6.9.200721 June 2005 Masaryk University, Brno 1

The magnetic fields of peculiar A and B stars in open clusters

John D LandstreetUniversity of Western Ontario

London, Upper Canada

6.9.200721 June 2005 Masaryk University, Brno 2

The Team:

• Stefano Bagnulo, Armagh Observatory (N Ireland)• Vincenzo Andretta, INAF (Italy)• Luca Fossati, Vienna Observatory (Austria)• Elena Mason, ESO (Chile)• Jessie Silaj, University of Western Ontario (Canada)• Gregg Wade, Royal Military College of Canada

6.9.200721 June 2005 Masaryk University, Brno 3

Stars in the H-R diagram• The H-R diagram contains

stars in their initial, H-burning phase, on main sequence

• Following central H exhaustion stars evolve to giant phases, burning He and heavier elements

• After all fuel is gone, stars become white dwarfs, neutron stars, or black holes

• On main sequence we expect to find variations of chemistry only due to chemical evolution (metal enrichment) of the galaxy, thus only in cool stars

6.9.200721 June 2005 Masaryk University, Brno 4

Chemically peculiar A and B stars• Lower main sequence stars do show variations in

chemistry, due to evolution of interstellar abundances with age of galaxy

• Upper main sequence stars show little chemical variety, due mainly to (a) extreme youth and (b) strong winds

• In 1897 Antonia Maury identified some middle main sequence A and B stars with very peculiar spectra

• We now know that these “Ap” stars do have very peculiar atmospheric chemical compositions

• They constitute about 10% of all “tepid” main sequence stars

6.9.200721 June 2005 Masaryk University, Brno 5

Types of peculiar A and B stars• Peculiarities are identified by

prominent spectral features (HgMn stars, SrCrEu stars, He-weak stars)

• Each peculiarity type is found in a restricted mass (or temperature) range of main sequence

• Limited range in Teff even on main sequence shows that peculiarities are a surface effect, not a bulk property of stars

6.9.200721 June 2005 Masaryk University, Brno 6

Are Ap stars interesting?

• Observed variety of chemical peculiarities of Ap stars is due to varying competition among gravitational diffusion, radiative levitation, turbulent & convective mixing, mass loss, etc (Michaud, Vauclairs, Alecian, Charland…)

• In cooler and hotter main sequence stars, this competition is overwhelmed by a single process (hot stars: mass loss; cool stars: deep convection zone)

• Thus Ap stars are unique laboratories of internal stellar hydrodynamics

• Furthermore, some (but not all) chemically peculiar A and B stars have strong, global magnetic fields, so we can study interaction of fields with hydrodynamics

6.9.200721 June 2005 Masaryk University, Brno 7

Magnetic fields in stars• Our work has so far focused on

magnetic field measurements, so review how this is done

• Magnetic fields in stars are detected by the Zeeman effect

• In a magnetic field, a single spectral line splits into pi and sigma components, with separation proportional to B

• Components are polarized, and both line of sight and transverse field components may be measured by spectropolarimetry

• Observed line profiles are result of summing over stellar disk

6.9.200721 June 2005 Masaryk University, Brno 8

Fields in Ap stars

• Fields are detected in some peculiar stars via directly observed line splitting (e.g.

HD 94660)• Most fields are detected by

polarization in spectral lines, as in NGC2244-334

• Field strengths are of order 100 - 30000 G (0.01 – 3 T)

• Fields appear to cover whole star, they are not spotty as on surface of Sun

6.9.200721 June 2005 Masaryk University, Brno 9

Features of magnetic Ap stars

• Field strength, photometric brightness, and spectrum are usually periodically variable

• Periods range from 0.5d to many years, inversely correlated with v sin i

• => Period is rotation period of star

• Specific angular momentum typically 0.1x or less of normal A star, => extra braking (when?)

6.9.200721 June 2005 Masaryk University, Brno 10

A model of magnetic Ap structure

• Observed relation between field modulus |B| and line-of-sight component Bz => field structure is “simple”

• Periodic Bz field reversal suggests dipole-like topology with dipole axis inclined to rotation axis: the “oblique dipole rotator”

• Spectrum and light variations => non-uniform distribution of various elements over stellar surface (“abundance patches”)

• Such patches are possible because magnetic field suppresses “weather” which would mix atmosphere horizontally

6.9.200721 June 2005 Masaryk University, Brno 11

Magnetic fields – some basic problems

• What is the nature of magnetic fields found in some A and B stars? How are they produced?

-- Long-term stability, simple structure, lack of “activity”, lack of correlation between B and rotation rate suggest that field is a “fossil” left from (at least) PMS

• Why do Ap stars have fields while other A & B’s do not? -- Not yet understood• How does the field evolve as the star evolves? -- If it is a fossil, there is ohmic decay plus distortion

and amplification due to stellar structure changes

6.9.200721 June 2005 Masaryk University, Brno 12

Observational study of Ap evolution

• At present, we know that at some unknown time in its main sequence life, a magnetic Ap star can have an observed field structure and surface chemistry

• To make observed characteristics of A stars into far more powerful probes of (M)HD processes, we want to be able to associate a particular field structure and chemistry with a particular mass and age, not just say that the observed state happen sometime, in a star of unknown age and mass

6.9.200721 June 2005 Masaryk University, Brno 13

Observational study of Ap evolution - 2

• We need to determine masses, ages, and fractional ages (= fraction of MS lifetime) for a substantial sample of Ap stars

• The obvious method is to determine Teff and log(L/Lo) for field stars from parallaxes and photometry, and then compare results with standard evolution models in HR diagram to determine mass and fractional age

• We can then associate a particular observed star with a particular age since the PMS – 106, 107, 108 yr, etc

• We can also look for statistical trends in field strength, chemical abundances, rotation periods, etc

6.9.200721 June 2005 Masaryk University, Brno 14

Hubrig Theory• Hubrig et al have placed a

sample of nearby (field) magnetic stars with Hipparcos parallaxes onto the HR Diagram

• They claim that they find evidence that magnetic fields first emerge in stars of M < 3Mo after 30% of the main sequence lifetime.

• Is this reasonable? Consider rotation

6.9.200721 June 2005 Masaryk University, Brno 15

Slow rotation of magnetic Ap stars

• Slow rotation of Ap stars probably due to magnetic coupling with circumstellar material (accretion disk, stellar wind)

• Even young Ap stars (in clusters, associations) have slow rotation, so loss of angular momentum probably occurs in PMS phase (North)

• Slowest rotators are cooler – lower mass – Ap stars that spend longer in PMS phase (Stepien)

• Hubrig et al result conflicts completely with this picture. If it is right, how do Ap’s lose angular momentum without a surface field present to couple to surroundings?

6.9.200721 June 2005 Masaryk University, Brno 16

New observational data• Hubrig results are quite uncertain, even with

Hipparcos parallaxes!

• (Realistic) errors of ~500 K for Te and ~0.1 dex for log(L/Lo) lead to large age uncertainties, especially if the bulk composition is not known – see figures at right

• How can we provide better masses and ages from observation to better constrain physical processes active in magnetic Ap stars?

6.9.200721 June 2005 Masaryk University, Brno 17

Cluster magnetic Ap stars

• Ap’s in clusters would have more accurate ages.• Two recent major advances have made cluster

magnetic stars accessible: -- New proper motions from Hipparcos and

especially Tycho-2 have greatly improved knowledge of cluster membership down to V~ 10 or fainter, and surveys by Maitzen group have greatly expanded information about probable Ap stars beyond classification by Abt, etc.

-- Powerful spectropolarimeters on large telescopes (FORS1 on ESO VLT, Espadons on CFHT) make possible field measurement at V ~ 10 or even 12

6.9.200721 June 2005 Masaryk University, Brno 18

Cluster star magnetic measurements• Bagnulo, Mason,

Wade, Silaj & I have been observing probable cluster Ap’s for fields using FORS1 at VLT, and ESPaDOnS at CFHT

• Candidates are from Δa or Geneva Z photometry, or spectral classifications

• We now have about 80 field detections in some 30 clusters and associations (including Orion and Sco-Cen)

6.9.200721 June 2005 Masaryk University, Brno 19

How to use these data?

• We determine membership using parallaxes from Hipparcos, proper motions from Tycho-2, and occasional radial velocities.

• To find mass of a cluster star, we use Te, determined from Geneva or uvby photometry

• We find luminosity using cluster distance and new BC’s.

• Then compare stars to evolution tracks (Geneva, Padova, etc.) as before, but now only for the cluster’s known age

6.9.200721 June 2005 Masaryk University, Brno 20

Results so far

• Sample is rich in massive and relatively young stars• Fractional ages of young cluster magnetic Ap stars are

much more precise than those of field stars• Fields and fluxes definitely decline with age for M above

3 Mo, beyond about 30 Myr, but hardly change with time for lower mass stars

• Contrary to Hubrig theory, plenty of magnetic fields in young 2 – 3 Mo stars

• Hardly any stars in sample below 2 Mo of any age, even though Ap’s occur down to 1.6 Mo among field stars!?

6.9.200721 June 2005 Masaryk University, Brno 21

Some conclusions

• It is now very practical to study fields of cluster Ap stars

• If we split samples by mass (say 2 - 3 or 3 - 4 Mo), we find statistical decrease of fields from young to old for stars above 3 Mo, but not below this mass.

• There is no obvious shortage of young magnetic stars between 2 and 3 Mo. The Hubrig et al hypothesis of late field emergence is not confirmed.

• However, there does seem to be a shortage of magnetic stars (and even Ap stars) with M<2Mo -- of all ages -- in open clusters, compared to the field where Ap stars down to 1.6 Mo are found.

6.9.200721 June 2005 Masaryk University, Brno 22

Next?

• The next step is to study how chemistry evolves with stellar age in our sample. We then have a powerful tool for using observed stars to inform theory about how both fields and atmosphere chemistry depend on mass and evolve with time

• This should be very helpful in using observed stellar characteristics to guide and test ideas about the interaction of various stellar hydrodynamic and magneto-hydrodynamic processes, our initial goal