R. Chary , B. Zuckerman , & E. E. Becklin
Division of Astronomy & Astrophysics, University of California, Los Angeles, CA 90095-1562, U.S.A.
The presence of metal absorption lines due to Mg, Ca, Fe etc. in the
photospheres of cool (T
eff<20000 K) hydrogen atmosphere DA-type white
dwarfs has been an unexplained problem for some time. The lifetime of the
elements before they diffuse through the photosphere is in the range of a
few days to thousands of years depending on the mass and temperature of
the white dwarf. Hence, it appears as if the metals are being supplied to
the white dwarfs on relatively short timescales. The existence of metals
in the atmospheres of much hotter (T
eff>50000 K) DA white dwarfs has
been explained through the theory of radiative levitation whereby the
metals are suspended in the photospheres by the radiation pressure of the
white dwarf. However, this is not feasible for the cooler white dwarfs
because of the diminished radiation pressure. It has been demonstrated
that the detected lines are associated with the white dwarf and not
produced by the intervening interstellar medium (ISM). For example, since
the MgII 4481 Å line observed in WD1337+70 has an excited lower level,
it must be photospheric in origin (Holberg et al. 1997, ApJ, 474, L127).
This evidence suggests that the metals are being accreted from the
interstellar medium by the white dwarfs. Dupuis et al. (1993, ApJS, 84,
73) showed that the metal abundances can be explained by episodic
accretion events whenever the white dwarf travels through relatively
overdense clouds with n
10 cm-3 at velocities of 
20
km s-1. However, most of the detected metal line white dwarfs are
within 50 pc of the Sun and recent maps of the sodium D lines towards
stars with Hipparcos distances within 70 pc does not show evidence for
such overdense clouds within the Local Bubble (Welsh et al. 1998, A&A,
333, 101).
The presence of infrared excess (Zuckerman & Becklin 1987, Nature, 330, 138) around the white dwarf G29-38 followed by the detection of metal lines in it's spectrum (Koester et al., 1997, A&A, 230, L57) provided the first clues to resolve this issue. The infrared excess is thought to arise from particulate matter orbiting the white dwarf (Graham et al. 1990, ApJ, 357, L21). To determine if there is a correlation between infrared excess and metal lines, we proposed extremely sensitive ISOCAM mid-infrared observations of 12 nearby white dwarfs, including G29-38, which show metal lines in their optical/UV spectrum. We find that none of the white dwarfs other than G29-38, shows a significant infrared excess over the photosphere and are able to place strong upper limits on the existence of dusty disks around the white dwarfs. We conclude, based on these observations, that ongoing accretion of a dust disk seems to be an unlikely scenario to explain the existence of metals in the photospheres of cool DA white dwarfs and suggest possible alternatives to explain this phenomenon.