Transcript White Dwarfs - Indiana University

History of White Dwarfs
•
Bessell (1844)
•
Alvan Clark (1862)
•
Procyon B found in 1895 at Lick
•
Spectrum of 40 Eri B observed – an A
star!
•
– Proper motions of Sirius and Procyon
wobble
– Suggested they orbited “dark stars”
– Found Sirius B at Northwestern’s
Dearborn Observatory
– Was it a star that had cooled and
dimmed?
– It must be hot
– Must have small radius to be so faint
– The first “white dwarf”
Adams found Sirius B is also an A star in
1915
– From luminosity, R~ 2 x Earth (actually ¾)
– From orbit, about 1 solar mass
– Density 105 x water (actually 106)
20th Century History
• Eddington
– Gas must be fully ionized so that nuclei could be compacted
together
– Conundrum – as the white dwarf cools, the atoms should
recombine, but they can’t because the star can’t swell against
gravity
• R. H. Fowler (1926)
– Recognized the role of degeneracy pressure in supporting the
star
• Chandrasekhar (1935)
– Upper limit to mass supported by electron degeneracy pressure
due to limit of velocity of light (1.4 solar masses)
• Zwicky (1930’s) - Degenerate Neutron Stars
• Schatzman (1958) – chemical diffusion in strong gravity (plus
radiative levitation, winds and mass loss, convective mixing,
accretion)
• Greenstein and Trimble (1967) - Gravitational redshift
• Hewish and Bell (1967) - Pulsars
Interiors in a Nutshell
•
•
•
•
•
•
•
•
•
•
•
Upper mass limit for white dwarf formation is somewhere between
5-9 solar masses – “Inside every red giant is a white dwarf waiting
to get out” (Warner)
Most have C-O cores, most massive may have O-Ne cores
In hot, pre-white dwarfs, neutrinos dominate energy loss
When nuclear burning stops, photon cooling dominates
interior becomes strongly electron degenerate, mechanical and
thermal states decouple, ions are a classical ideal gas
Ions eventually crystallize but we still have no empirical evidence
for this
Crystallization releases latent heat and carbon and oxygen may
undergo a phase separation on crystallization may also provide heat
which would prolong cooling times
after crystallization, heat capacity drops, cooling times shorten
Interplay of gravitational settling of heavier species and turbulent
energy transport (convection) may affect surface abundances
As the degeneracy boundary moves outward, it eventually halts the
convection
At cool enough temperatures H2 forms, and possibly even He2
Masses of White Dwarfs
• Methodology
– orbital solutions or binary stars
– measurements of surface gravity (with a massradius relation)
• model atmospheres with photometry, parallaxes
• gravitational redshifts
– asterseismology
• <M> = 0.58 – 0.59 solar masses
• About 1/6 of (presumed) single white
dwarfs show radial velocity variations
White Dwarfs
•
•
White Dwarfs – DO, DB, DA, DF, DG, DM, DC
Classifications NOT analogous to MS – reflect compositions, not temperature
•
•
Heavier atoms sink in gravitational field
Above 15,000 K, 15% are non-DA, below 15,000 K, half are non-DA. How do the
stars do that?
NO DB stars between 30,000 and 45,000 K
•
–
–
–
–
–
DA – hydrogen lines (no other lines, pure H atmosphere)
DB – neutral He lines (no hydrogen at all, pure He)
DO – ionized He lines (no hydrogen at all, hotter DBs)
DC – continuous spectrum, no lines
DF, DG, DM (can’t discriminate DA or DB)
Surface Compositions
• DA (80% of WDs) and non-DA
• Most WDs have pure or nearly pure H or
He atmospheres
• DAs found from hottest to coolest
• Non-DAs start with hot stars
– DOs for Teff > 45,000K with He II or He I and
He II
– DBs for Teff < 30,000 with He I only
– DCs (featureless) for Teff < 11,000
– NO He-rich WDs between 45,000 and 30,000K
Why the DB Gap?
•
Simple picture of parallel sequences of H and He-rich objects
doesn’t work
– Accelerated evolution of DBs between 45,000 and 30,000K doesn’t
make sense
– Change in ratio of DAs and DBs around 10-15,000K also hard to explain
– Mean masses of DAs and DBs are the same
•
Theory of spectral evolution – Fontaine and Wesemael
•
But this model doesn’t work
– All WDs have a common origin (PNN) with some hydrogen, upper limit of
10-4 solar masses to 10-15 solar masses of hydrogen (recall that 10-4 is
the limit where H burning stops)
– Only about 10-15 is needed to produce an optically thick H layer at the
surface
– Diffusion brings H to surface; by Teff=45,000 K, all WDs have
hydrogen atmospheres, so there are no DBs
– At 30,000 K, the formation of an He ionization zone creates turbulence
which mixes the H with He, and leads to He stars (stars with more than
10-13 H have too much H to form a sufficient convection zone, and they
remain DAs)
– Change in DA/non-DA ratio at 11,000 K results from onset of convection
from H ionization zone, increases mixing, and more DBs appear
Spectral Evolution Model
• What’s wrong with the spectral evolution model?
– model suggests DAs should have a wide range of
hydrogen layers, from 10-4 solar masses to 10-13 solar
masses of hydrogen
– Asteroseismology results suggest all DAs have thick
hydrogen layers
– The model also predicts trace amounts of H in the
hottest DB stars (just at the cool edge of the DB gap)
– H was found with GHRS on HST but at a level way to low
(<10-18 solar masses) to have ever permitted this DB to
have been a DA in the DB gap
• The WDs are fed by other sources than PNN
– subdwarf O and B stars (whose origin is still not clear)
– IBWDs (interacting-binary white dwarfs)
• Both of these enter the cooling curve somewhere
along the spectral sequence
• Maybe the DBs come from the IBWDs, and all the
DOs become DAs at 45,000 K (and stay that way)
Variable White Dwarfs
• Asteroseismology with the Whole Earth Telescope
(WET)
– Determine masses, hydrogen masses, rotation rates,
magnetic fields
• ZZ Ceti Stars – extension of Cepheid instability
strip
–
–
–
–
Hydrogen ionization zone below the photosphere
Temperature range from 10,500 to 13,000 K
Amplitudes of 0.01 to 0.3 mag
Periods of 3-20 minutes
• DB pulsators from He ionization zone
– T ~ 20,000K
• PG 1159 stars – also pulsators
– T ~ 130,000
– Oxygen ionization zone drives pulsations
– Periods of minutes
Rotation
• Measuring rotation rates (vsini)
– shapes of hydrogen line cores
– rotational variation of polarization in magnetic white
dwarfs
– asteroseismology
• vsini measurements suggest rotation rates < 8-40
km/sec – very slow! (where does the angular
momentum go?)
• Periodic changes in polarization gives two groups,
those with rotation periods of a few days, and
those with periods >100 years
• Asteroseismology also gives slow rates
Class Problem – What is the approximate rotational
velocity of a star with a rotational period of 2
days? (assume we are observing it in its equatorial
plane)
Why Is Slow Rotation a Problem?
• Assume a solar rotation period of 30
days, conserve angular momentum,
and estimate the rotation rate if the
Sun were shrunk to the radius of the
Earth...
Magnetic Fields
• Broadband circular polarization
•
•
•
•
•
– detects fields > 50 MGauss
Circular spectropolarimetry
limits to 1-50 kG
Detected fields range from 3 kG to 1 GG
asteroseismology suggests fields of 1 kG
Magnetic fields detected at all
temperatuers, but more and stronger
fields in cool WDs (<16,000K)
• Does a dynamo form when convection
starts?
• Oblique rotators again?
Neutron Star Oddities
• The non-pulsar neutron star (Geminga)
–
–
–
–
–
discovered from x-ray brightness
imaged by HST in 1998 (V~25)
distance <~400 pc (in front of IS cloud)
Teff > 106 K
Probably lots of these around
• The Black Widow Pulsar
– Eclipsing double, companion R=0.2 RSun, mass of 0.02 MSun
– Mass transfer spins up pulsar
– Pulsar is eroding away the companion
• The Magnetar
– Magnetic field of 1014 G
– field cracked pulsar’s crust, producing gamma and x-ray
burst
– burst partially ionized the upper atmosphere of Earth
• Quark Stars?
Ages from White Dwarfs
• Age of the disk – from the coolest WDs found
• Liebert, Dahn, and Monet (1988, etc) used a
sample from the Luyten Half-Second catalog
• Oswalt from common proper motion binaries
• Observational problems
– completeness
– undetected binaries
– small sample statistics, especially for the coolest,
faintest white dwarfs. Need larger samples!
• Many remaining issues in WD cooling physics
– C/O ratio in core, phase separation at crystallization
– Settling of heavier species (22Ne, 56Fe)
– depth of He layer
• Age estimated around 10 Gyr
• Age of the halo and globular clusters still to be
done
Degenerate Binaries
• Novae
–
–
–
–
–
10 magnitudes or more increase in brightness over a day or two
Drop typically 3 mags in a month or two
Back to original brightness after a few years or decades
White dwarf – low mass main sequence binary
Began as wider binary, then common envelope evolution tightens
the binary
– Recurrent novae, dwarf novae
• Symbiotic Stars – binary separation sufficient that stars
don’t interact until companion becomes a giant
– Spectrum is a cool star + hot accretion disk
– Mass loss from giants feeds an accretion disk around the white
dwarf
– Nova-like eruptions – due to white dwarf mass accretion or to
instabilities in the accretion disk
• X-ray binaries – neutron star + companion
R Corona Borealis (and other) Stars
• RCorBor Stars & Extreme Helium Stars
–
–
–
–
–
–
–
–
A to G-type supergiants
Occasionally dim by ~8 magnitudes
Recovery can take a year
Veiling by carbon dust from mass loss
Highly deficient in hydrogen
Helium dominates
Carbon greatly enriched
Cepheid-like pulsations
– Heating by 30K per year, shrinking
• Post-AGB stars?
• Coalesced white dwarf binaries?
White Dwarf Merger Scenario
•
The camera aspect remains the same, but moves back to keep the
star in shot as it expands. After the star reaches 0.1 solar radii, an
octal is cut away to reveal the surviving disk and white dwarf core.
The red caption (x) is a nominal time counter since merger. A rod of
length initially 0.1 and later 1 solar radius is shown just in front of
the star.