Commentary – hot stars (abundance anomalies)
Hot stars (hotter than early F stars on the main sequence) generally differ from the cooler stars in that the outer layers transport energy by radiation rather than by convection (mixing) as with the cool stars. For those hotter stars that express a strong magnetic field (among these are the peculiar A stars or Ap stars), the star rotates like a solid in the sense that the equator rotates at the same rate as the polar regions unlike the cooler stars where there is a difference between the polar and equatorial rotation rate (differential rotation). The Ap stars are named peculiar because they have very unusual apparent abundances at the level of their surfaces (stellar atmospheres) compared with the solar abundances and the lines representing these elements in the spectrum vary in strength as the star rotates indicating that the unusual abundances are distributed unevenly over the stellar surface. Only about 10% of the stars in the A and B star classification are of the chemically peculiar type. The unusual appearance of the various elements is thought to arise through a process where radiation pressure (light pressure) drives specific atoms from the deeper regions of the star up into the atmosphere where the spectrum lines are formed. The fixed pattern of the distribution of elements at the surface is thought to arise because the magnetic field guides the rising elements into the perceived pattern. Because the magnetic fields are almost certainly rarely a simple dipole field and the process of guiding the elements may vary as the star ages, the patterns of element distribution are rarely simple. A rare example of a simple pattern is the oxygen distribution on the star epsilon Ursae Majoris (Alioth) as shown below.
Epsilon Ursae Majoris and its extreme surface Oxygen distribution!
This star is an Ap magnetic variable star with common name Alioth, the star in the handle of the Big Dipper that is closest to the bowl. It has had a variation of the Doppler imaging code applied to numerous spectra taken of the oxygen triplet of lines at 7775 angstroms. The map of oxygen generated for this star from the Doppler imaging code shows that the oxygen abundance is typical for a normal star (i.e. the Sun) in a ring about what is known now to be the magnetic equator of the star but the oxygen is depleted by a factor of about 100,000 in large areas centred on the magnetic poles. (See Astronomy and Astrophysics 326, 988, 1997). This is probably about the best and simplest map of the abundance of any element on an Ap star yet. The simplicity of the abundance map could arise from a variety of factors that are characteristic of this star. Alioth is evolved somewhat more than the average magnetic Ap star since it is now a short distance off the main sequence (in other words, slightly depleted in its core hydrogen fuel). Further, the magnetic field is rather small for a typical magnetic star being only around a global field of about 100 Gauss. The final characteristic is that it seems the field itself is globally close to a simple dipole field. A video of this star is available as an MPEG (epsilon UMa). Note the variation in the shape of the triplet of oxygen lines displayed at the bottom of the video showing how they vary as the star rotates.
Now we note the distribution of the element Chromium on epsilon UMa for comparison with the distribution of Oxygen. See the video of the Cr distribution as presented in the oxygen paper cited above (an image determined from the Cr 4558 line). If you compare carefully the images by stopping the animation for each at specific phases starting phase, quarter rotation etc., you will see that the Cr abundance is at a minimum roughly along the same belt that O is at a maximum. Thus it appears that Cr is minimized along the magnetic equator for epsilon UMa.
The image of oxygen on the surface of Theta Aur represents an example of a star with a more complex magnetic field than epsilon UMa. The maximum of observed positive magnetic field occurs near phase 0.82 (295 degrees longitude on our image). The negative maximum (and therefore the rough location of the negative magnetic pole) occurs near phase 0.38 or 135 degrees of longitude. If the image of Cr abundance for theta Aur by Hatzes (MNRAS 248, 487, 1991) is used as a guide, the symmetry suggests the positive pole then is about latitude negative 55 degrees with longitude near 295 degrees and the negative pole is about +55 degrees latitude and longitude 135 degrees. (See a postscript image of the Cr abundance). For the Cr distribution, strong Cr abundance occurs near the positive pole with weak abundance in a somewhat sub equatorial belt and a maximum of Cr abundance at the negative pole. It is difficult to relate the Oxygen distribution shown in the video with the Cr abundance because of the tendency to see the bright colour of the oxygen video (signifying strong oxygen) as though it should be compared with the dark (low abundance) part of the Cr map. It is easier to see the comparison using the postscript figure of the oxygen distribution of our paper which is laid out much like the Cr map. The oxygen map shown in postscript here is much more complex than for epsilon UMa. We wrote in our paper on the distribution of oxygen on theta Aur (Astronomy and Astrophysics 424, 237, 2004) that this oxygen image shows a double spot of pronounced oxygen depletion at roughly longitude 135 and latitude 55 where the negative magnetic pole occurs and the double depletion spot is surrounded by a network of roughly solar abundance. Note the extended peak of Cr abundance at this pole. At roughly longitude 270 to 300 degrees and latitude -30 (barely visible) we see another elongated zone of very depleted oxygen where we would expect the positive magnetic pole . Once more the Cr is enhanced at this pole and also once more the depleted region of oxygen may be surrounded by some spots of more normal oxygen. While this pattern is far more complicated than epsilon U Ma, we might see some similarity if we look at the two zones of more normal oxygen surrounding the two poles as more like a double band of normal oxygen on each side of the magnetic equator. Much as with epsilon U Ma, it seems as though the Cr map is close to a negative of the oxygen map.
The overall conclusion of our paper was that the magnetic field for theta Aur probably has a strong quadrupole component in contrast to the field of epsilon U Ma where we appear to have a global field that is primarily dipole in nature.