exp ✓ 2 hσvifo hσvith ◆

Asymmetric Dark Matter

• What if dark matter doesn’t have any antiparticles?
• Freezeout in the presence of an asymmetry

(Graesser, Shoemaker, Vecchi 1103.2771 & JHEP; Imminiyaz, Drees, and Chen 1104.5548 & JCAP)

⟨σv⟩ann ~ few × ⟨σv⟩th ⟹ “no” antiparticles n(¯ χ) n(χ)

• t→∞

' exp ✓ 2 hσvifo hσvith ◆

CMB Constraint on ADM

Lin, Yu, Zurek, 1111.0293 & PRD

Light ADM

• ADM requires a larger annihilation cross section
• Also need a model for the asymmetry generation

Outline

• Energy Injection During CMB
• P-wave annihilation
• Hidden Sector
• Asymmetric Dark Matter
• Freeze-in

Freeze-In

• What if dark matter was never equilibrated?

Freeze-In

• What if dark matter was never equilibrated?
• example: Dodelson-Widrow sterile ν mechanism —
• scillation of SM ν to sterile (Dodelson & Widrow, hep-ph/9303287 & PRL)

Freeze-In

• What if dark matter was never equilibrated?
• example: Dodelson-Widrow sterile ν mechanism —
• scillation of SM ν to sterile (Dodelson & Widrow, hep-ph/9303287 & PRL)
• generalized by Hall et al. 0911.1120 & JHEP

Freeze-In vs. Freeze-Out

• Freezeout: abundance inversely related to coupling
• Freeze-in: abundance directly related to coupling

Freeze-In vs. Freeze-Out

• Freezeout: no sensitivity to initial conditions (relic

density set entirely by ⟨σv⟩)

• Freeze-in: initial conditions determine the couplings

that are required to get ΩDM

Freeze-In vs. Freeze-Out

• Freezeout: easy to decouple σDD and ⟨σv⟩ (wide

range of targets)

• Freeze-in: generically relate σDD and ⟨σv⟩ (narrow

range of targets)

Freeze-In vs. Freeze-Out

• Freezeout: easy to decouple σDD and ⟨σv⟩ (wide

range of targets)

• Freeze-in: generically relate σDD and ⟨σv⟩ (narrow

range of targets)

Dvorkin, Lin, Schutz, 1902.08623

Conclusions

• CMB constrains s-wave thermal dark matter below

10-20 GeV (depending on final state)

• p-wave annihilation lessens strength of constraint
• Asymmetric dark matter mostly avoids constraint
• Hidden sector freeze-out works, but can be difficult

to make compatible with large DD signal

• Freeze-in is a very natural target we haven’t hit (yet)