Detec%ng Gravita%onal Waves from Cosmological Phase Transi%ons. Jonathan Kozaczuk (UMass Amherst) KICP, 10/14/16

Transcription

1 Detec%ng Gravita%onal Waves from Cosmological Phase Transi%ons Jonathan Kozaczuk (UMass Amherst) KICP, 10/14/16

2 Preface Primary focus of this talk: gravita'onal waves from the electroweak phase transi'on (I ll comment on some other PT scenarios not related to electroweak symmetry breaking as well) Most of this talk based on work with members of LISA Cosmology Working Group: Caprini, JK et al, JCAP 1604 (016) no.04, 001 Kozaczuk

3 Mo%va%on The standard picture of electroweak symmetry breaking: V e ( ) At high temperatures, a background field with electroweak quantum numbers stabilized at the origin Kozaczuk 3

4 Mo%va%on The standard picture of electroweak symmetry breaking: V e ( ) At zero temperature, the background field is stabilized away from the origin and EW symmetry is broken Kozaczuk 4

5 Mo%va%on What happened in between? V e ( )?? First or second order transi%on? At what temperature? Other fields involved? Kozaczuk 5

6 Mo%va%on The SM features a smooth cross-over Kajan%e et al, hep-ph/ V e ( ) Kozaczuk 6

7 Mo%va%on Scenarios beyond the SM can instead accommodate a first order EWPT V Φ Φ Kozaczuk 7

8 Mo%va%on Understanding the electroweak phase transi%on is important! First order PT can allow for successful electroweak baryogenesis Kozaczuk 8

9 Mo%va%on Understanding the electroweak phase transi%on is important! First order PT can allow for successful electroweak baryogenesis V Φ h i =0 h i6=0 h i6=0 h i6=0 Φ h i6=0 Kozaczuk 9

10 Mo%va%on Understanding the electroweak phase transi%on is important! First order PT can allow for successful electroweak baryogenesis f f h i6=0 = h i6=0 Kozaczuk 10

11 Mo%va%on Understanding the electroweak phase transi%on is important! First order PT can allow for successful electroweak baryogenesis f, f B/ h i6=0 Non-local EWB Kozaczuk 11

12 Mo%va%on Understanding the electroweak phase transi%on is important! First order PT can allow for successful electroweak baryogenesis f, f B / =0 /B 6= 0 B/ Non-local EWB h i6=0 B 6= 0 Kozaczuk 1

13 Mo%va%on Understanding the electroweak phase transi%on is important! First order PT can allow for successful electroweak baryogenesis Can affect cosmology (e.g. the abundance of thermal relics) See e.g. Profumo + Wainwright, Kozaczuk 13

14 Mo%va%on Understanding the electroweak phase transi%on is important! First order PT can allow for successful electroweak baryogenesis Can affect cosmology (e.g. the abundance of thermal relics) See e.g. Profumo + Wainwright, We should understand where we are on the EW phase diagram Kozaczuk 14

15 Mo%va%on How might we test for a strong first-order (electroweak) phase transi%on in our past? Two arenas: gravita'onal wave astronomy and colliders (This talk) (Ask me later!) Kozaczuk 15

16 Gravita%onal Waves from Strong PTs A first-order cosmological phase transi%on can source a stochas%c gravita%onal wave background in different ways: Bubble Collisions Sound Waves h i6=0 h i6=0 Turbulence Kozaczuk 16

17 Gravita%onal Waves from Strong PTs Various contribu%ons to the gravita%onal wave spectrum depend on the phase transi%on proper%es Bubble Collisions: h GW ' h + h sw + h turb Efficiency factor Bubble wall velocity 1 h env (f) = H apple vw 3 S 1+ g vw env (f) Dura%on of the PT ds ' dt t=t Latent heat Frequency dependence Kozaczuk 17

18 Gravita%onal Waves from Strong PTs Various contribu%ons to the gravita%onal wave spectrum depend on the phase transi%on proper%es Bubble Collisions: Sound Waves: Turbulence: h GW ' h + h sw + h turb h env (f) = H apple 1+ h sw (f) = H applev 1+ h turb (f) = H appleturb g v 3 w g v w 1 3 vw S sw (f), 1/3 100 v w S turb(f) g S env (f) Kozaczuk 18

19 Gravita%onal Waves from Strong PTs Which contribu%ons dominate depends on the PT dynamics Three general scenarios: 1. Non-runaway bubbles h GW ' h sw + h turb. Runaway bubbles in plasma =0 =0 Friction V T =0 h GW ' h + h sw + h turb 3. Runaway bubbles in vacuum h GW ' h, v w When considering specific BSM scenarios, it s important to know which applies! Kozaczuk 19

20 Gravita%onal Waves from Strong PTs How should we look for this signal? Christopher Moore, Robert Cole and Christopher Berry hmp://rhcole.com/apps/gwplomer/ EWPT Characteris%c frequency set by H. For EWPT, f peak ~ 10-3 Hz so LISA is the way to go Kozaczuk 0

21 Gravita%onal Waves from a Strong EWPT LISA (Laser Interferometer Space Antenna), formerly known as elisa hmp://lisa.jpl.nasa.gov/gallery/lisa-waves.html Leading candidate for the ESA s Cosmic Vision Program L3 experiment Proposed launch in 034 LISA Pathfinder launched last year, demonstra%ng feasibility of LISA technology Kozaczuk 1

22 Gravita%onal Waves from a Strong EWPT How well can LISA observe a stochas%c background origina%ng from cosmological phase transi%ons? Study commissioned by the Gravita%onal Observatory Advisory Team (GOAT): Caprini, JK et al, JCAP 1604 (016) no.04, 001 Answer depends on the experimental configura%on and mission %meline h ΩGW(f) f[hz] C1 C C3 C4 Name C1 C C3 C4 Full name NA5M5L6 NA1M5L6 NAM5L4 N1A1ML4 #links Arm length [km] 5M 1M M 1M Duration [years] Noise level N N N N1 Preliminarily confirmed by LISA Pathfinder Kozaczuk

23 LISA Sensi%vity to Stochas%c Background Primary astrophysical foreground expected to be unresolved galac%c white dwarf binaries Dis%nguish between primordial and unresolved background using spectral differences and annual modula%on Adams + Cornish, 013 Extra-galac%c sources can also be isolated from anisotropies in the foreground Kozaczuk 3

24 Gravita%onal Waves from a Strong EWPT Prospects for LISA are model-dependent. Strategy: consider several representa%ve BSM scenarios New physics can yield a strong 1 st order PT through several different mechanisms: V 1 loop e = V 0 + V1 T =0 + V1 thermal New bosonic DOFs can contribute a sizeable thermal cubic term Loop effects can reduce the energy difference between vacua BSM physics can give rise to new terms in the scalar poten%al New physics can also produce a first-order PT via strong dynamics Kozaczuk 4

25 X W VF = = HuT "Hd + S + S + + S (Hu Hu + Hd Hd ) (13) i bounce minimizes the four-dimensional Euclidean action at T ), it should be noted that this n i model can allow for a metastable electroweak-symmetric phase to persist to zero temperature. The D-terms are the same as MSSM This suggests that very strongly supercooled transitions occurring in vacuum (Case 3) may 1X a 1 1 a be notu Hbeen previously analyzed, but gathe ( i T resulting Hdthis + case + g )(H H (14) VD possible. = g Huin (g1 have i )( j T GW j ) =signals u d Hd ). i,j 8 would be worthwhile to consider in future work. Gravita%onal Waves from a Strong EWPT The soft-breaking terms corresponding to the super potential is New scalars charged underathe SM B gauge group are also phenomenologically and theoret- well-motivated. Whentransform transforming nontrivially under the electroweak gauge group, Representa%veically scenarios particles are T H [GeV] 6 +trivially V = massumed H to + m63 H H m S S under the new Z symmetry. The most general soft H u u u Hd d d S such scalars can participate in electroweak symmetry breaking,aswith a potentially significant renormalizable scalar in 1 case3 can be written potential this T 3 + electroweak ( A S + b0 )H A S + m7 S + mof + h.c.. (15)of a twod + simplest u "H 9 S this impact on the PT. The realization scenario is that 3 /H = µ (H + the (H SM H) Higgs + asector (H H)S + b S by + a bsecond Higgs Higgs-doublet-model portal: V (H, S) (HDM), in H) which is enlarged scalar(30) 4S. /T Higgs Then we can get the scalar potential of the most general NMSSM: doublet. These scenarios can accommodate electroweak baryogenesis [64] and can result in Dilaton-like Potentials and Naturally Supercooled Transitions Here, theelectroweak CP-even neutral component H is identified with the 15 GeV Standard Model Table 5: Characteristics of the PT in the SM plus a of dimension-6 e ective operator forhaving the generation of GWs at the electroweak PT (with preliminary studies been carried V = The VF +discrete VD + Vsoft NMSSM: like Higgs. symmetry ensures that S is stable. The singlet can thus contribute two benchmark points taken broken from Ref. [36]: 600 GeV 1 (A) symmetry and 576 GeV a(b), see Section Models with a spontaneously (approximate) conformal feature pseudoout in [65]). The scalar potential is T = with Hudark "H + S + Sabundance. + this + field S Note (H that Hu +SHcan ) the observed relic also be a component of a scalar d Hd Nambu-Goldstone bosontoassociated thed matter broken symmetry; the h isu known ias other 1 µ(e.g. charged symmetries hidden group). VV(H = Hdµ H µ H Hd sector Hd)+ gauge h.c. H1 One + should H 4also bear in 1), H 1 (g+ a dilaton. TheHDM: scalar potential (under ofg)the be by a + scale + 1 H + field, 1H u Hudilaton 1 +,gcan )(H u Hparametrized 8 model parameter space without the mind by that this scale-dependent model resides in a subspace of a larger invariant function modulated weakly function: 5 + m m H H d + +inmthe S SH + approach, for instance by adding dimension-6 operators Higgs potential allowing for. Hu H u Hu 3+ H H S d + H H H H + h.c. d 1 Z limit 4 1opens 1 parameter Going beyond Z symmetry. the upadditional space for a(31) strong 11 a negative quartic coupling T 3 3 V ([35, )+ =70]:4( AP S ( + ) b0where (33) (16) )Hu "Hd + A S + m7 S + m9 S + h.c. first-order PT. After electroweak symmetry breaking,3the presence of the two doublets H, H yields three 1 6 This Higgs portal model can give rise to a strong electroweak PT in primarily two 4 ± A particularly interesting and well-motivated of whenh:the quadraticscalar Vstates ( ) =in µ addition class to scenarios + 15.arisesHiggs (3)H and two neutral Dim-6 EFT: new physical the GeV a charged ways. If the by parameter can be destabilized from the origin at finite < term for the Higgs field controlled the VEVbof the0,dilaton : 5the singlet statesis H 0, A0. This model illustratestemperature the typical (along correlations small values of HS /H term and then provides a cubic the Sexpected directionbetween in field space). The H In this class of scenarios, a strong electroweak PT is driven by a decrease in the free-energy Dilaton: ( /H and ) hence (34) (,the) = V of ( ) +and large mentioned earlier. were computed in the (m, ) plane term VtoContours e ective potential, can contribute to a barrier along the direction h 4 di erence between the electroweak-symmetric local maximum and electroweak-broken phase for the one-loop finite temperature e ective potential associated with Eq.=(3) in is another example connecting the hsi potential 6= 0 and electroweak vacua (in which hsi 0). This where iscomplete a constant. is precisely 5D Randallat TIn =particular, 0, w.r.t. this the SM prediction. The that e ectofofthe this decrease on the strength of PT is Ref. [35]. The [73], region to sectors a sizable GW the electroweak ofcorresponding a two-step transition. Alternatively, if from b >to 0, problem the singlet willpt beisspecula%ve) stabilized at the origin Also consider strongly coupled not related EWSB Sundrum models which providedark an elegant solution to signal the hierarchy of(more the 1 We assume for simplicity CP conservation, as well as a Z symmetry (softly broken by µ in Eq. (31)) confined to values of at below 1 TeV. TheNevertheless, tension between a low cuto and thee ects LHC can lower the SM-like all temperatures. largesuch zero-temperature loop SM. for phenomenological reasons, namely the absence of flavour-changing neutral currents in the Higgs sector. bounds remains to beis investigated. Nevertheless, ther offrom GW(Tthe signal, consider Higgs quartic coupling to increase the on value the bubble. Finite-temperature Assuming the Higgs localized on the IR brane at focusing a distance y= UV we brane n ) inside These scenarios also span the different possibili%es for bubble expansion (nontwo benchmark points where is around 600 GeV [36],is(alongside both in the relativistic, non-runaway 30the e ects can also play a role SM gauge, Higgs, and Goldstone bosons) in (localized at y = 0), the loop 4D e ective action for the Higgs runaway, runaway in plasma, runaway inbetween vacuum) case and in the runaway with finite (the predicted wall velocity andremains to be studied contributing to acase barrier the origin the electroweak vacuum in this case. vp k r µ 4k r µ Lin4 = e non-renormalizable Dµ H D H To e models, ( H ) runaways =Kozaczuk Dµ HD Hthe H (35) origin 5 such although seem likely, given the tree-level of illustrate thevpcharacteristics electroweak k PT in this scenario, we consider four of the barrier betweenbenchmark vacua). The resulting the electroweak PT are quoted in = 50 GeV, and corpoints. Thesefeatures points, of labeled A D, are chosen such that m

26 Gravita%onal Waves from a Strong EWPT Results (Case 1: Non-Runaway Bubbles) Representa%ve models: HDM, NMSSM, Dark Sector (?), Dim-6 EFT Primary source: sound waves Kozaczuk 6

27 Gravita%onal Waves from a Strong EWPT Results (Case : Runaway Bubbles in Plasma) Representa%ve models: Higgs Portal, NMSSM, Dark Sector (?), Dim-6 EFT Primary source: sound waves + scalar field Kozaczuk 7

28 Gravita%onal Waves from a Strong EWPT Results (Case 3: Runaway Bubbles in Vacuum) Representa%ve models: Dilaton, Dark Sector (?) Primary source: scalar field Kozaczuk 8

29 Gravita%onal Waves from a Strong EWPT Conclusions of our study: Caprini, JK et al, JCAP 1604 (016) no.04, 001 For typical electroweak phase transitions, in which all relevant dimensionful parameters are near the electroweak scale, the configuration C1 is definitively better than the others, while the configuration C4 is decisively unsatisfactory. The performances of the configurations C and C3 are similar, but C has the ability to test a larger fraction of the considered benchmark points. For phase transitions not strictly related to electroweak symmetry breaking, e.g. involving a dark sector or a dilaton, the predicted characteristics of the phase transitions exhibit more variation. Larger GW signals are allowed than in the electroweak case. For certain scenarios, the resulting GW spectrum can be probed by all considered elisa designs. The C1 configuration, however, has the potential to test a much wider region of the parameter space in such models than do C C4. Kozaczuk 9

30 Gravita%onal Waves from a Strong EWPT C1 configura'on is similar to the original LISA design. Feasible if NASA decides to par%cipate The takeaway: LISA will likely have a real chance at detec%ng gravita%onal waves from a strong first-order phase transi%on, providing valuable informa%on about our cosmic history. In some cases (e.g. for highly decoupled new physics), LISA is likely the only op%on. Kozaczuk 30

31 Gravita%onal Waves from a Strong EWPT Only very strong phase transi%ons yield observable gravita%onal radia%on. What would a signal imply for models electroweak baryogenesis? Can electroweak baryogenesis source both detectable gravita'onal waves and the observed baryon asymmetry? Depends on the scenario Kozaczuk 31

32 Implica%ons for Electroweak Baryogenesis Usual charge transport mechanism could produce both, provided significant plasma velocity in front of the bubble J.M. No, 011 v w -v + relevant for EWB Not very generic, but possible v w relevant for gravita%onal waves Kozaczuk 3

33 Implica%ons for Electroweak Baryogenesis Other possibili'es: Bubbles inside bubbles (Caprini + No, 011) Baryogenesis from bubble collisions -Natural Cold EWB (Konstandin + Servant, 011) -Par%cle produc%on from elas%c collisions (Katz + Riomo, 016) Kozaczuk 33

34 Complementarity GW interferometers will complement collider searches in probing a strong first-order EWPT Huang et al, 016 Kozaczuk 34

35 Beyond the EWPT Other cosmological PT scenarios might also predict a signal at LISA and other GW experiments E.g. Monodromy infla'on Hebeker et al, 016 If more than one vacuum is populated in a Hubble patch, can give rise to a strong first order cosmological PT For other infla%onary signals expected at LISA, see Garcia-Bellido et al, 016 and Marco s slides Kozaczuk 35

36 Beyond the EWPT Other GW experiments can probe PTs at different scales Dev et al, 016 Schwaller, 015 Higher temperature phase transi%ons can predict signal at aligo. Lower temperature phase transi%ons can predict signal at pulsar %ming arrays Kozaczuk 36

37 Takeaways LISA can probe cosmological phase transi%ons in many BSM scenarios. The closer to the original LISA design the bemer. A signal in LISA can be compa%ble with viable electroweak baryogenesis PTs at other scales may yield a signal in advanced LIGO or pulsar %ming arrays Kozaczuk 37