sign in sign up
<<
Home , Dark matter, Dark photon

Dark Sectors 2016

! Theory and Visible Dark Primer

Natalia Toro ! Prepared with Rouven Essig & Philip Schuster Welcome!

2 Looking Back In 2009, SLAC held a “Dark Forces” workshop. At that time…

• data spurred renewed theoretical interest

Dark Forces Workshop 2009 • Re-analysis of data from multi- purpose detectors opened window to the dark sector

• First-generation experiments were starting to take shape.

3 Searching in Dumps Now is a greatdiscoverable with new, low- time! power beam dump 0.01 0.1 1 106 seconds @ 100 nA, 6 GeV 0.01 0.01 10-4 Scalar Thermal Relic DM Broad, one signal event per hour KLOE -3 10 LHC HADES KLOE -5 BaBar -4 3 3 10 a,5 APEX 10 10 10 PHENIX Test -5 LEP favored A1 10 * a,±2 NA48/2 E774 -6 10-6 10 ae international 4 4 4 E774 CRESST II 10 10 ) 10-7 BaBar E141(10cm, A' -7

m -8 10 / 10 XENON 10 5 .3 mC) 5 2 -9 ⇥ ⇥

10 10 10 m activity ( -8 E141 -10 E137 10 D 10 E137

-11 Density 6 6 2 10 Relic (200m, 30 C) 10 10 -9 -12 LSND 10 = 10

+CERN, milli- y searching for -13 MegaWatt x Year 10 7 7 -10 x SIDM 10 charge 10 10 10-14 LSND lower limit Orsay U70 10-15 -11 dark sectors! 8 8 10 10-16 10 10 -3 -2 -1 (dump10 experiments10 also constrain10 1 1 10 102 103 0.01 0.1 1 longer-lived decaym modes,A' [GeV ] m (MeV) mA' GeV see Schuster, NT, Yavin to appear) *E774: 20cm, .3 mC

8 Many of the first-generation dedicated experiments are still ongoing. Their reach can be anticipated reliably, and it’s clear what limits they run into. It’s a perfect time to ask: Where do we go next?! That’s the focus of this workshop.

Next 3 talks: assessing the state of the field today, to set stage for these discussions. 4 Where to now?

• What new ideas will propel us past the obstacles found in designing the first-generation experiments?

• What important possibilities are we overlooking, that we should be searching for?

• What opportunities do we have with upcoming facilities and/or new technologies that didn’t exist 7 years ago?

• What do we want this field to look like in 5-10 years? What is within reach? What should we focus on?

5 Outline

• Why Dark Sectors?

• Organizing Questions

• Visibly Decaying Dark Searches – Properties – Search Strategies – Outstanding Challenges

6 Why Dark Sectors?

• Nature seems well described by SU(3)xSU(2)xU(1) – We need to check this assumption!!

• Precedent in for rare/subtle effects to be important (Radioactivity, Rutherford scattering, , parity violation…)

• We already know we are missing something…

7 There is a dark sector!

Baryons Dark Dark

• What is it?

• Where did it come from?

8 Theory Prejudice

Least addition of new ? ! Motivates SM weak multiplet or

[SM itself is far from minimal in this sense] Theory Prejudice

Least addition of new Parallel structure to particles? Standard Model ⇒ ! Motivates SM weak Motivates weak multiplet or axion coupling searches at modest energy

[SM itself is far from minimal in this sense] Looking for Dark Sectors Whatever is in the dark sector, only three sizeable interactions allowed by Standard Model symmetries:

1 Y µ Vector Portal 2 Y Fµ F 0

2 2 Higgs Portal h ⇥ exotic rare Higgs decays h | | | |

Neutrino Portal ⌫ (hL)⇥ not-so-sterile neutrinos

10 New gauge interactions

¯ µ Gauge Portal g qi i¯ iVµ i X A new vector can also couple to -free global symmetries of the Standard Model (or, with appropriate new matter, to anomalous symmetries) e.g. B-L, Lμ-Lτ, … Looking for Dark Sectors Whatever is in the dark sector, only three sizeable interactions allowed by Standard Model symmetries:

focus for this talk – 1 Y µ Vector Portal 2 Y Fµ F 0 big opportunity for low- energy experiments 2 2 Higgs Portal h ⇥ exotic rare Higgs decays h | | | |

Neutrino Portal ⌫ (hL)⇥ not-so-sterile neutrinos

¯ µ Gauge Portal g qi i¯ iVµ sometimes “bycatch” in vector portal searches, but i X some interesting cases 12 require dedicated searches! Sources and Sizes of Kinetic 1 Y µ Mixing 2 Y Fµ F 0 • If absent from fundamental theory, can still be generated by perturbative (or non-perturbative) quantum effects – Simplest case: one heavy ψ with both EM charge & dark charge ψ γ e gD A0

egD m 2 4 generates ✏ 2 log M 10 10 ⇠ 16⇡ ⇤ ⇠ 13 Sources and Sizes of Kinetic 1 F Y F µ Mixing 2 Y µ 0 • If absent from fundamental theory, can still be generated by perturbative (or non-perturbative) quantum effects – In Grand Unified Theory, symmetry forbids tree-level & 1-loop mechanisms. GUT-breaking enters at 2 loops ψ γ X e gD A0

3 5 generating ✏ 10 10 7 ⇠ (→ 10 if both U(1)’s are in unified groups) 14 Effects of Kinetic Mixing

Regardless of where it comes from, kinetic mixing can always be re-interpreted as (mainly) giving matter of electric charge qe an A′ coupling ∝ qεe A′

e + e e µ,⇡,... Dark-sector matter can have an independent coupling gD to A′ Hints of Dark Forces?

• High-energy anomalies • Direct detection anomalies • g-2 • Muonic • 8Be decay resonance • Astrophysical hints of self-interaction? • 511 keV line • keV γ-ray excess • excess

16 Taxonomy of Dark Sectors

A few simple questions about the new …

1. What is the mediator?

Kinetically mixed ? Scalar? Flavor gauge ? SM dipole interaction?

2. Is it lighter or heavier than…

– any dark matter in dark sector? (invisible decays)

– relevant SM particles? 2 me (visible decays) – new unstable particles? (“rich” multi-body decays)

17 Taxonomy of Dark Sectors

If dark matter is relevant…

3. How is it produced?

– Thermally (through to SM)?

– Asymmetric?

– Other?

May imply constraints/relations among dark-sector couplings

4. or scalar?

5. Elastic (-diagonal) or inelastic (mass-off-diagonal) interactions?

Affect signals in various experiments and the comparisons among them

18 Taxonomy of Dark Sectors

Kinetically mixed dark photon decays to SM Not produced on-shell * can still mediate dark sector production heavy

Visible decays – 2 width Γ~αmAʹϵ 2 me

Dark photon mass Long-lived Aʹ➝3γ

0 Dark sector millicharges Taxonomy of Dark Sectors

Kinetically mixed dark photon decays to SM Not produced on-shell * can still mediate dark sector production heavy

Visible decays – Visible & missing 2 width Γ~αmAʹϵ mass searches 2 me Cosmological & stellar

Dark photon mass Long-lived Aʹ➝3γ bounds, solar production, light thru walls 0 Dark sector millicharges Taxonomy of Dark Sectors

Kinetically mixed dark photon coupled to DM Not produced on-shell * can still mediate DM production heavy

⇒ invisible signal Aʹ DM Visible decays – m Decays mainlyD to 2 ʹ = 2 m width Γ~αmAʹϵ DM anti-DM pair, mA 2 me width Γ~ α + subleasing visible decay Long-

Dark photon mass lived Aʹ➝3γ 0 Dark matter mass Taxonomy of Dark Sectors

DM Signals at colliders

Contact operator

heavy DM production

Resonant DM DM

production ʹ = 2 m mA 2 me Near-threshold DM production

Dark photon mass (off-shell light mediator) 0 Dark matter mass Sub-MeV Dark Photons

[Figure from 2013 Intensity Frontier report – Javier Redondo] Sub-MeV Dark Photons

Light thru walls

Cavity searches

Solar+scatter

4 4 frequency Hz kHz MHz GHz !4 1 inside the detector. Given the significant enhancement 10 Jupiter Earth 0.1 in the low-energy part of the solar dark photon spec- Coulomb ALPS CROWS CROWS 10[Figure-2 from 2013Coulomb Intensity Frontiertrum, report Fig. 1, a detector– Javier with a Redondo] low threshold energy of !6 limit 10 CAST 10-3 O(100) eV will have a clear advantage. To date, the only using published 10-4 longitudinal work that considers limits on dark photons from direct !8 -5 10 10 limit DM detection is by HPGe collaboration, Ref. [9]. How- Sun mode 10-6 CMB ever, it used incomplete calculations of the solar flux, and HB

this Κ ∂ Future HB ! XENON10 10-7 work as we will show in the following, the low-energy ioniza- 10 10 high CoGeNT - Sun 10 8 -Q H tion signals by the XENON10 [10] and CoGeNT [11, 12] microwave -9 collaborations yield far more stringent limits. 10 L ALPS II !12 cavity 10 10-10 The XENON10 collaboration has published a study experiment 10-11 on low-energy events in [10]. With 12.1 eV !14 10-12 ionization energy, the absorption of a dark photon with 10 10-13 300 eV energy can produce about 25 . To get a -16 -15 -14 -13 -12 -11 -10 -9 -8 -7 -6 -5 -4 -3 ! ! ! 10 10 10 10 10 10 10 10 10 10 10 10 10 10 conservative constraint we count all the ionization events 10 6 10 4 10 2 100 102 104 mg' eV within 20 keV nuclear recoil equivalent in Ref. [10], which mV !eV" FIG. 2: The reach of cavity searches for hidden photons, taking advantage of the improved transmissioncorresponds of the to a signal of about 80 electrons. The total @ D longitudinal modes. The blue region is the published limit from the CROWS experiment [29],number calculated of using events is 246, which indicates a 90% C.L up- FIG. 2: Constraints on κ as functions of mV .Thesolid, the results of [25]. The solid blue line shows the bound we obtain by reanalyzing the CROWSper results, limit taking on the detecting rate to be r<19.3events dashed, dot-dashed and dotted curves show constraints from into account longitudinal hidden-photon modes. The dashed blue line shows the reach of a realistic− future1 − cavity1 experiment. See section VI for more details. The gray shaded regions show various preexisting astrophysicalkg day and(similar to limits deduced in Ref. [13]). In the energy loss of the Sun by requiring that the dark photon laboratory constraints, compiled from Refs. [9, 34], while the gray dashed line shows the projectedthe reach region of the12.1eV< ω < 300 eV the ionization process luminosty does not exceed 10% of the standard solar lumi- ALPS-II experiment [39]. dominates the absorption, and therefore Br in this region nosity [16], energy loss in horizontal branch (HB) , the XENON10 experiment and the CoGeNT experiment, respec- can be set to unity. The 90% C.L. upper limit on κ as a tively. The thick curves are for the SC, whereas the thin the kinetic terms of (1), we obtain function of mV is shown by the dot-dashed black curve curves are for the HC with e′ =0.1. For comparison, the cur- in Fig. 2, where we can see that it gives the most strin- rent bound (gray shading) from the LSW-type experiments 1 µ⌫ µ⌫ 1 2 µ µ = F F + F 0 F 0 + m A0 A0 eJ A + " A0 gent constraint(2) in the SC. To arrive at these limits we are shown (see Ref. [17] for details). The conservative con- µ⌫ µ⌫ 0 µ EM µ µ L 4 2 reconstruct ε for Xe from the real and imaginary parts straint from the CAST experiment [18] by considering the r where Aµ and Aµ0 are of course linear combinations of aµ and aµ0 . The propagating massof eigenstates the refractive are index of Xe. The imaginary part can contributions from only the transverse modes [4] is also shown the transverse photon and hidden-photon modes A and A0 , along with the longitudinal hidden-photon in green shading.The orange shaded region is excluded from | T i | T i be extracted from the total photoabsorption cross sec- mode A0 (there is no longitudinal mode for the massless photon). In this basis, the linear combination tests of the inverse square law of the Coulomb interaction [19]. | Li tion [14, 15], and the real part can be calculated from the AT + " AT0 directly interacts with charges, while the linear combination AT0 " AT is sterile. | Leti us| firsti focus on the transverse modes. In a “Light-Shining-Through-A-Wall”| i | experiment,i imaginary the linear part using the Kramers-Kronig dispersion rela- combination A +" A0 is first produced. But, this is not a mass eigenstate - the two statestions.A and TheA improvement0 over other experimental probes κ e m | T i | T i | T i | T i TABLE I: Sensitivities to and eff in the small V region. have di↵erent leading to a di↵erential phase developing between them. These states areis relativistic quite significant, and considering that the signal scales as 1 2 i 24(m /!)L hence after traveling a distance L (where the wall is placed), the state evolves to AT + " eκ . We0 alsoAT0 collate main constraints in Table I. | i | i Model param. Sun HB XENON10 CoGeNT (up to an overall irrelevant phase), where ! is the energy of the produced state. Notice that this is similar m − − − − The published data from the CoGeNT DM experiment SC, κ × V 4 × 10 12 4 × 10 11 3 × 10 12 8 × 10 11 to the phenomenon of neutrino oscillations. The linear combination AT + " AT0 is absorbed by the wall, eV | i | i have a threshold of about 450 eV [12]. In this region, the −14 −15 −13 −13 while the sterile component AT0 " AT passes through the wall. The amplitude of this sterile component HC, eeff 3 × 10 8 × 10 1 × 10 4 × 10 is proportional to "m2 L/!.| Whilei | thisi sterile component travels through the wall, it cannotdark directly photon be flux from the Sun drops almost exponen- 0 detected by the instruments on the other side of the wall. For detection, we need the sterile statetiallyA0 with" A energy, whereas the observed spectrum in Co- | T i | T i to partially oscillate back into the state AT + " AT0 . This will happen because the sterileGeNT state is is relatively also flat. Therefore, in order to optimize not a mass eigenstate. The component of| thei state| thati overlaps with the interaction state A + " A theT sensitivity,T0 we only use the event counts in the inter- the associated line is shown as the thin dotted curve in after traveling a further distance L is again proportional to "m2 L/!. Hence, the signal in| thisi setup| isi 0 val 450 500 eV and the 90% upper limit on the back- Fig. 2, and included in Table I as limit on e .Inboth, proportional to "2m4 L2/!2. For an apparatus of fixed physical size L, this implies that the sensitivity− for eff 0 ground subtracted rate is r<0.6eventskg−1day−1.The the HC and the SC, CoGeNT does not have sensitivity low mass hidden photons (m0 1/L) drops sharply. With this scaling the unconstrained parameter space that can be probed by these experiments⌧ is limited. resulting sensitivity is shown as the thick dotted purple to constrain dark photons since the required flux is not Naively, one may think that this discussion should also apply to the longitudinal modecurve of the in hidden Fig. 2, which is far weaker than the constraint supported by the Sun. photon. There is however a very important di↵erence - the massless photon Aµ does not havefrom a longitudinal the energy loss of the Sun. Conclusions We point out that the unprecedented mode. Since the hidden photon is coupled to electromagnetic currents (see Eq. 2), these currents will source In the HC, Nexp in Eq. (21) must include a contribution sensitivity of some of the DM experiments to ionization 2 2 from (20), and it dominates in the region mV ω ∆εr , allows to turn them into the most sensitive dark photon 2 2 ≪ | | but is subdominant if mV ω ∆εr . Since the flux of helioscopes. By directly calculating the ionization signal, V (h′) in the HC is mainly∼ contributed| | from conversion we show that the ensuing constraint from the XENON10 of transverse photons in the Sun, the spectral distribu- experiment significantly surpasses any other bounds on tion reflects the solar temperature, Fig. 1, with the cutoff dark photons, including very tight stellar energy loss con- −5 above 1 keV. A dark photon of 1 keV energy can at most straints in the mV -interval from 10 to 100 eV. In the produce about 60 electrons in . For e′ =0.1, case of “mini-charged” particles (equivalent to the Hig- the 90% C.L. upper limit is shown as the thin dot-dashed gsed version of dark photons), we also derive a strin- −13 curve in Fig. 2. For the sensitivity from CoGeNT we take gent bound, eeff < 10 , which is second only to the into account all the events from 450 eV to 1 keV and constraint from the energy loss of the horizontal branch Taxonomy of Dark Sectors

Kinetically mixed dark photon coupled to DM Not produced on-shell * can still mediate DM production heavy

⇒ invisible signal Aʹ DM Visible decays – m Decays mainlyD to 2 ʹ = 2 m width Γ~αmAʹϵ DM anti-DM pair, mA 2 me width Γ~ α + subleasing visible decay Long-

Dark photon mass lived Aʹ➝3γ 0 Dark matter mass Heavy photon production & decays in a fixed target experiment electron beam-fixed target is analogous to bremsstrahlung:

•prefers x~1Decay (i.e.Heavy EA’ = Ephotonbeam Modes) lifetime & Lifetimes •small angle emission dominates Aʹ→Standard Model 10-4 BaBar WASA KLOE -5 10 am, 5 s

am,±2 s favored Aʹ decays back to charged SM -6 10 ae with BFs taken from E774 Prompt 10µm + − + − + − lower ε, lower mass 10-7 Displaced

R(e e →/e e →µ µ ) 2 e -8 ! → longer lifetime 10 E141 1mm caveat: if there is a dark sector particle 10-9 lighter than Aʹ, dominant decay will be …this is why beam dump experiments are so effective at low mass/coupling. 10-10 1cm Orsay ! U70 invisible (I think we’ll hear more about Ebeam=2.2GeV Hard to get theassumes 10µm-1cm Aʹ can’t regime… decay to 10-11 these scenarios) 10-3 10-2 10-1 1 particles Far from Z-pole, same mA' GeV New Perspectives in Dark Matter breakdownMathew Graham,as e+e– SLAC 8 annihilation modes H L New Perspectives in Dark Matter 26 Mathew Graham, SLAC 9 Visible Dark Photons

• Decay Modes and Lifetime

• Search Strategies!

• Outstanding Challenges

27 Dark Photon Searches

Production Modes Detection Signatures

• Electron- annihilation • Pair resonance

• Decays • Beam-dump late decay

• Drell-Yan (collider or fixed • Inclusive missing mass target) • Reconstructed displaced • Bremsstrahlung vertex Dark Photon Searches

Production Modes

• Electron-positron annihilation ~constant yield up • Meson Decays to threshold

• Drell-Yan (collider or fixed target) Yield falls steeply with mass • Bremsstrahlung Production: e– bremsstrahlung

Distinctive kinematics: for me≪mAʹ≪Ebeam, Aʹ carries most of beam energy & very forward-peaked

3/2 mA l+ E (narrow) ⇥ e A mA 1/2 E mA l E ⇥ e(wide) EAʹ≃Ebeam-mAʹ

Ee-≃mAʹ Dark Photon Search: Current Constraints

10-4 KLOE Red/green: e,μ -5 HADES KLOE BaBar 10 a,5 APEX PHENIX Test anomalous dipole A1 a,±2 favored -6 NA48/2 10 a moments E774 e

10-7 2

All other colors: Pair -8 10 E141 resonance searches 10-9

10-10 Gray: Beam Dump Orsay U70 10-11 10-3 10-2 10-1 1

mA' [GeV] Two HPS searches: Bump-hunt and Vertexing

• two types of searches → two kinematic fits →two mass New Dark Photon Searches:!+ resolutions Large coupling search • Large coupling Aʹs decay in the target → constrain the + − − e & e to originate from beamspot Visible Vertex Searches! •very good constraint on angles •Small coupling Aʹs decay outside of target → point 40µm !+ decay products back to target (vertical) •good at removing poorly reconstructed tracks − Look for Aʹ decay displaced ! from target by reconstructing Small coupling search σm(NC) ~ 2-3 MeV What an HPS search looks like:final-state Vertexing region tracks σm(BSC) ~ 0.7-2.7 MeV

(after mass cut) 4000 bkg events (after vertex cut) (50-100MeV) 500 A’ at 80MeV Small Coupling 50 A’ at 80MeV −8 + 10M bkg α~5×10 ! −8 α~5×10 events toy MC for example only... does not reflect reality − ! Large Coupling

− !

not included yet…recoil electron! 2D searchEnables in mass & vertex substantial position (z) background → small coupling region (α~10−8 − 10−10) ⟹adds mass resolution/BH discrimination ! reduction ➝ sensitivity to lower vertex resolution is the key here! New Perspectives in Dark Matter Mathew Graham, SLAC 18 ϵmass (assuming resolution is secondary no other decays that broaden Aʹ resonance) New Perspectives in Dark Matter Mathew Graham, SLAC 20 New Dark Photon Searches: Fixed Target Missing Mass Searches

New technique, being pursued by several proposed experiments: γ “Easier” missing mass: e+ e– beam on atomic e–: measure γ [e.g. MMAPS, VEPP-3] e+ Aʹ – “Harder” missing mass: e Aʹ beam on H, reconstruct e– & – – [e.g. DarkLight, in conjunction e e with visible resonance or veto p p to reject background] Ongoing Efforts

Broad worldwide effort to search for dark forces! (this is the list of ongoing searches for visible dark photons)

• DarkLight • VEPP-3 • APEX • MESA • HPS • Future MAMI-A1 • CMS & ATLAS • SHiP jets & exotic Higgs • SeaQuest decays • KLOE-II • LHCb • NA64 • Belle II • ???? • MMAPS Ongoing Efforts CMS High-energy High intensity colliders colliders PHENIX

KLOE at DAΦNE

ATLAS

APEX Fixed MAMI A1 Target The NA48/2 detector JLab CEBAF 2003–2008: charged kaonMiniBooNE beams, the NA48 detector HPS WASA@COSY Narrow band K beams:

PK= 60 (74) GeV/c, PK/PK ~ 1% (rms).

Maximum K decay rate ~100 kHz; NA48/2: six months in 200304; NA62-R : four months in 2007. K NA48/2 Principal subdetectors:

Magnetic spectrometer (4 DCHs) 4 views/DCH: redundancy efficiency; p/p = 0.48% ⨁ 0.009%p [GeV/c] (in 2007)

Scintillator hodoscope (HOD) Fast trigger, time measurement (150ps).

Liquid Krypton EM calorimeter (LKr) High granularity, quasi-homogeneous; Vacuum beam pipe 1/2 E/E = 3.2%/E ⨁ 9%/E ⨁ 0.42% [GeV]; = =4.2mm/E1/2 ⨁ 0.6mm (1.5mm@10GeV). x y 4 E. Goudzovski / Messina, 25 September 2014 Beam What Next? • Opportunities – Are these lists missing something? Can we combine them in new and exciting ways?

!Production Modes Detection Signatures • Electron-positron annihilation • Pair resonance ! • Meson Decays • Beam-dump late decay • Drell-Yan (collider or fixed target) • Inclusive missing mass ! • Bremsstrahlung • Reconstructed displaced vertex • Challenges – What do present experiments teach us about obstacles we’ll have to confront to go further? – What milestones do we really want to hit? 36 The Limit to Bump Hunts?

This is what a signal looked Detection requires excellent like in APEX test run statistics & mass resolution and exquisite bkg model! ! theory & exp uncertainties ➝ most analyses parametrize bg ! %-level σm Not unique challenge ~106 events (cf h➝γ γ), but future dark photon searches push it even further. ! What’s the ultimate limit? Closing “Mont’s” Vertexing Gap Expected HPS reach

A' Æ Standard Model 2015 Running (Yellow) 10-4 ! 1 week with 50nA @ 1.1 GeV am, 5 s 1 week with 200nA @ 2.2 GeV -5 Closing from above (bump 10 KLOE 2 weeks with 300nA @ 4.4 GeV favored am,±2 s MAMI ! 10-6 BaBar hunt)! runs into systematics E774 APEX MAMI 2017 Running (Blue): Test Runs ae ! 10-7 ê 2 weeks with 200nA @ 2.2 GeV 2 weeks with 300nA @ 4.4 GeV 2 e 3 weeks with 450nA @ 6.6 GeV -8 10 E141 Closingtime given is beam timefrom=floor time/2 below requires Orsay HPS 10-9 great vertex resolution in 10-10 e.g. from HPS…U70 high luminosity environment 10-11 10-3 10-2 10-1 1 1/5 lifetime m GeV A' 1/50 the energy

H L orders of magnitude lower S/B New Perspectives in Dark Matter Mathew Graham, SLAC 23 Extending to Higher Masses

How to reach lower ϵ at e– brem A' yield arb. units

O(GeV) masses? L 105

units H L . arb

• H Yield… 1000 yield ' A – Fixed-target yields falling 10

brem 11.1 GeV –

precipitously (form factors, e 4.4 GeV 0.1 pdfs, …) 0.01 0.02 0.05 0.10 0.20 0.50 1.00 2.00 A' Mass GeV Drell-Yan (120 GeV p on target) – Still mostly “low-pT” for LHC H L – Is Belle II best we can do?

• Hadronic channels start to ϵ =0.01 dominate branching ratios plot by Stefania Gori 39 Conclusion

• Exploring Dark Sectors is an important and growing element of physics beyond the Standard Model – Requires searching for several complementary signatures (more in subsequent primer talks) – A wealth of exciting ongoing experiments (more in WGs) – Important hurdles to surpass for the next generation of experiments (more in WGs)

• Workshop aims to stimulate thinking about the next steps. Thanks for coming!

40