VV Cep - certain open questions for discussion

VV Cep 2017-2019 Campaign
Moderator for this forum: Ernst Pollmann
Post Reply
Thilo Bauer
Posts: 37
Joined: Sun May 31, 2015 4:45 pm
Contact:

VV Cep - certain open questions for discussion

Post by Thilo Bauer »

Dear all,

At the OHP I was able to get some nice spectra of VV Cep. Having an image at hand, probably helps to follow the discussion about spectral profiles of VV Cep, beta Lyr (shelyak), etc. So, I created a simple mount of certain raw stellar spectra, that I’ve taken at the OHP with my Vixen 8“ Cassegrain, Alpy 600 and modified Canon EOS 60D DSLR with Astronomik UV/IR L-Filter. Although, wavelength range is uncalibrated, Balmer lines of alpha Lyr will yield good watermarks for typical wavelengths of spectral properties.

This is an invite to discuss (and correct me, if I'm wrong, of course)!

From comparison of these stars, some of them Be stars, like Sheliak, VV Cep yields very confusing aspects from having a first look onto stellar spectrum obtained:
  • Prominent emission lines
  • No He emission
  • Very low UV excess at Balmer lines below H-gamma (434 nm)
This is just what was my first impression: Hohle et al. (2010) reported spectral type M2epIa and 64 solar masses for the prominent red supergiant and B8Ve/B9 and 35 solar masses for the hot companion. Although, late spectral type Be might be thought and indicated by emission lines, it seems more likely, that the companion is not a Be star. Also note the missing He emission line compared to beta Lyr, P Cyg and CH Cyg, which also has an M type primary.
See: Hohle, Neuhäuser, Schulz, 2010: Astron. Nachr. / AN 331, No. 4, 349 – 360 (2010)

Any additions or comments on this remark?

From other papers, it is believed, the companion should have spectral type of a late B or A type. Taking into account given mass ratio of the dominant supergiant (64 solar masses) and the hot companion (35 solar masses), the hot companion of the VV Cep binary system also doesn't really seem to be a "light-weight" main sequence star.

I found a nice animation, how stars like the two components of VV Cep may evolve over time:
http://spiff.rit.edu/classes/phys230/le ... vol_hr.swf

I tried the animation to compare two stars with 30 and 70 solar masses. Luminosity and color of both stars will yield different results. The heavy 70 solar mass star will yield a huge diameter and red color, while the 35 solar mass companion should end up with a hot surface and lower overall brightness compared to the more heavy super giant. This could be an indication, that the companion will not directly be visible in the composite stellar spectrum of the binary system. Indeed, there is a publication, where authors tried to evaluate the two composite spectra. The authors reported, that there was no success to find a match for the companion (N. Ginestet and J. M. Carquillat, 2002. The Astrophysical Journal Supplement Series, Volume 143, Issue 2, pp. 513-537).

Simulated luminosities from the animation expressed in solar luminosities L_sun yield about 1.4 Mio L_sun for the red supergiant and 0.33 L_sun for the hot companion at similar time scales. However, this may indicate, brightnesses aren't as different, as one might expect. Thus, in a later state, like that we observe now for the M-type supergiant, the difference in brightness of both binary components should be about 1.5 magnitude difference, which is not really a big difference of apparent magnitude in the night sky.

From my spectrum, however, it is very impressive, how faint the blue and UV part of the spectrum (below 434 nm) really is. In this wavelength range one should expect the maximum of light of the hot companion. Please consider, the image intensities are already enhanced by amplifying low image intensities using a square-root like intensity correction. Thus, taking into account a difference in luminosity (apparent magnitude) the companion should show up a little brighter, like CH Cyg. Of course, this is not absolutely correct, when talking about "luminosity" or "bolometric magnitude" vs. absolute (or apparent) magnitudes of the two binary components of VV Cep. But, taking into account, that a hot star shows a more blue compressed spectrum, the blue part from the hot companion of type B should show up a little brighter compared to the broad M spectrum. To evaluate this, look how different spectra of an A type spectrum of alpha Lyr look compared to VV Cep or CH Cyg. Nevertheless, even after amplification of the spectral intensities the blue part of the spectrum of VV Cep remains "dark". Please correct me, if I'm wrong here.

Questions:

What might be reasons, why the two spectral components could not be obtained from matching synthetic stellar spectra?

Is there a professional here in the group, able to provide more precise estimates to the luminosities or absolute magnitudes of the two binary components?

What might be the reason not to find more obvious hints from a binary companion, whose magnitude difference shouldn't be as low, as seen from my spectrum?

What do you think?

Cheers,

Thilo

Image
Attachments
ComparisonOfSpectra.jpg
Ernst Pollmann
Posts: 461
Joined: Mon Sep 26, 2011 7:16 pm

Re: VV Cep - certain open questions for discussion

Post by Ernst Pollmann »

Hi Thilo,
this is an interesting and fundamental consideration.
To get an overview of the differences in brightness of both components, I used my own (Wuppertal-seminar) tool (see attachments), in order to "play" with the radii, masses, and temperature references we found in the literature (see my distributed pdf-paper, written by Ph. Bennet, J. Hopkins and me).

For the primary I used (the following averages):
- 1500 solar radius
- 19 solar masses
- 2500 K

for the companion (the following averages):
- 17 solar radius
- 16 solar masses
- 15000 K

The use of these average parameters did lead to the following luminosities:

Primary = 80000 solar luminosity
Secondary = 13000 solar luminosity

By using of delta mag = -2.5 * lg (80000/13000), the brightness difference between the both stars is 1.97 mag. This "small" difference contradicts the difficulties in identification the spectrum of the companion.
On the other hand, Hack et al. (Astron. Astrophys. Suppl. Ser. 95, 1992) has been able to analyse spectra differences such as: Sp(B+M)-Sp(M) at different phases of the orbit. His identification in that way of the Balmer sequence, FeII and HeI, leads to the assumption, that the companion should be a "normal" (?) B star. But the usage of the stellar parameter above leads to the assumption, that it's more a giant than a main sequence star.

Ernst
Attachments
M-star.PNG
M-star.PNG (178.25 KiB) Viewed 3847 times
companion.PNG
companion.PNG (174.28 KiB) Viewed 3847 times
Last edited by Ernst Pollmann on Tue Sep 01, 2015 6:58 am, edited 1 time in total.
Thilo Bauer
Posts: 37
Joined: Sun May 31, 2015 4:45 pm
Contact:

Re: VV Cep - certain open questions for discussion

Post by Thilo Bauer »

Hi Ernst,

I'm always impressed, how different the numbers taken from simulations of star evolution will appear. However, this ratio of the luminosities is within certain limit of what I also expected to find from independent computations. I roughly estimated 1.5 and 2 magnitudes difference in brightness, when asking my stomach. So this is more or less confirmed by both of our quick approaches based on computations.

The error of both stellar masses evaluated from the papers seems quite large: 64 solar masses vs. 19 solar masses. I'm wondering if the primary star may have such a large diameter of 1500 solar radii taken into account only 19 solar masses for the red (super?) giant. But this is more an intuitive argument against the numbers found. Anyway, this is just a fraction of solar masses published by Hohle et al. Most of the work to achieve data from the secondary component has been done in the UV below 200 nm.

Seems we don't know much about the system and even the basic properties of both stars of the binary system are not well understood.

What could be done regarding observations in the visual?

Thilo
Post Reply