Magnetic fields in the early universe

Size: px

Start display at page:

Download "Magnetic fields in the early universe"

Brooke Gibbs
6 years ago
Views:

1 Magnetic fields in the early universe KandaswamySubramanian a a Inter-University Centre for Astronomy and Astrophysics, Pune , India. Heidelberg Joint Astronomical Colloquium, 1st December, 2009 p.0/28

2 Plan Universe is filled with magnetic fields Magnetic fields in galaxies and clusters Why early universe? Early Universe Generation Magnetic signals in the CMB Primordial fields and early structure formation K. Subramanian, Magnetizing the Universe, PoS proceedings, arxiv: K. Subramanian, Magneic fields in the early universe, AN(in Press), arxiv: Heidelberg Joint Astronomical Colloquium, 1st December, 2009 p.1/28

t r y y t i c o l e v Measuring B fields:synchrotron Radiation B t r c e E l n o or t c e j a (http://www.statemaster.

3 t r y y t i c o l e v Measuring B fields:synchrotron Radiation B t r c e E l n o or t c e j a ( encyclopedia/faraday-effect) n o t r c e E l n o i t a er l e c c a ' n s o t r c e E l us o e n a t n a t ns i k H E FaradayRotationgives B RM = 0.81 R n e B dl (rad m 2 ) Synchrotronpolarizationgives B Heidelberg Joint Astronomical Colloquium, 1st December, 2009 p.2/28

4 Galactic Magnetic Fields: Observations Synchrotron polarization and Faraday rotation probe Bfields. M51at6cm (Fletcher and Beck) Few-ten µg mean Fields coherent on 10 kpc scales Howdosuch large scale galactic fields arise? Early universe? Dynamos? Heidelberg Joint Astronomical Colloquium, 1st December, 2009 p.3/28

Cluster magnetic fields ClusterfieldsfromRadiohalos(B total )andfaradayrotation(b coherent ) Galaxy clusters host few µg fields correlated on 10 Kpc scales Hydra Cluster Statistical RM study, Clarke

5 Cluster magnetic fields ClusterfieldsfromRadiohalos(B total )andfaradayrotation(b coherent ) Galaxy clusters host few µg fields correlated on 10 Kpc scales Hydra Cluster Statistical RM study, Clarke et al, e 11 k E B(k) 3 [erg/cm ] 1e 12 (Vogt and Ensslin, 05) Magnetic Power Spectrum of Hydra A Cluster Kolmogoroff spectrum 1e 13 B = ( ) µ G l B = ( ) kpc 1e k [kpc ] Heidelberg Joint Astronomical Colloquium, 1st December, 2009 p.4/28

6 StrongBinhighzgalaxies? Universe increasingly Faraday opaque at high z? Bernetetal,Nature,454,302,2008;Kronbergetal.2008 Heidelberg Joint Astronomical Colloquium, 1st December, 2009 p.5/28

7 Why Early universe? WHY NOT? Scalar/Tensor perturbations generated. Why not B fields? In an Early universe Phase transition? Coherent Galaxy and Cluster fields? Dynamos?? Brandenburg& Subramanian, Phys. Rep. 417(2005) Seed field + amplification by motions(turbulence/shear) Can turbulent dynamos generate required coherence? Could probe early universe physics CMB Puzzles? CBI Excess Power? Alignments? Non Gaussianities? Fluxfreezing: B(t)a 2 (t) =constant,so B(z) = B 0 (1 + z) 2 B Gongalacticscales,interestingforGalaxy formation +galaxy/cluster B?(ρ B = ρ γ implies B 3µG). How can One Constrain/Detect Primordial B fields? Heidelberg Joint Astronomical Colloquium, 1st December, 2009 p.6/28

Early Universe timeline (http://www.damtp.cam.ac.

8 Early Universe timeline ( Heidelberg Joint Astronomical Colloquium, 1st December, 2009 p.7/28

Quantum fluctuations to Galaxies (http://wmap.gsfc.nasa.

9 Quantum fluctuations to Galaxies ( Heidelberg Joint Astronomical Colloquium, 1st December, 2009 p.8/28

10 Primordial fields from Inflation? Origin during Inflation:(Turner and Widrow, PRD, 1988) Rapid expansion vacuum fluctuations stretched to long wavelength classical fluctuations Negligible charge density breaks flux freezing. BUT Need to break conformal invariance of ED ConsiderEMaction(F µν = A ν;µ A µ;ν = A ν,µ A µ,ν ): S = Z g d 4 x 1 4 F µνf µν = Z g d 4 x 1 4 gµα g νβ F µν F αβ Considerconformaltransformation: g µν = Ω 2 g µν :implies g = Ω 4 g, g µα = Ω 2 g µα, A µ = A µ. S = S. FRWisconformaltoflatspace cantransformem conformally to flat space. So no amplification of EM waves. Heidelberg Joint Astronomical Colloquium, 1st December, 2009 p.9/28

11 Primordial fields from Early Universe? Many ways considered for breaking CI during inflation: S = Z g d 4 x [ f(φ, R) 1 4 F µνf µν bra 2 +gθf µν F µν Dµ ψ(d µ ψ) ] Coupling of EM action to inflaton, dilaton, curvature invariants, axion, charged scalars,... EM wave amplified from vacuum fluctuations Afterreheating Eshortedoutand Bfrozenin. Exponentiallysensitivetoparameters,asneed B 1/a ǫ B 10 9 to G,for f(φ) e αφ,for α 20 0(Ratra,1992) Needhugegrowthofcharge:aProblem?(Demozzietal,2009) Heidelberg Joint Astronomical Colloquium, 1st December, 2009 p.10/28

12 A worked example Consider f = a α, a(η) = a 0 (η/η 0 ) 1+β,(ηisconformaltime). For γ = α(1 + β) 1/2,andusing (k/ah) = ( kη), (KS,2009) dρ B d ln k = C(γ) «4+2γ k 2π 2 H4 = C(γ) ah 2π 2 H4 ( kη) 4+2γ 9 4π 2H4 (for γ = 2,c.fnearlyexponentialexpansion β 2and α 2) Using ρ B (0) = ρ B (a f /a 0 ) 4,instantaneousreheating, gt 3 a 3 conservation afterwards, gives «B H G 10 5 M pl When f = f 0 isconstant: F µν ; ν = (4πJ µ )/f 2 0,OR e N e/f 2 0 But f = f i (a/a i ) α a 2 duringinflation;sotheoryeitherstrongly coupled initially or very weakly coupled at end! Heidelberg Joint Astronomical Colloquium, 1st December, 2009 p.11/28

13 From Electroweak/QCD Phase transition? Correlationscaleusuallytiny: H 1 1cm(EW)or 10 4 cm QCDphasetransitionorcomoving R H 100AU/0.1pc Unless Helicity generation/conservation leads to Inverse Cascade(Brandenburg et al, PRD 96, Banerjee& Jedamzik, 2004) MagneticHelicity H = R V A B dv, A = B Aisvectorpotential, V isclosedvolume Measureslinksandtwistsin B Helicity is nearly conserved even when energy dissipated HelicitygenerationduringEWbaryogenesis: H/V n b /α! (Vachaspati, 2001; Copi et al 2008; Diaz-Gil et al, 2008) Heidelberg Joint Astronomical Colloquium, 1st December, 2009 p.12/28

14 Inverse cascade of helical B (Christensson, Hindmarsch, Brandenburg, 2001; Brandenburg 2001) Assuminghelicityconservation, H LB 2 LE constant. so de/dt E/(L/v) E 5/2 /H L B 2 t 2/3 (Sim. L t 1/2 ). Heidelberg Joint Astronomical Colloquium, 1st December, 2009 p.13/28

15 Magnetic field evolution Banerjee and Jedamzik, PRD, 70, , 2004 B 2 /(8πρ rad ) 10 7 (B 0 /10 9 G) 2 Conductivity high; Viscosity important around neutrino/photon decoupling. Fromtoptobottom,(a) h = 1 r = 10 2,(b) h = 10 3 r = 10 2 n = 3,(c) h = 0 r = 10 2 n = 3,(d) h = 1 r = 10 5 n = 3. T g = 100GeV. Heidelberg Joint Astronomical Colloquium, 1st December, 2009 p.14/28

16 Probing Early Universe B B 2 /(8πρ rad ) 10 7 B 2 9. Here B 9 = B 0 /(10 9 G) Magnetic stress metric perturbations, including Grav. Waves Lorentz force J B/c almost incompressible motions Overdamped by radiative viscosity, unlike compressible modes.(jedamzik et al, 1998; KS, JDB 1998) Survivesdampingfor L A > (V A /c)l Silk L Silk CMB signals from metric and velocity perturbations Postrecombination: n rad /n b 1 compressiblemotions seeds δρ/ρ First Structures B field Dissipation Ionization, Heating, Molecules Coherent primordial ng fields potentially detectable Heidelberg Joint Astronomical Colloquium, 1st December, 2009 p.15/28

17 CMB ANISOTRPIES LAST SCATTERING SURFACE IONIZED GAS SPATIAL VARIATION AT LSS θ SEEN AS oo ANGULAR ANISOTROPY TODAY T T (θ, φ) = X lm a lm Y lm (θ, φ); a lm a l m = C lδ ll δ mm. ( T) 2 T 2 = X l C l 2l + 1 4π Z l(l + 1)Cl 2π dln l Heidelberg Joint Astronomical Colloquium, 1st December, 2009 p.16/28

CMB Anisotropies Observations WMAP http://map.gsfc.nasa.gov Readheadetal.

18 CMB Anisotropies Observations WMAP Readheadetal.,ApJ,609,498,2004 CBI Excess Power? TTandTEPowerspectrum Heidelberg Joint Astronomical Colloquium, 1st December, 2009 p.17/28

19 CMB signals from tangled B fields Scalar Modes Subdominant to Inflation generated signal, (Shaw and Lewis, 2009, M. Giovannini, 2008, Yamazaki et al. 2008) VorticalmotionoffluidatLSS(Vectormodes)(KS&Barrow1998, KS, Seshadri, Barrow 2003) Tensors Significant at l < 100,(Durrer, Ferreira, Kahniashvilli, ) Polarization B(Curl) modes due to Vectors/Tensors Scalars only induce E(Gradient) modes (Seshadri&KS,2001;Macketal2002;Lewis2004;Gopal&Sethi,2005) FaradayRotaion ConvertsEtoBmodesignals Helicalfieldscanalsocause T B, E Bcrosscorrelations! Non Gaussian Statistics(Seshadri, KS 2009, Caprini et al 2009) Primordial few ng magnetic fields potentially detectable using the CMB Heidelberg Joint Astronomical Colloquium, 1st December, 2009 p.18/28

20 Vector mode CMB Signals B = 3 nano Gauss CBI BIMA n= 2.9 n= 2.5 Vortical motion of fluid at LSS(Vector modes) Can partially explain CBI Excess? NonGaussianStatisticsbecause F L B 2 KS,JDB, PRL, 81, 3375(1998); MN, 335, L57(2002); KS,TRS,JDB, MN, 344, L31(2003) Heidelberg Joint Astronomical Colloquium, 1st December, 2009 p.19/28

21 CMB Polarization Vector Modes B = 3 nano Gauss n= 2.5 LENSING n= 2.9 FOREGROUND Thomson Scattering + quadrupole at LSS Polarization B- Type Polarization(unlike inflationary Scalar Modes) SignalsbelowSilkDampingscale(l 10 3 ) TRS, KS, PRL, 87, (2001); KS,TRS,JDB, MN, 344, L31(2003) Heidelberg Joint Astronomical Colloquium, 1st December, 2009 p.20/28

22 CMB signals: scalar+tensor + Vector B λ = 4.7µG, n 3,Includingpassivecomponent, Shaw&Lewis,arXiv: TT Mode EE Mode /2πl(l K 2 +1)C l / Primary Scalar Magnetic Vector Magnetic Tensor Magnetic Scalar Passive Tensor Passive l BB Mode /2πl(l K 2 +1)C l / l TE Mode /2πl(l K 2 +1)C l 10-2 / / /2πl(l K 2 +1)C l Lensed Scalar l l Heidelberg Joint Astronomical Colloquium, 1st December, 2009 p.21/28

23 Structure formation signals Extra power in the matter power spectrum on small scales (Gopal, Sethi, 2003) Damping of B field energy by ambipolar diffusion and decaying turbulence, can heat/ionize universe at z 100 (Sethi, KS 2005; Schleicher, Banerjee, Klessen, 08). Altered Visibility function Additional CMB anisotropies Firstdwarfgalaxiesformathigh z > 10evenfor B 0.1nG,but for masses larger than magnetic and thermal Jeans mass. B field induced first structures Reionization? (Sethi, KS 2005, Tashiro, Sugiyama, 2006; Schleicher, Banerjee, Klessen, 08). Influence formation of first structures through catalyzing Molecule formation(sethi, Nath, KS 2008; Schleicher et al 2009) Probe through redshifted HI 21 cm signals (Tashiro, Sugiyama, 06;Schleicher, Banerjee, Klessen, 09; Sethi, KS 09) Heidelberg Joint Astronomical Colloquium, 1st December, 2009 p.22/28

24 Modified matter power spectrum Gopal,Sethi,JAA,2003; 3nG, n = 0, 1, 2, 2.9;LCDM Heidelberg Joint Astronomical Colloquium, 1st December, 2009 p.23/28

25 Post Recombination Blues VISIBILITY FUNCTION B = 3 ng, n = 2.8 Ambipolar damping and Turbulence decay gradual re-ionization Modified Visibility Function (Sethi, KS MN, 2005) REIONIZATION AMBIPOLAR DIFFUSION DECAYING TURBULENCE NORMAL VISIBILITY FN. (Sethi, KS, MN, 2005) FIRST STRUCTURES SEEDED BY PRIMORDIAL B FIELDS Even B = 0.1nG can induce 10 6 M dwarf galaxy collapse at high z > 15 causing early re-ionization (Sethi, KS MN, 2005) z = 10 z = 15 z = (SETHI,KS, MN, 2005) n = 2.8 n = 2.9 NONLINEAR LINEAR σ α 1/R^2 Heidelberg Joint Astronomical Colloquium, 1st December, 2009 p.24/28

26 Global 21 cm signals from reionization Sethi,KS,2009; 0, 0.5, 1nG HI global signal only seen in emission in magnetised models Heidelberg Joint Astronomical Colloquium, 1st December, 2009 p.25/28

27 HI correlation signals from reionization Sethi,KS,2009; 0.5, 1, 3nG Both ionization and density inhomogeneities contribute Heidelberg Joint Astronomical Colloquium, 1st December, 2009 p.26/28

28 Final Thoughts? Universe is full of magnetic fields! Origin from the early universe phase transitions? Primordial fields will leave signatures in the CMB, Structure formation. Redshifted 21 cm signals detectable with upcoming radio telescopesfor B 0 0.5nG Other Probes: Radio RMs(SKA) and High energy CRs Dynamos certainly needed to maintain fields BUT Need to understand their saturation better Heidelberg Joint Astronomical Colloquium, 1st December, 2009 p.27/28

29 THANKYOU! Heidelberg Joint Astronomical Colloquium, 1st December, 2009 p.28/28

Constraining primordial magnetic fields with observations

Constraining primordial magnetic fields with observations Kerstin Kunze (University of Salamanca and IUFFyM) KK PRD 83 (2011) 023006; PRD 85 (2012) 083004; PRD 87 (2013) 103005; PRD 89 (2014) 103016, arxiv:1703.03650