PDA

View Full Version here: : BL Tel -A slow eclipsing binary


Terry B
25-06-2013, 09:38 PM
Dear All
BL Tel is an interesting binary star. It has a very long period of 778 days. Most eclipsing binary stars have periods between half a day and a few days. At the period increases the chance that binary star happen to allign up so we see an eclipse becomes very small. This makes BL Tel very unusual. The stars are currently undergoing an eclipse. The last eclipse was in May 2011. I took photometry images of the 2011 eclipse and a grapg of the eclipse is below.
For this eclipse I am also taking spectra. The 2 stars are different spectral classes. The brighter star is an F4 star and the eclipsing star is a M star. In the uneclipsed state, the F star spectra dominates but during the eclipse, the M star spectra starts to become visible. The stars will also become more red. I have attached a spectra showing the stars before the eclipse on 29/5 and one from last night. The eclipse is not at maximum yet so should the difference should become more apparent in a few weeks.

The red graph is vert typical of a F3 star with deep Ca II lines called K and H between 3900 and 4000A. Ha, Hb and Hg are also clearly visible.
In the blue graph, the overall colour has shifted to the red. The Ca lines are weaker and there are brouad absorptions visible that are caused by TiO.
I will continue to take spectra weather permititng to hopefully show the further changes in the spectra.

Terry

DavidTrap
25-06-2013, 09:48 PM
Fascinating stuff Terry.

Where do you find such targets?

DT

Weltevreden SA
26-06-2013, 07:18 AM
Thanks for the alert and spectra, Terry. Plus, your interpretations are very informative. I'm curious about a couple of things. In the AAVSO eclipse plot, starting about 04/20 the 'V Prevalidated' brightness seems to be a consistent ~0.2 mags fainter than the 'Visual Prevalidated'. Is that a real difference or a property of the technical definitions of 'V' and 'Visual'. Spectrally, is there a bandwidth distinction between the two?

Second, could you recommend a source for interpreting bands by star type, atomic mass, and isotope? You mentioned the broad TiO band in the M star. If I were to see the spectra without your interpretation I wouldn't offhand know which band is the TiO and what constitutes broad or narrow absorption. It would be handy to see a spectrum and have a resource by which could identify a particular abs or emiss feature. This comes up a lot as I read the literature and become curious about a band that isn't discussed in the text.

Thanks, Dana in SA

Terry B
26-06-2013, 08:53 AM
Dana
Take a look at this page
http://www.ursusmajor.ch/astrospektroskopie/richard-walkers-page/index.html
Scroll down to the "English documents" and download the spectroscopic atlas and the interprestaion document. Both are great resources that will explain lots.
The difference of 0.2 between the V and Vis measurements os within the error margins. The Vis results are from 1 observer only and this is a pretty good result. There is more error when the stars are very red but this star is not red.
Cheers
Terry

Terry B
26-06-2013, 08:54 AM
Thanks
The target came up on the VSS alerts in 2011 and also this year.

Terry

Weltevreden SA
27-06-2013, 08:01 AM
Hi again, Terry . . . Thanks for the ursusmajor.ch link and the info I found there. Now I'm wondering where the Ti comes from in an M star. I assume you mean an M red dwarf and not, as I occasionally see, a red giant whose color temp is the same as an M dwarf but is 5 - 6 mags brighter. A red dwarf core converts via the p-p chain and I'm wondering where the spare neutrons come from to generate Na, Si, all the way to Ti. I looked up Ti synthesis via CNO cycling and it comes to ~2.4 Myr and requires T on the order of ~40 million K. So far as I know, M dwarfs don't reach the CNO cycle temps. Does the Ti come from several prior generations of Type 2 SN that the M dwarf accreted in the pre-MS phase? Does your spectrum show significant SI2, Na, Al, or Mg lines? The last 3 would come via the Ne/Na and Mg/Al cycles, which occur via He burning in the horizontal branch, massive RG, or AGB stars. Your spectrum has prompted some fun head-scratching over here, but so far all that has fallen out is (yet more) grey hairs. Thanks . . . Dana in SA

Merlin66
27-06-2013, 08:24 AM
Dana,
Walker's Atlas is very good!
The other references you could consider:
Kaler's "Stars and their spectra"
Gray & Corbally " Stellar Spectral Classification"
Tennyson's "Astronomical Spectroscopy"
and
D F Gray's "Stellar Photospheres"

Terry B
27-06-2013, 09:04 AM
The Ti and all of the heavier elements came from prior supernova and are not made in the star that we are observing. Thre are lots of lines of "metals" in our sun as well but none have been made in the sun.
The spectra of a M dwarf vs an M giant is quite similar even thought they are completely different stars. The reason that TiO becomes significant even though it is quite an uncommon element is that it is a very good absorber of light- hence the reason we use it as sunscreen.
At high temperatures, TiO is ionised along with lots of other elements. This is why hot stars like A and B stars have very few absorption lines (except hydrogen).
Cooler stars allow molecules to form near the surface and these absorb the blackbody light continuum resulting in absorption lines. TiO is just very good at absorbing light so it is very promonent in the spectra.
Na lines are also visible but I haven't identified then specifically.
Cheers

Terry

Weltevreden SA
28-06-2013, 08:34 AM
06/28/2013:

Hi again, Terry . . . OK, the SN origin of the TiO abundances is clear now. I figured your ‘M’ identification probably referred to Class VII dwarfs and not the similarly coloured Class I to II supergiant RGs. The sawtooth-shaped diffuse lines and lower luminosity than the K star pointed to a dim M dwarf, not an M supergiant 6+ mags brighter and sporting a multitude of crisp skinny lines.

Alas, I now have even more questions than I started with. (I understand this business of one answer leading to two questions started around the time of Archimedes and has led mainly to the building of ever-larger telescopes.) If Ti and other heavy metals, e.g., VaO, are abundant via accretion from earlier SN ejecta, do these metals convect into the M star’s interior? I ask because M stars are wholly convective <0.35 Msolar, which lifts He as high as the photosphere and would show up as He lines in your spectrum. Given the line broadening I can’t identify any, compared with the Lalande 21185 and Wolf 359 sequences in the 3900-7100 nm range. If accreted metals are convected inward the same way He is convected to the surface, the cores of such stars would expand as opacity and temp increases, thereby shutting off the transport mechanism. Would you happen to have your M dwarf’s B-V and/or U-B color indexes? I am curious about the star’s He lines because if they are present it would point to a Msol of <0.25 and it would end up a He dwarf.

Unfortunately the 04/2013 Walker ‘Guide to Spectral Classes’ in your link provides two misleading spectra under the rubric ‘Spectral Class M’ starting p.87. The text refers to stars Msol 0.5 down to 0.08. but the spectra in Table 60 are two red supergiants, Antares and Alpha Herc. Walker doesn’t provide a spectrum for an M dwarf; I looked one up in Rojas-Ayala et al 2011 (http://arxiv.org/abs/1112.4567).

On the other hand, the TiO bands in the Alpha Herc spread between 4000 and 7000 point to TiO’s impressive ubiquity. To look at the r-hand side of Chart 60, one could conclude the whole dadgum star is hot TiO.

All this is fascinating. Thanks for the tip that led to the quest. Somehow Archimedes doesn’t seem quite so ancient any more.

=Dana in SA

madbadgalaxyman
28-06-2013, 08:36 PM
Hi Dana, and all of our other illustrious 'spectro-people' ,

I am really enjoying these detailed spectro-discussions as I am definitely a 'babe in the woods' when it comes to spectroscopy.

An up-to-date and extremely detailed professional-level reference book about the observed spectra of stars of all spectral types, and about spectral classification, is:

"Stellar Spectral Classification"
by Richard Gray and Christopher Corbally
It was published in 2009, in the Princeton Series in Astrophysics.
(ISBN: 0691125112)

( Merlin66 = Ken also recommends this book, as per his previous post in this thread. )

The paperback version of this book is cheaper, and still quite physically robust (not a cheap and nasty binding)

As a galaxies-specialist who needs to understand relevant aspects of stellar astronomy, I find this to be a clearly written, useful and accessible Reference Work for anyone from advanced amateur astronomers through to professional astronomers;
this book is much more about the observed spectra and about how they are classified, than about a heavy physico-mathematical treatment. It essentially brings the Morgan-Keenan system of stellar spectral classification up-to-date, by using recent data and by including the more recently discovered spectral types.

For those of you who know the wonderful semi-popular book about the spectral sequence by James B. Kaler ("Stars and Their Spectra"), this book somewhat resembles a much more detailed version of the Kaler book.

Cheers,
Robert

Terry B
29-06-2013, 05:02 PM
Richard Gray's book is also available online here in an abridged form.
http://ned.ipac.caltech.edu/level5/Gray/frames.html

Terry

Weltevreden SA
30-06-2013, 08:09 AM
Good old Level 5 comes through again. I hadn't looked into the Stellar Spectra part of its huge list of goodies before. Thanks to your suggestion, I downloaded the abridged Gray survey. Alas, it's a little too abridged. Now I have to start saving the pennies to buy the original, which in SA will take a lot of pennies.

I have a couple of questions. First, I notice in Ch. 11 that the bottom three stars, Vega, A Leonis, and HD 23194, all have broad-sloped lines in the Hbeta at ~4100 and Hdelta ~4348 bands. These shoulders are not so pronounced in other bands in the same spectra. Only a few other spectra, e.g. 23585 in Ch. 18, exhibit this feature, one of the lines being the ~4100 Hbeta. Would broad shoulders like this indicate high rotation rates in the photospheres?

Since the pattern is not mimicked cross all the bands, I wonder if we have a case of magnetic field circulation in the stellar photospheres. In thick stellar atmospheres, an upper altitude (not core-to-surface) convection cycle can develop, which is analogous to the Earth's upper atmospheric weather pattern cycling from the equator to the poles and back. As with the earth and its weather, stars with polar circulation are affected by the Coriolis Force, which in stars produces magnetic turbulence akin to our earthly weather fronts. The effect would be greatest in mid-latitudes and least at the equator and poles. The effect on spectra could be variable line smearing.

Second, starting with the bottom 3 spectra in Ch. 24 and becoming notable by Ch. 26 HR 3577 and BK Vir, the gradual-rise, sharp-drop sawtooth effect I mentioned yesterday becomes very evident. This becomes even more evident in Ch. 29 with S and R Leonis (though HR 3577 and BK Vir are repeated as well). Other lines in the same spectra don't feature this sawtooth effect. Also, the sawtooth lines are the strongest of all lines in all the affected charts. This is a feature common to most M dwarfs. It can't be due to rotation because other lines don't repeat the pattern, but I can't figure out what might cause the behaviour.

In Ch. 31 V Cancri has a remarkably spiky, bright Hbeta emission line. This appears again in both spectra in Ch. 34. What would cause such a bright peak in Hbeta that isn't reflected in other spectra or even in other lines in these same spectra?

An observation: the He-rich subdwarf in C.33 is probably a He white dwarf dropping off the bottom of the Horiz Branch "Blue Tail" into true WD territory. These are unusual stars and there are not many of them. They are normal Horiz Branch blue stars, all of which have a core mass of 0.487 Msol, but have lost their entire H atmospheres and become almost totally He. They have no more heat input from nuclear processes (too cold for Carbon ignition) so cannot ascend the AGB branch and become planetary shedders like most pre-WDs. They are produced mostly in low-metallicity, and therefore low-opacity, GCs, which convert atmospheric mass into solar wind more easily. Compare this spectra with Barnard 26 in Ch. 32, which is a former Horiz Branch star that retained enough of a He atmosphere to evolve upward into the AGB, where it is now busily becoming a planetary-nebula parent.

Thanks again for all the links. I already have Kaler's 'Stars and Their Spectra', so the Gray is a welcome expansion into the territory. It's fun being introduced to new stuff.

=Dana in SA

Merlin66
30-06-2013, 08:42 AM
Dana,
The questions you raise are associated with the modelling of the star and it's photosphere....

The definative work in this area is "The Observation and Analysis of Stellar Photospheres" by David F Gray.
to quote
""Topics covered include radiation transfer, models of stellar photospheres, spectroscopic equipment,....and techniques for measuring stellar temperatures,radii, surface gravities, chemical composition, veleocity fields and rotation rates""

Andrew Pearce
08-07-2013, 03:49 PM
Hi All

BL Tel appears to have reached it's minimum. I've had a very good run with weather over here in Perth and have managed a number of observations. It has gone down to magnitude 9.3 over the last few nights and this corresponds to the prediction for mid eclipse of July 6th as deteremined by Peter Williams. Attached is a light curve plot from the AAVSO web site.

Regards
Andrew Pearce
Perth, WA