Forum der Vereinigung der Sternfreunde

An Be-Stern-Interessierte der FG!
Folgende Nachricht sendete ich an die Be-Stern-Mailingliste:
--------------------------------------------------------------------
An: <mailto:BE-STARS@JISCMAIL.AC.UK (BE-STARS@JISCMAIL.AC.UK)>
Betreff: He lines in delta Sco

The He emission lines 6678 in delta Sco exhibit considerable changes between
February and July this year (see attachment):

1. change of the V/R ratio from V<R to V>R
2. change of the radial velocity of both peaks

Question:
Is there someone who can confirm this ?
Is there someone who is able to explain that phenomenon ?
-----------------------------------------------

Hier statements von Philippe Stee bzw. Richard Townsend:
-------------------------------------------------------- --------------------------------------------

---Ursprüngliche Nachricht---
Von: "Philippe Stee"
Betreff: Re: He lines in delta Sco
Datum: 26. Jul 2008 22:17

Hi,

based on interferometric measurement from the GI2T, we estimate the
size of the
emitting region in He6678 to be 2.3 R* (<8.5 R* in Hbeta and 18 R* in
Halpha) for gamma
Cas see Stee et al. 1998, A&A, 332, 268.

Thus He 6678 must originate very close to the star's surface. I would
favor its formation
thanks to a recent outburst in an rotating ring not centred on the
central star...

Is the double pics separation of He 6678 close to its vsin i ?

You may also have a look to our recent paper on Achernar which try to
explain the relation
between the line intensity and the double peak separation see:
Kanaan et al. 2008, A&A, 486, 785

Cheers,

Phil

______________________________
Philippe STEE
Chercheur - CNRS - Directeur PNPS
UMR 6525 UNS-CNRS-OCA
Responsable Equipe ESOM
Observatoire de la Côte d'Azur
Site de Grasse-Roquevignon
Avenue Copernic
[url=callto:06130]06130[/url] GRASSE - FRANCE
Tel: (33) 4 93 40 53 52
Fax: (33) 4 93 40 53 33

______________________________


-------------------------------
weitergeleitet von E.P.

---Ursprüngliche Nachricht---
Von: "Thom Gandet"
Betreff: Re: He lines in delta Sco
Datum: 26. Jul 2008 17:49

Hello,

Yes! Halonen et al (2008)
(http://fr.arxiv.org/PS_cache/arxiv/pdf/ ... 0988v1.pdf) found
that the disk density between about 1-4R* had changed since Zorec et al
(2007, A&A, 470, 239)
determined the density variation. (This is based on H-alpha.) Would not
HeI follow the same
general density variations as H-alpha? Has the width between the peaks of
the HeI 6678 line
changed, Ernst?

A couple of remarks about the binary:

There is still controversy over the orbit of the binary. The Sixth
Catalogue of Orbits of Visual
Binary Stars gives i=48° and the magnitude difference between the components
as 2.2
(based on minimum mag of A). Halonen found i=20° to fit their observations
somewhat
better. Isn't the periastron distance is more like 1R*? It would be very
interesting to
know what the gravitational field strength is like near periastron, compared
to the Roche lobe
constraints. Perhaps also non-radial pulsations could, given surface
rotational velocities near
critical and a boost from the nrp, Del Sco form a disk during periastron
(comments, Rich?).
It would be very interesting to see some orbital dynamical models for this
highly eccentric binary.
An updated interferometric orbit would be very helpful, too, since the last
one was around 2000 (?).

Best regards,
Thom Gandet
Lizard Hollow Obserrvatory
Tucson, AZ USA
http://www.lhobs.org




----- Original Message -----
From: "Ernst Pollmann" <mailto:ernst-pollmann@T-ONLINE.DE (ernst-pollmann@T-ONLINE.DE)>
To: <mailto:BE-STARS@JISCMAIL.AC.UK (BE-STARS@JISCMAIL.AC.UK)>
Sent: Saturday, July 26, 2008 2:22 AM
Subject: Re: He lines in delta Sco


Dear colleagues!

First of all many thanks for all the comments !

Regarding the formation of a ring in this binary system, my friend and
colleague A. Miroshnichenko gave at the spectroscopy conference in
Heidelberg (in May this year) during his lecture the following assumption:

- the Halpha line width decreased (no contribution from rapidly rotating
part of
the disk);

- the line equivalent width (EW) decreased;

- the brightness decrease in 2004/5 has been due to a decreasing mass loss
from the primary

Possible explanations: the disk became a ring !

The binary is coming to its next periastron in February - May 2011.
Distance between the stars at periastron is d = 24 R1
Primary’s Roche lobe size ~ 0.6 d or ~ 15 R1

Consequences for Hálpha:
Some ring material may flow into the secondary’s Roche lobe
Ring may become denser and line emission will rise
Single- or triple-peak profiles may be observed

O.k., may be this will happen, we will see....

But we discuss about HeI6678 !

If in the current time situation, tidal effects because of the large
distance
between both components is hardly likely, and outbursts rather to be the
reason, what could be the explanation for the change of the V/R ratio ?

An elliptical rotating ring where the ring is not centred on the central
star ?

Ernst
-----------------------------------------------------

Dear colleagues!<?xml:namespace prefix = o ns = "urn:schemas-microsoft-comoffice" />
Comments of Thom:
Has the width between the peaks of the HeI 6678 line changed, Ernst?
...here the distances related to λ0 (6678.149) in km/s:
2008-02-02 > 230
2008-04-01 > 230
2008-07-23 > 190
What is the background of your question?
Comments of Philippe:
Thus He 6678 must originate very close to the star's surface. I would favor
its formation thanks to a recent outburst in an rotating ring not centred on the central star...
I agree, there are different other reasons too, which implicate the formation near the surface of the primary.
Comments of Philippe:
Is the double pics separation of He 6678 close to its vsin i ?

I´m sorry, but I don´t know yet the vsin i value of these lines.
Regarding elliptical orbit:
I can imagine an elliptical orbit which is responsible for the change of the V/R-ratio.
But this kind of orbit does not explain the monotonous shifting of the lines to higher wavelengths (or ?) !

Best wishes,
Ernst

---Ursprüngliche Nachricht---
Von: "Philippe Stee" <Philippe.Stee An: <BE-STARS Betreff: Re: He lines in delta Sco Datum: 27. Jul 2008 22:54
Hi Ernst,

the important point is not really the change of the radial velocity of both peaks but rather
the change in the peaks separation.


Event if it is more difficult to estimate the peaks separation for the 2008/07/23 spectrum
(since the profile appears not really double peaked but rather with 3 peaks) it seems that
the peaks separation taken 2008/07/23 is larger that those for 2008/04/01 and 2008/02/08
this is exactly what must be observed if the He6678 line is emitted in a rotation disk or ring
that is slowly expanding from the central star.


When it is ejected from the stellar surface its rotational velocity must be close to the stellar photosphere
rotation, i.e. close to the vsini of the star which is about 165-175 km/s (from Simbad) than if the disk or
ring is expanding and assuming a Keplerian rotation, its rotational velocity must decrease
(as a function of the radius ^(-0.5)) and thus the double peaks separation must decrease
as well as a function of time.


If you told me that for 2008/07/23 the double peaks separation is about 190 km/s, this is very
close to the vsini of the central star and this is another evidence that He6678 was just formed
and ejected from the stellar photosphere, maybe thanks to a recent "outburst".


The problem I see is that you also told me that for 2008/04/01 and 2008/02/08 the double
peaks separation is 230 km/s which is unlikely since it should be smaller that 190 km/s...


On the other hand a possible explanation is that we just see the disk/ring expansion as a function
of time and thus the double peaks separation is also decreasing as a function of time (from 230
km/s to 190 km/s) and should still decrease till it will be diluted and will be no more observable.
In the hypothesis, the vsini of the central star should be at least 230 km/s but it would not be the first
time that the true rotational velocity is underestimated (for instance the inclination angle i is
usually poorly estimated).


Cheers,


Philippe



______________________________
Philippe STEE
Chercheur - CNRS - Directeur PNPS
UMR 6525 UNS-CNRS-OCA
Responsable Equipe ESOM
Observatoire de la Côte d'Azur
Site de Grasse-Roquevignon
Avenue Copernic
06130 GRASSE - FRANCE
Tel: (33) 4 93 40 53 52
Fax: (33) 4 93 40 53 33

______________________________-------------------------------------
E.P

"Dietrich Baade" > schrieb: ...hallo Herr Baade,
vielen Dank für Ihre kritische Anmerkung, die ich gottlob wie folgt beantworten kann:
Seit ich mit neuem C14 in der STW zur Spaltspektroskopie Nov./2007 übergegangen bin, wird nach jeder Spektrenserie eine Neon-Kalibrationsaufnahme bei ebenfalls stets gleicher Belichtungszeit (60 sec) angefertigt.
Dies geschieht in immer gleichbleibender Position des Teleskops plus Spektrograph (im Cassegrainfokus) an Teilkreisen: im Meridian bei Deklination: -30° (delta Sco: ~ 23°)

So ist es mir heute möglich, die Kalibrationsgenauigkeit, hier (zunächst) ausgedrückt als Wiederholgenauigkeit der Dispersion (in Angstr./Pix) im Kalibrationsspektrum, für den fraglichen Zeitraum (02/2008-07/2008) zu dokumentieren:

0.245086 0.245303 0.245086 0.244000 0.244761 0.244491 0.244882 0.244285 0.244411 0.244594 0.244306 0.244110 0.244702 0.244922 0.244410 0.243275 0.246268
Die Standardabweichung vom Mittelwert (0.24469719) beträgt danach 0.00062861 (= 0.25689 %)

Mit dieser Dispersionsgenauigkeit wird ein jeweiliges delta Sco-Summenspektrum in Wellenlänge skaliert. Vergleiche der gemessenen Wellenlängen der Wasserdampflinien (hier verwende ich die Wellenlängen der UVES-Spektren von Herrn Hanuschik aus 2006) mit diesen UVES-Wellenlängen, führte bisher zu einer Standardabweichung zwischen 0.0039 - 0.015 Angstr.

Die Größenordnung dieser Standardabweichung bei der Bestimmung der Wellenlängen der HeI6678-Linien liegt somit deutlich unterhalb des gemessenen Verschiebungsbetrages (zwischen 2008/02/08 und 2008/07/23) von 2.2 Angstr.

Beste Grüße,
Ernst Pollmann

Hallo Roland!

Z.Zt. wird ein Beobachtungsergebnis von mir - Hel6678-Emissionen in delta Sco - in der Be-Star-Mailingliste diskutiert.
U.a. geht´s dabei um den Verschiebungsbetrag (Radialgeschwindigkeit) der Doppelpeakemission (siehe Anhang).

Herr Dietrich Baade (ESO) schrieb nun in diesem Zusammenhang:
man kann zur genauen Bestimmung von Verschiebungen als sehr bequemes Werkzeug eine Kreutzkorrelationsfunktion berechnen.

Da diesbezüglich meine mathematischen Kenntnisse wie Möglichkeiten begrenzt sind, würde ich dich fragen wollen, inwieweit du hier behilflich sein kannst.

Beste Grüße,
Ernst

Hallo Ernst,

das wäre kein Problem. Da ich davon ausgehe, dass die zeitlichen Änderungen von Interesse sind, könnte man als Korrelationsmaske eines der delta Sco - Spektren verwenden. Man bekäme dann die Verschiebungen in Bezug auf diesen Zeitpunkt.
Ich benötige dafür die Spektren in ganz normalem Textformat (Wellenlänge, Intensität).
Du kannst mir ja mal einige Spektren mailen. Wenn der Test positiv verläuft, ist es für mich keine große Arbeit, eine ganze Zeitserie auszuwerten.

Gruß
Roland

Weil ich weis, dass es unter den Forumslesern einige Teilnehmer gibt, die durchaus an Interesse daran haben, hinsichtlich des Themas im Betreff auf dem Laufenden zu bleiben, leite ich die jüngste Stellungnahme von Th. Gandet aus der Be-Star-Mailingliste hierhin weiter:
----------------------------------------------------------------------------

Mustn't any model take into account the multiple
components of the emission lines? Looking at the spectra, it seems a difficult probalem
to determine what is a separate "ring" and
what are V/R components of other "rings".
Measurements of the separation of the peaks
may provide a way of sorting that out.
If we are seeing rings, each will rotate at
a somewhat different velocity. If a disk
is involved, then the changes will tell if
the disk is expanding or collapsing.

There seems to be two physical effects
going on in Delta Sco, yes? One is the
ejection of material from the photosphere,
the other is the formation of "blobs" of gas
in the disk or ring(s) 9probably from viscosity
affects?). The blobs then rotate in a more or
less Keplerian fashion, eventually to fall back
closer to the star.

About the orbit: the orbit inclination is
not well determined - values between 20° and
48° are in the literature. The only directly
observed value of it, from interferometry,
depended primarily on incomplete coverage of the
orbit because of the high eccentricty and long
period.

There also remains a lot of uncertainty about the
orbital eccentricity, which will affect the
periastron distance. It would be good to have i
and e better constrained!

Would the periods and modes of nrps be affected
by the secondary at periastron passage? There
would have to be a change in gas denisty in the
photospher, is that correct? My question is
a separate one from whether the nrps contributed
to the disk formation.

Sorry for the long posting and probably dumb
questions!


There is a very interesting set of spectra
that shows the changes in various lines in
Delta Sco between 1995 and 2006. H-beta
and HeI 5876 are shown, and it's interesting
to compare them with Ernst's recent profiles.
The paper is by Galazutdinov and Krelowski,
in ApJ, 2006, 637, 342.


Cheers,
Thom Gandet


------------------------------------------------------------

Hallo Ernst,

hier die Ergebnisse meiner Auswertung im Anhang
als Excel-File. Ich habe sowohl die Halpha- als
auch die HaI6678-Emissionen mit SpecRaVE (also
Gaußfit) und Kreuzkorrelation (KK) ausgewertet.
Alle Ergebnisse sind von der Erdbewegung befreit,
also heliozentrisch. Da die KK nur
Geschwindigkeitsänderungen gegenüber einer
Referenz liefert, habe ich alle Werte einheitlich
auf dein erstes Spektrum JD=2454505.3 bezogen.
Beide Methoden reagieren natürlich auf etwas
unterschiedliche Weise auf Profiländerungen,
dennoch zeigen sie die gleichen Tendenzen. Mehr
würde ich erstmal nicht herauslesen wollen, da
man aufgrund der wenigen Messpunkte noch keine
Aussagen über die Streuung treffen kann.

Herzlicher Gruß
Roland

Am 13:01 29.07.2008 schrieben Sie: