eos vorstellung
Transcription
eos vorstellung
Weak magnetic fields in early-type stars Jon Braithwaite Bonn University Santiago, 7th May 2012 Mittwoch, 9. Mai 12 Contents Review of observations of early-type stars − Intermediate-mass stars − High-mass stars Hypotheses to explain nature and/or presence of magnetic fields: − Fossil fields and failed fossils − Subsurface convection − Other possibilities Summary Mittwoch, 9. Mai 12 Contents Review of observations of early-type stars − Intermediate-mass stars − High-mass stars Hypotheses to explain nature and/or presence of magnetic fields: − Fossil fields and failed fossils − Subsurface convection − Other possibilities Summary Mittwoch, 9. Mai 12 Upper-main-sequence stars spectral type field detected Steady, strong, large-scale fields Rest of population Mittwoch, 9. Mai 12 A and late B O and early B Upper-main-sequence stars spectral type A and late B field detected Steady, strong, large-scale fields Rest of population Mittwoch, 9. Mai 12 ~10% of stars, the so-called Ap/Bp stars 200G < B < 100kG O and early B A and late B stars: Ap/Bp stars First magnetic fields found outside solar system (Babcock 1947) Constant (> decades) 200G-100kG strength Large-scale (often dipolar) 53 Cam (Kochukhov et al 2004) Mittwoch, 9. Mai 12 α2 CVn (Kochukhov et al 2002) Upper-main-sequence stars spectral type A and late B field detected Steady, strong, large-scale fields ~10% of stars, the so-called Ap/Bp stars 200G < B < 100kG Rest of population Weak fields found in two bright A stars. Present in all other stars too? Mittwoch, 9. Mai 12 O and early B A and late B stars: the rest Bimodality: No stars with fields 10 - 200 G? (Aurière et al 07) Recent detections: (Lignières et al 2009, Petit et al 2010, 2011) name type mass /Msun v sin i /km/s spin period /hours age /Myr field strength measured Vega A0V ~2.5 22 12 ~400 0.6 ± 0.3 G Sirius A1V 2.1 16 24 - 120? ~200 0.2 ± 0.1 G Time variability: uncertain Geometry: uncertain but not very small-scale Mittwoch, 9. Mai 12 Observations of Vega Mittwoch, 9. Mai 12 Contents Review of observations of early-type stars − Intermediate-mass stars − High-mass stars Hypotheses to explain nature and/or presence of magnetic fields: − Fossil fields and failed fossils − Subsurface convection − Other possibilities Summary Mittwoch, 9. Mai 12 Upper-main-sequence stars spectral type A and late B O and early B field detected Steady, strong, large-scale fields ~10% of stars, Some fraction the so-called Ap/Bp stars (10, 20%?) 200G < B < 100kG B ~ 1kG? Rest of population Weak fields found in two bright A stars. Present in all other stars too? Mittwoch, 9. Mai 12 Observations of main-sequence stars: O and early B Some fraction have Ap-like fields (all Of?p magnetic?) Detection limit is much higher than in A stars Typically kG strength Field topology of τ Sco, a B0 main-sequence star (MV=2.8) (Donati et al. 2006, using Zeeman-Doppler imaging) Mittwoch, 9. Mai 12 Upper-main-sequence stars spectral type A and late B O and early B field detected Steady, strong, large-scale fields ~10% of stars, Some fraction the so-called Ap/Bp stars (10, 20%?) 200G < B < 100kG B ~ 1kG? Rest of population Weak fields found in two No fields detected, but bright A stars. Present in indirect suggestions of all other stars too? magnetic activity in whole population Mittwoch, 9. Mai 12 `Non-magnetic´ massive stars Most stars have no detected/ detectable field Various observational phenomena: discrete absorption components (DACs), line profile variability (LPV), wind clumping, solar-like oscillations, red noise, photometric variability, Xray emission Some of these phenomena are ubiquitous -- cannot be explained by large-scale 'fossil' fields Could be caused by: magnetic activity at surface: reconnection heating etc. wind shocks caused by linedeshadowing instability, but seeded by magnetic activity at surface? (DACs: Kaper, Henrichs et al. 1999) Mittwoch, 9. Mai 12 Contents Review of observations of early-type stars − Intermediate-mass stars − High-mass stars Hypotheses to explain nature and/or presence of magnetic fields: − Fossil fields and failed fossils − Subsurface convection − Other possibilities Summary Mittwoch, 9. Mai 12 Evolution of magnetic field in absence of driving Put arbitrary magnetic field into a star du = dt U2 1 1 ⇤P + g + (⇤ ⇥ B) ⇥ B ⇥ 4 ⇥ cs2 vff2 vA2 (=B2/4πρ) Lorentz force is balanced by inertia Therefore evolution takes place on Alfvén timescale τA = L / vA Reconnection occurs, magnetic energy is destroyed, field strength drops Magnetic relaxation involves motion on spherical shells Mittwoch, 9. Mai 12 Evolution of magnetic field in absence of driving Put arbitrary magnetic field into a star du = dt U2 1 1 ⇤P + g + (⇤ ⇥ B) ⇥ B ⇥ 4 ⇥ cs2 vff2 vA2 (=B2/4πρ) Lorentz force is balanced by inertia Therefore evolution takes place on Alfvén timescale τA = L / vA Reconnection occurs, magnetic energy is destroyed, field strength drops Eventually, an equilibrium is reached: Lorentz force balanced by non-spherical deviations of P and ρ Mittwoch, 9. Mai 12 Put arbitrary magnetic field into a star U2 cs2 From momentnced by inertia Therefore evolution takes vff2 Reconnection occurs, mag Initial conditions Mittwoch, 9. Mai 12 vA2 Simulations of magnetic relaxation to equilibrium (Braithwaite 2008) Evolution of magnetic field in absence of driving Put arbitrary magnetic field into a star du = dt U2 1 1 ⇤P + g + (⇤ ⇥ B) ⇥ B ⇥ 4 ⇥ cs2 vff2 vA2 (=B2/4πρ) Lorentz force is balanced by inertia Therefore evolution takes place on Alfvén timescale τA = L / vA Reconnection occurs, magnetic energy is destroyed, field strength drops Eventually, an equilibrium is reached: Lorentz force balanced by non-spherical deviations of P and ρ Energy (and field strength; E=∫B2/8π dV), fall a lot during relaxation. Therefore, time taken to reach equilibrium depends on final Alfvén timescale (confirmed in simulations) In an Ap star with B ~ 1 kG, τA ~ 10 yr But during relaxation, age of star t ~ τA Mittwoch, 9. Mai 12 Importance of rotation Could fields of Vega and Sirius be dynamically evolving? Assuming age ~ τA , B ~ 25 and 50 µG in Vega & Sirius Too weak! Mittwoch, 9. Mai 12 Importance of rotation Could fields of Vega and Sirius be dynamically evolving? Assuming age ~ τA , B ~ 25 and 50 µG in Vega & Sirius Too weak! Missing ingredient is rotation du = dt U2 1 1 ⇤P + g + (⇤ ⇥ B) ⇥ B - 2Ω x u ⇥ 4 ⇥ cs2 vff2 In regime Ω << 1/τA , τevol ~ τA In regime Ω >> 1/τA , τevol ~ τA2 Ω vA2 (LΩ) U (τevol = L / U) Again assuming age ~ τevol , fields of Vega and Sirius should be ~ 10 G. Better! Mittwoch, 9. Mai 12 Improving the numbers First estimate gives too strong fields Possible explanations: Mittwoch, 9. Mai 12 Field strength higher in interior than at surface Observations underestimate field strength (cancellation effects) Characteristic length scale L is smaller than R What determines what happens? Initial conditions? Strength and form of equilibrium depend on initial conditions, particularly on magnetic helicity H ≡ ∫A.B dV, where A is vector potential given by B = ∇xA Units: H = E L H is roughly conserved, and at equilibrium Eeq = H / Leq ≳ H / R Possible scenarios for initial conditions: − Large H, large E: high-energy equilibrium forms fast. At time of observation, τevol << age − Small H, large E: evolves quickly at first, then more slowly. No equilibrium found during lifetime of star. τevol = age. − Small H, small E: evolves slowly right from beginning. No equilibrium found. τevol > age. Problematic for star formation! Reference: Braithwaite & Cantiello 2012 Mittwoch, 9. Mai 12 Contents Review of observations of early-type stars − Intermediate-mass stars − High-mass stars Hypotheses to explain nature and/or presence of magnetic fields: − Fossil fields and failed fossils − Subsurface convection − Other possibilities Summary Mittwoch, 9. Mai 12 Subsurface convection Subsurface convective layer Possible dynamo activity Field could rise to surface via buoyancy Cantiello et al. 2008 Mittwoch, 9. Mai 12 Subsurface convection Subsurface convective layer Possible dynamo activity Field could rise to surface via buoyancy Cantiello et al. 2008 Mittwoch, 9. Mai 12 Subsurface convection in massive stars Dynamo field may fluctuate on timescale ~ τconv ~ day Field reaches surface via buoyancy, since field provides pressure without mass Buoyant rise at Alfvén speed Mittwoch, 9. Mai 12 Field strengths expected Assuming equipartition in convective zone: B2 / 8π ~ ρ u2 / 2 B (G) 0 500 120 MSun 6.0 2000 2500 150 1000 0 500 0 2000 1500 00 35 MSun 500 5.0 Log L 1500 10 5.5 1000 1000 20 MSun 4.5 4.0 10 MSun 3.5 3.0 4.8 500 0 GAL 7 MSun 4.6 4.4 logTeff 4.2 Cantiello & Braithwaite 2011 Mittwoch, 9. Mai 12 4.0 Field strengths expected Further assuming B ∝ρ B (G) 2/3 on way to surface Fields can have significant effects on wind 10 40 80 320 160 40 16 0 35 MSun 1020 5 40 5 20 20 MSun 4 5 10 MSun 3 4.8 GAL 7 MSun 4.6 4.4 logTeff 4.2 Cantiello & Braithwaite 2011 Mittwoch, 9. Mai 12 > 320 80 120 MSun 20 10 But small-scale fields are not directly detectable with Zeeman effect 6 5 Log L 0 4.0 Spots on surface? Magnetic pressure causes gas pressure in magnetic feature to be lower than in surrounding gas Consequently photosphere is lower So spot is hot (In the Sun, convection is inhibited so spot is colder) For fields of ~ 100 G emerging at the surface this leads to a temperature increase of ~ 500-1000 K. A hot, bright spot Mittwoch, 9. Mai 12 Appearance of spots Spot in convective star Mittwoch, 9. Mai 12 Spot in radiative star? Spots in a O8V star? HD 46149 (Degroote et al. 2010, using CoRoT) Mittwoch, 9. Mai 12 Spots in a B0.5IV star? HD51756 (Pápics et al. 2011, using CoRoT) Mittwoch, 9. Mai 12 Spots in a B0.5IV star? HD51756 (Pápics et al. 2011, using CoRoT) Mittwoch, 9. Mai 12 Subsurface convection in A and late B stars Same principle as in massive stars, except Opacity bump and convective layer due to helium ionisation Very thin layer, 1% of radius below surface Can produce fields at photosphere of a few gauss (subject to assumptions e.g. equipartition field, B~ρ2/3 as field rises buoyantly through overlying radiative layer, etc.) Origin of Vega & Sirius fields? Mittwoch, 9. Mai 12 Comparison of subsurface convection and failed-fossil fields Geometry of field: length scale Failed fossil Subsurface convection L ~ R/5 ? small-scale? Geometry: latitude dependence? stronger at pole? probably Time variability & (weak) x-rays no yes Field strength predicted in Vega & Sirius < 10G ~3G ? Correlation with age yes not much Correlation with spin period yes ? dynamo theory Likely in massive stars too? no yes Mittwoch, 9. Mai 12 Comparison of subsurface convection and failed-fossil fields Strengths Failed fossil Subsurface convection produces right length scales we know dynamos work produces right field strengths right field strengths Weaknesses ignores meridional circulation ? dynamo theory ? etc still have issue of bimodality in initial conditions how to get larger length scales? ignorance of initial conditions details of convective hampers predictive power layers still uncertain no confirmation yet of τevol ~ τA2 Ω from simulations Mittwoch, 9. Mai 12 Comparison of subsurface convection and failed-fossil fields Strengths Failed fossil Subsurface convection produces right length scales we know dynamos work produces right field strengths right field strengths Weaknesses ignores meridional circulation ? dynamo theory ? etc still have issue of bimodality in initial conditions how to get larger length scales? ignorance of initial conditions details of convective hampers predictive power layers still uncertain no confirmation yet of τevol ~ τA2 Ω from simulations Mittwoch, 9. Mai 12 Further thoughts (on [failed] fossils) Unknown initial radial magnetic energy distribution Why bimodality in initial helicity? Assume decaying T-S dynamo and guess initial rotation profile? Mergers? But all merger does is create differential rotation. Why any different from primordial diff. rot.? Blue stragglers an important clue? Do simulations to confirm τevol ~ τA2 Ω Relaxation -- motion on spherical shells: what forces (components perpendicular to gravity)? pressure gradient Lorentz force. How many degrees of freedom? Coriolis --OR-- inertia Simulations with barotropic e.o.s.? Mittwoch, 9. Mai 12 Contents Review of observations of early-type stars − Intermediate-mass stars − High-mass stars Hypotheses to explain nature and/or presence of magnetic fields: − Fossil fields and failed fossils − Subsurface convection − Other possibilities Summary Mittwoch, 9. Mai 12 Further thoughts regarding Vega & Sirius Meridional circulation could drive dynamo directly? modify a failed fossil drive differential rotation? Differential rotation could drive dynamo BUT: dynamo vanishes towards surface So how to get field to surface? bouyancy instability buoyancy, diffusive ...? Mittwoch, 9. Mai 12 Contents Review of observations of early-type stars − Intermediate-mass stars − High-mass stars Hypotheses to explain nature and/or presence of magnetic fields: − Fossil fields and failed fossils − Subsurface convection − Other possibilities Summary Mittwoch, 9. Mai 12 Summary Weak magnetic fields evolve very slowly in rotating stars Vega and Sirius fields could be `failed fossils´, continuously evolving come from a subsurface convective dynamo Either way, all A & late B stars without fossil fields should have fields of at least ~1G unless v.slowly rotating Subsurface dynamo in O and early B stars could produce ~100 G fields at the surface, and play a role in the various unexplained observational phenomena in these stars Mittwoch, 9. Mai 12 Early-type stars: Summary of observations and theories spectral type A and late B O and early B field detected Steady, strong, large-scale fields ~10% of stars, Some fraction the so-called Ap/Bp stars (10, 20%?) 200G < B < 100kG B ~ 1kG? Theory: fossil field Theory: fossil field (?) Rest of population Weak fields found in two No fields detected, but bright A stars. Present in indirect suggestions of all other stars too? magnetic activity in whole population Theories: Failed fossil, subsurface convection? Theory: Seems likely all other A Subsurface convection? stars have these fields. Mittwoch, 9. Mai 12 Mittwoch, 9. Mai 12 Brainstorming J: just done D: develop F: future Theoretical considerations Observational puzzles Subsurface convection Drop in flux during Ap star lifetime F: observational consequences Fossils and failed fossils D: formation: what to expect given different radial energy distributions and helicity. what are the degrees of freedom in initial conditions? field becomes potential near surface? meridional circulation? Origin of Ap-minority, variety in pulsar fields, etc. Merger hypothesis? Field strengths in accreting NSs and in former accretors D&F: diffusive evolution: different in different stars. santiago gang know about hall drift etc etc F: think about initial conditions? T-S dynamo dies away leaving what? Likely radial energy distribution? Magnetic effects with differential rotation (v. brief because of henk‘s talk) mention work of Luis what else? Mittwoch, 9. Mai 12 44 Abundance patterns on HR 3831 (Kochukhov et al) Observations in main-sequence stars: A and late B Some fraction display fossil fields, so-called Ap/Bp stars − Steady over > decades − Strong: 200G to 100kG − Large-scale, often roughly dipolar Rest have fields < 10 gauss − No stars with fields 10 to 200 G? (Aurière et al 07) − Recent detection of weak fields in Vega and Sirius: 0.6 and 0.2 G. Great variety of chemical abundances in all A stars: lots going on with mixing, radiative levitation, gravitational settling, selective evaporation, etc. Lots of physics! Mittwoch, 9. Mai 12 Opacity 46 Mittwoch, 9. Mai 12 Opacity 46 Mittwoch, 9. Mai 12 Opacity Strong Z-dependency of all phenomena connected to FeCZ 46 Mittwoch, 9. Mai 12 Surface Turbulence Matteo Cantiello Mittwoch, 9. Mai 12 Near-Surface Convection in Massive Stars KITP – October 24th 2011 3D Hydro Simulations Pencil Code Setup: piecewise polytropic (stableunstable-stable) Cartesian grid 128 x 128 x 256 Fcon/Frad ~ 0.3 Re ~ 80 Shown is vertical (Brandenburg & Dobler 2002) Preliminary, low resolution runs!!! (Cantiello, Käpylä, Brandenburg et al. In Prep.) Matteo Cantiello Mittwoch, 9. Mai 12 Near-Surface Convection in Massive Stars KITP – October 24th 2011 Microturbulence Is the additional broadening coming from nonthermal motions varying on a small scale in the region of line formation. Observed line profile Intrinsic line profile ΔλD = Matteo Cantiello Mittwoch, 9. Mai 12 λ c V 2 therm Near-Surface Convection in Massive Stars +ξ 2 turb KITP – October 24th 2011 Microturbulence Is the additional broadening coming from nonthermal motions varying on a small scale in the region of line formation. Matteo Cantiello Mittwoch, 9. Mai 12 Near-Surface Convection in Massive Stars KITP – October 24th 2011 Microturbulence Is the additional broadening coming from nonthermal motions varying on a small scale in the region of line formation. To fit stellar spectra of hot stars microturbulence (~0-25 km/s) is needed Matteo Cantiello Mittwoch, 9. Mai 12 Near-Surface Convection in Massive Stars KITP – October 24th 2011 Microturbulence Is the additional broadening coming from nonthermal motions varying on a small scale in the region of line formation. To fit stellar spectra of hot stars microturbulence (~0-25 km/s) is needed Used as a fudge-factor. Unknown physical origin Matteo Cantiello Mittwoch, 9. Mai 12 Near-Surface Convection in Massive Stars KITP – October 24th 2011 Microturbulence Is the additional broadening coming from nonthermal motions varying on a small scale in the region of line formation. To fit stellar spectra of hot stars microturbulence (~0-25 km/s) is needed Used as a fudge-factor. Unknown physical origin But recently a correlation between near-surface convection and microturbulence has been found! Matteo Cantiello Mittwoch, 9. Mai 12 Near-Surface Convection in Massive Stars KITP – October 24th 2011 Energetic considerations 2 Eg ρ c ⎛ v c ⎞ ≈ M c ⎜ ⎟ Et ρ s ⎝ ξ ⎠ € Theoretical models Cantiello et al. 2009 Matteo Cantiello Mittwoch, 9. Mai 12 Near-Surface Convection in Massive Stars KITP – October 24th 2011 Energetic considerations 2 Eg ρ c ⎛ v c ⎞ ≈ M c ⎜ ⎟ Et ρ s ⎝ ξ ⎠ € Theoretical models + VLT-Flames observations Cantiello et al. 2009 Matteo Cantiello Mittwoch, 9. Mai 12 Near-Surface Convection in Massive Stars KITP – October 24th 2011 Results ξ = f (L,Teff ,Z) € Matteo Cantiello Mittwoch, 9. Mai 12 Near-Surface Convection in Massive Stars KITP – October 24th 2011 Results ξ = f (L,Teff ,Z) € Matteo Cantiello Mittwoch, 9. Mai 12 Near-Surface Convection in Massive Stars KITP – October 24th 2011 Results ξ = f (L,Teff ,Z) € Matteo Cantiello Mittwoch, 9. Mai 12 Near-Surface Convection in Massive Stars KITP – October 24th 2011 Results ξ = f (L,Teff ,Z) € Matteo Cantiello Mittwoch, 9. Mai 12 Near-Surface Convection in Massive Stars KITP – October 24th 2011 Solar-like oscillations Matteo Cantiello Mittwoch, 9. Mai 12 Near-Surface Convection in Massive Stars KITP – October 24th 2011 Near-Surface Convection (Solar Z) Matteo Cantiello Mittwoch, 9. Mai 12 Near-Surface Convection in Massive Stars KITP – October 24th 2011 Solar-like oscillations in massive stars Cantiello et al. 2009 Suggest that near-surface convection in hot, massive stars could cause stochastically excited pulsations Belkacem et al. 2009 Corot detection of solar-like oscillations in the massive star V1449 Aql (B type Star) [However, see Aerts et al. 2011] Belkacem et al. 2010 Theoretical calculations of stochastically excited modes from sub-surface convection. Degroote et al. 2010 Corot detection of solar-like oscillations in an O-type star Matteo Cantiello Mittwoch, 9. Mai 12 Near-Surface Convection in Massive Stars KITP – October 24th 2011 Solar-like oscillations in O star HD 46149 (Degroote et al. 2010) Matteo Cantiello Mittwoch, 9. Mai 12 Near-Surface Convection in Massive Stars KITP – October 24th 2011 Photometric variability: HRD location HD46149 (Degroote+ 2011) Matteo Cantiello Mittwoch, 9. Mai 12 Near-Surface Convection in Massive Stars KITP – October 24th 2011 Red Noise Matteo Cantiello Mittwoch, 9. Mai 12 Near-Surface Convection in Massive Stars KITP – October 24th 2011 “Red Noise” in O stars Variability in the CoRoT photometry of 3 hot O- type stars (Blomme et al. 2011) Matteo Cantiello Mittwoch, 9. Mai 12 Near-Surface Convection in Massive Stars KITP – October 24th 2011 “Red Noise” in O stars Variability in the CoRoT photometry of 3 hot O- type stars No clear pulsations detected (Blomme et al. 2011) Matteo Cantiello Mittwoch, 9. Mai 12 Near-Surface Convection in Massive Stars KITP – October 24th 2011 “Red Noise” in O stars Variability in the CoRoT photometry of 3 hot O- type stars No clear pulsations detected Variability of stochastic nature (Blomme et al. 2011) Matteo Cantiello Mittwoch, 9. Mai 12 Near-Surface Convection in Massive Stars KITP – October 24th 2011 “Red Noise” in O stars Variability in the CoRoT photometry of 3 hot type stars No clear pulsations detected Variability of stochastic nature Near-surface convection, granulation or wind inhomogeneities O- (Blomme et al. 2011) Matteo Cantiello Mittwoch, 9. Mai 12 Near-Surface Convection in Massive Stars KITP – October 24th 2011 Photometric variability: HRD location HD46150 HD46223 (Blomme+ 2011) HD46966 Matteo Cantiello Mittwoch, 9. Mai 12 Near-Surface Convection in Massive Stars KITP – October 24th 2011 Magnetic fields / Spots Matteo Cantiello Mittwoch, 9. Mai 12 Near-Surface Convection in Massive Stars KITP – October 24th 2011 B fields in massive stars (direct evidence) About a dozen magnetic OB stars found (e.g. Donati, Hubrig, Neiner, Petit) Detection through Zeeman spectral signature Bias toward strong, large scale fields Origin unclear. Likely Fossil (Wade et al. 2010) Tau Sco Credits: Jardine & Donati Important evolutionary consequences (e.g. ud-doula & Owocki 02, Meynet et al. 2010) Matteo Cantiello Mittwoch, 9. Mai 12 Near-Surface Convection in Massive Stars KITP – October 24th 2011