Jump to content

Banner.jpg.b89429c566825f6ab32bcafbada449c9.jpg

jsandse

Members
  • Posts

    326
  • Joined

  • Last visited

Blog Entries posted by jsandse

  1. jsandse
    Hi all,
    Time for another blog.
    If anyone is interested in my blogs please
    comment or I may think I am speaking to
    myself.:eek::eek:
    Anyway Wolf-Rayet stars - what are they
    what do their spectra look like and
    moreover why are they interesting?
    Wolf-Rayet stars are extemely hot stars
    (surface temperatures 25000 to 50000
    celsius compared to the suns cool 6000
    celsius) which have evolved from hot O
    stars - O stars are the ones that look blue
    when you look through the telescope.
    First the O star expands to become a red
    giant as it starts to run out of hydrogen
    and then it evolves to become a Wolf-
    Rayet star when it burns heavier elements
    in its core.
    So O stars as they start to die (run out of
    hydrogen to burn) evolve into Wolf-Rayet
    stars. The stars then spend 10% of their
    lives in this Wolf-Rayet state before they
    finally go bang - and I mean BANG!!! -
    when we end up with a
    supernova/gamma ray burst. Apart from
    the big bang thats near the biggest bang
    you can get.
    Another thing about Wolf-Rayet stars is
    that they are rare - approximately 300
    found in our galaxy so far and they are
    not bright so spotting them is difficult
    In the northern hemisphere the brightest
    are just under magnitude 7 which means
    unless you have a very dark site then
    you'll need a telescope just to look at
    them.
    Because they are dim and as they are rare
    it wasn't until Charles Wolf and Georges
    Rayet discovered some of them in Paris in
    1867 - thats under 150 years ago.
    Thats enough history - the thing that
    excites people like me is their spectra.
    These stars instead of having the usual thin
    dark absorption lines that our more
    common stars have, have extremely wide
    and bright emission lines!
    The reason for this is that the stars are
    suffering from extreme mass loss through
    the large stellar winds they have (500-
    2500km per second) - these winds are
    powered from the burning taking place in
    the core of the star.
    There are two main types of Wolf-Rayet
    star the Nitrogen type called WN
    characterised by it helium and nitrogen
    emission lines and the carbon type called
    WC characterised by its strong carbon,
    helium and oxygen emission lines.
    The type of the star really depends on
    what the burning process is that is going
    on in the core of the star.
    For the WN type there is still some
    Hydrogen left in the core and it combines
    with carbon to eventually produce Helium
    and Nitrogen. I'll save you from all the
    equations but this is called the CNO
    process
    For the WC type there ain't the hydrogen
    there so we have full on Helium burning in
    the star so the Heliums fuse together to
    form eventually Carbon. Depending on the
    conditions in the core of the star we can
    sometimes get Carbon fusing with Helium
    to get Oxygen. This is called the triple
    alpha process (an alpha particle being a
    helium nucleus hence the name).
    Ok now for the star I took the spectra of.
    For all of you out there who like beautiful
    images here is a link to a picture of WR
    136 taken by the Isaac Newton telescope
    showing the star and its surrounding
    nebula
    APOD: 2009 September 15 - NGC 6888: The Crescent Nebula
    Ok now time for the spectra - this star is
    mag 7.5 so its a pretty challenging target
    I took a couple of half hour exposures to
    get this spectra which was not like any I
    had taken before:

    And the line profile from the spectra is
    here:

    Ao we have a couple of very broad
    emission lines. So what are they?
    Fortunately several professional
    astronomers have done extensive
    examinations of Wolf-Rayet stars in
    general and this star in particular. One
    called Hamann in 1993 produced a model
    for the star showing that it was mostly
    Helium with the remainder being 12.5%
    hydrogen and 1.5% nitrogen.
    So the two emission lines predicted from
    the model are for the bright one a mixture
    of Helium and Hydrogen (mainly helium)
    and for the second line Helium alone.
  2. jsandse
    Hi all,
    In my last blog I mentioned that you can generate spectra from "man-made" stellar models. In this blog I will talk a little bit more about my experience with these models.
    Just to rewind a bit say you have gone to the bother of capturing a stellar spectra. So you have this spectra with a number of lines in it. So what? what does it mean its just a bunch of lines - not very exciting...or is it?
    An example of the spectra you actually capture looks like the ones I took of Mizar in my album
    http://stargazerslounge.com/members/jsandse-albums-spectra.html
    That picture actually has 6 spectra in it which are aligned one on top of the other.
    But you get what I mean what can I do with this spectra?
    Well you can convert it into a line profile using some software (I use vspec) and then calibrate it so that you know what the wavelengths of the line in the spectra are.
    This calibration can be done using a spectra with a reference lamp with known lines in it.
    Ok so you have a line profile of a spectra with a number of dips in it corresponding to the dark lines (also called absorption lines) in the original spectra. An example of this
    is the one in my album of epsilon auriga
    What is causing these lines and what does this tell me about the nature of the star I am looking at? I'll just answer the first part of the question in this blog. Leave the second part for another time.
    What is causing these lines are atoms on the surface of the star absorbing some of the light before it escapes from the star. This part of the star is called the photosphere. It's thin the sun's photosphere is only 1000km thick.
    Each type of atom (element) absorbs light of a unique characteristic wavelength. Anyway what this means is that if I know the wavelength of a dark line in my spectrum then I can in principle find out the atom responsible.
    So back to stellar models. Given three basic properties of a star:
    a) temperature
    b) logG - surface gravity of the star - this is how heavy things are on the surface of the star (eg the surface gravity on the moon is less than that onthe earth - just look how long those astronauts can jump and hit a golf ball!)
    c) metallicity - ratio of amount of metals (to astronomers this means any elements that are not hydrogen or helium - crazy isn't it?) to amount of hydrogen in the star
    a stellar model can calculate for most stars
    the physical properties of the star (pressure, temperature at different depths in the star) and generate a stellar spectra for it.
    I won't go into how we find out what the temperature, surface gravity and metallicty of the star is here - thats another blog...
    I will just say that you need these three properties to tell you what types of atoms are in the stellar photosphere (mainly comes from metallicity) and how they behave - which depends on temperature of course and the pressure they are under ( this is related to surface gravity)
    Right so I have captured my spectra of Vega, I know what its temperature, surface gravity and metallicity is.
    I can then use some software (I use Atlas) to generate a model for it and to produce the spectra from it.
    Now comes the exciting part of comparing the spectra I took with the model
    In my album please look at the picture at this link:
    http://stargazerslounge.com/members/jsandse-albums-spectra-picture8198-comparison-detail-spectra-vega-i-took-against-synthetic-spectra-produced-atlas-stellar-model-vega-similarity-between-two-spec
    So in the picture I have placed two line profiles next to each other. One in red which was generated from the model. And the other in black I captured myself from Vega.
    It is remarkable how similar they are - virtually all the absorption lines in the model correspond with those that I had captured from the real star.
    The benefit of using the model is that it gives me the atoms that cause the absorption lines. I have annotated the picture with details of some of these elements for the more prominent lines - includign ones caused by Hydrogen (H), Iron (Fe), Magnesium (Mg) and Titanium (Ti).
    How cool is that! I can decode the spectrum of a star that is 150,000,000,000,000.000 miles away into the types of atoms that are absorbing the light at its surface.
  3. jsandse
    Thought I would take the opportunity to provide some notes of my experiences in
    astronomical spectroscopy to date with current projects carried out and others I would like to do in the hope it may help/motivate other people.
    My simple aim when I started out on astronomical spectroscopy was to take
    spectra and to understand how the spectral lines in stars are formed. So thats
    two objectives the practical side of doing spectroscopy and the theoretical side
    of finding out why things are
    the way they are.
    Sounds simple enough but when you start delving into the subject
    there is a lot to learn - however taking the spectra is far easier
    than understanding why the lines are formed.
    So I started with a couple of projects:
    - before going near the sky calibrate the spectroscope using the internal
    calibration lamp on my L200 and also some external
    calibration bulbs - I used the ones from Habitat detailed on Buils website at
    http://astrosurf.com/buil/calibration/lamp1.htm
    - take spectra of a bright star across its continuum and the sun/moon and learn
    a spectra data reduction process to process the spectra
    - take spectra of a spectroscopic binary to see the radial motion of
    the stars - my results are here for this task:
    http://f1.grp.yahoofs.com/v1/QHP_TEwOlSOe0oy6uHI3PfjBV0-yD2UBS3dWed50Kqlz8CIAbID\
    3q6UT9y4lCnNvjP3LxcdcPxtdwnEi7phTlstrhkgGiw/John_s/mizargrouporiginal.jpg
    - take spectra of epsilon aurigae and submit to the current campaign
    that Robin is coordinating
    What I can say about the projects is that I have managed to complete
    them all. However there is room for improvement for what I have done:
    increasing the resolution of the spectra I have taken by going to
    second order on my L200 or using a higher resolution grating - would get me to R
    ~ 15000 and would give me more a chance to get reasonably accurate measurements
    of periods of binaries and give me more accurate wavelength measurements for
    taking high resolution spectra of bright stars suchas epsilon aurigae. These are
    projects for me next year.
    As well as improving on the above there are other projects which
    I have been thinking about:
    Project A
    - measure continua of stars using ultra-low resolution spectroscopy
    from say 2 to 50 Angstroms resolution. Key stellar paramaters can
    be calculated from these type of measurements - temperature, gravity
    and metallicity. Two significant surveys have been carried out which
    provide reference field stars and data for doing these surveys: the LICK survey
    and more recently the MILES survey.
    Under 1000 stars has been carried out in each of these
    surveys and it would be interesting to carry out a similar analysis on
    stars that were not covered and which are in the range of amateurs
    such as myself
    Project B
    - using data from above to practice classifying star spectral
    types would also be interesting and perhaps looking at rules for
    classifying stars where our cameras have highest quantum efficiency
    as currently most of the classification rules are at the blue end
    (3800-4500A)for historical reasons or up at infra-red or beyond.
    So plenty of projects to carry out then even at low resolution :0)
    Lets talk about the second objective - the theory behind how spectra
    are formed. Basically what I want to do is before I go out in the
    field to take detailed spectra of a star I want to know what I
    should expect to see and why so that if I do see something
    unexpected I can talk about it to other amateurs.
    Finding out what I should expect to see is easier than understanding
    why its that way so I will cover that first.
    There are several ways of doing this but I use VOSPEC which is
    a great tool and it can be found at
    http://www.sciops.esa.int/index.php?project=ESAVO&page=vospec
    along with some great flash videos showing you how to use the tool.
    Benefit of this tool is it accesses loads of archives for stars and
    you can also look at spectra of synthetic stars built out of stellar models as
    well.
    So thats great as long as the star is there but if you want to know
    things like what elements are for what lines. You can look at a
    library of elements such as you do in VSPEC but then you normally
    have options as several elements have absorption lines around a
    specific wavelength.
    So this is where stellar modelling comes in.
    There is available on the net free software which will allow you to produce your
    own stellar models :0)
    There are several available out there but I use Kurucz stellar
    modelling software called ATLAS which has been updated by Castelli to
    run on linux. This models the main sequence stars which are not too
    hot eg O stars or the ones that are too cool).
    Before you go - oh no not linux I need another PC and how much is
    this going to cost I will say that -to use Atlas you can run it on your own PC
    and all the software is free providing you have an internet connection and
    enough spare disk space on your PC)
    How you do it is you download Virtual PC (if you have good old XP like me or you
    can use the Windows 7 equivalent which is also free)and install it on your
    machine.
    You then install linux on your machine - I use fedora - you can
    download the software from their site.
    Next you install two compilers a free c one the GNU version and the intel
    fortran compiler details of how to do this are on the intel web site.
    Next you follow the instructions linked to the Atlas cookbook and you
    download the atlas code and data and compile it.
    Link to Atlas cookbook is here:
    http://wwwuser.oat.ts.astro.it/atmos/atlas_cookbook/Atlas_Cookbook.html
    There are two versions of Atlas - Atlas9 and Atlas12. Atlas9 is what
    I have used up to now and it appears to work - The main atlas program
    generates the stellar model then you have two other programs in the Atlas9 sweet
    called Width and Synthe which let you specify the resolution and then generate
    the spectra. The data file which has the spectra in it has the great advantage
    of telling you what elements are contributing to the spectra :0)
    Although as input into Atlas you need to have a starting model for
    your star - you can access grids at kurucz web site
    http://kurucz.harvard.edu/
    However to choose a grid you need to know the temperature, gravity
    and metallicity of the star you are studying. So how do you know that?
    Well you can get the data from simbad at
    http://simbad.u-strasbg.fr/simbad/
    But if you can't find the data online you'll have to work it out for
    yourself - that means probably doing Project A which I mentioned
    above.
    Thats enough on Atlas.
    If you want to look at hot stars and stars with accretion disks then
    you could try TLUSTY which can be found at http://nova.astro.umd.edu/
    although I have not tried to get that program working on linux yet.
    Ok so we are now in a position to produce synthetic spectra for a
    star and compare it against the spectra we see.
    Now for the tricky question why does the spectra appear in the way
    it does.
    I'll now list a set of books which will help you on your way to
    answering this question...and will comment on them
    Non-technical books:
    - Keith Robinsons books appear to be an excellent non-technical
    introduction to the theory behind spectroscopy and starlight. I have
    only flicked through them myself but only some vey basic school
    algebra is all you require - his books are called Starlight and
    Spectroscopy: Key to the stars and can be found on amazon.
    - Obviously there will be ken's book when it comes out but I believe
    it focusses on the practical aspects of spectroscopy and not the
    theory behind why stars have the spectra they have
    Technical books:
    To get the most out of the following books you will need to understand
    quite a bit of calculus - Schaums outline in calculus and advanced
    calculus should suffice for these.
    Here are two undergraduate books on astronomy:
    An Introduction to stellar Astrophysics - Francis Le blanc - this is
    an excellent introduction and the chapter on stellar atmospheres
    is excellent
    An introduction to Modern Astrophysics - Carroll/Ostle - also called
    the big orange book is excellent in its encyclopaedic coverage of
    astrophysics although in my opinion some bits could be more simply
    explained
    Now onto the postgraduate level stuff
    Stellar Photospheres - David Gray - I have only just bought this and
    in my opinion this is THE book on stellar spectroscopy if you want
    to see how the professionals do it - it covers a lot of ground and is
    quite readable and has plenty references out to the professional
    literature. I am ploughing through it and this book represents
    about how far my undertanding has got to in why spectra are the way
    they are.
    Now for the advanced stuff:
    Gray covers only slightly more than Local Thermodynamic Equilibrium models so
    where the non-local thermodynamic equilibrium models are described have to head
    to
    Rutten: Radiative Transfer in Stellar Atmospheres - this is freely
    available on the web and will be the next book I want to study after
    completing Gray
    And finally the book we have all been waiting for which is the new
    edition of Stellar Atmospheres which is written by Mihalas and Hubeny
    which is due out on amazon any month now....this one is not for the
    light hearted...
    I think thats enough for now any comments from people out there
    welcome I am interested to see who out there is trying to climb
    this slippery pole of finding out what the spectra of stars is all
    about
    John
×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue. By using this site, you agree to our Terms of Use.