Akkretionsscheibenmodelle mit äußerer Einstrahlung

DSpace Repositorium (Manakin basiert)


Dateien:

Zitierfähiger Link (URI): http://nbn-resolving.de/urn:nbn:de:bsz:21-opus-13411
http://hdl.handle.net/10900/48632
Dokumentart: Abschlussarbeit (Master)
Erscheinungsdatum: 2004
Sprache: Deutsch
Fakultät: 7 Mathematisch-Naturwissenschaftliche Fakultät
Fachbereich: Sonstige - Mathematik und Physik
DDC-Klassifikation: 520 - Astronomie, Kartographie
Schlagworte: Akkretionsscheibe , NLTE , Spektrum , Astrophysik
Freie Schlagwörter: Röntgendoppelsternsystem , ultra-kompakt , 4U1626-67 , 4U0614+091 ,
accretions disc , irradiated , synthetic spectrum , LMXB , ultra compact
Lizenz: http://tobias-lib.uni-tuebingen.de/doku/lic_ubt-nopod.php?la=de http://tobias-lib.uni-tuebingen.de/doku/lic_ubt-nopod.php?la=en
Zur Langanzeige

Inhaltszusammenfassung:

Vor einigen Jahren wurde die Existenz einer anderen Art von Doppelstern- systemen voraus gesagt, in denen die Scheibe H und He frei sein sollte. Inzwischen hat man sechs dieser Systeme gefunden, deren Scheibe haupt- sächlich aus Metallen wie O, Ne und Mg (Juett, Psaltis & Chakrabarty 2001; Schulz et al. 2001) besteht, was darauf schließen läßt, daß der Donorstern der Kern eines C/O Weißen Zwerges ist. Aufgrund der Tatsache, daß die Komponenten des Systems kompakte Objekte sind, was dazu führt, daß das ganze System äußerst kompakt ist, bezeichnet man solche Systeme als ultra-kompakte Doppelsternsysteme. Der prominenteste dieser ultra-kompakten Doppelsternsysteme ist der LMXB (low mass X-ray binary) 4U 1626-67. Das System besteht aus einem 7.66 s Röntgenpulsar mit einem Weißen Zwerg von ca. 0.02 Sonnenmassen (Chakrabarty 1998) als Begleiter. Der Neutronenstern besitzt ein starkes Magnetfeld, welches von Orlandini et al. (1998) anhand von Zyklotronresonanzlinien zu 3E+12 G bestimmt wurde. Das System hat eine kurze Bahnperiode von ca. 42 min (Middleditch et al. 1981), ein starkes Indiz, daß der Begleiter extrem wasserstoffarm sein muss (Paczynski & Sienkiewicz 1981). Chakrabarty (1998) ermittelte eine Röntgenleuchtkraft zw. 7E+36 und 5E+37 erg/s, womit man eine Akkretionsrate von ca. 2E-10 Sonnemassen/Jahr erhält. Das Doppel- sternsystem hat bei einem Bahnabstand von 300 0000 km (Schulz et al. 2001 und eine Akkretionsscheibe,die vom Gezeitenradius bei 200 000 km bis hinein zum Korotationsradius bei 6 500 km reicht, wo die Scheibe durch das Magnetfeld des Neutronensterns zerstört wird. In solch kompakten Doppelsternsystemen kann man die äußere Bestrahlung der Akkretionsscheibe durch die Röntgenstrahlung, welche durch die Akkretion auf die Polkappen des Neutronensterns entsteht, nicht mehr vernachlässigen. Weite Teile des Scheibenspektrums sind durch Effekte, welche die Bestrahlung verursacht, dominiert. Als Beispiel kann man hier Emissionslinien von hochionisierten Elementen, wie OV oder CIV nennen. Eines der größten numerischen Probleme, welches bei der Berechnung von bestrahlten Akkretionsscheibenmodellen auftritt, ist die ungewöhnliche Temperaturschichtung und Struktur der Scheibe, welche starke Probleme mit den Besetzungszahlen der atomaren Energieniveaus in unterschiedlichen Tiefen mit sich bringt. Durch die Implementierung eines so genannten tiefenabhängigen Modellatoms sollen diese Probleme in meiner Diplomarbeit gelöst werden. In der Dissertation von Nagel (2003) wurde das Programm-Paket AcDc entwickelt, mit dem man die Vertikalstruktur und das synthetische Spektrum einer Akkretionsscheibe berechnen kann. Dazu wird eine radiale Scheiben- struktur entsprechend dem alpha-Scheiben-Ansatz von Shakura & Sunyaev (1973) angenommen und die Scheibe anschlieflend in eine Zahl konzentrischer Ringe zelegt. Für diese werden dann, unter der Annahme von plan-paralleler Geometrie, die Gleichungen des hydrostatischen- und des radiativen- Gleichgewichts zusammen mit den non-LTE Ratengleichungen und dem Strahlungstransport gelöst. Dabei wird die beschleunigte Lambda Iteration (ALI) (Werner & Dreizler 1999) verwendet. Die kinematische Viskosität wird durch die Reynoldszahl parametris ert, wobei für den vertikalen Verlauf die Beschreibung von Hubeny & Hubeny (1998) verwendet wird. Während der Diplomarbeit wurde ein tiefenabhängiges Modellatom (DEPTOM) entwickelt und als Programm-Modul in den AcDc-Code implementiert. DEPTOM vergleicht die Besetzungszahlen der einzelnen Ionisationsstufen während der ALI-Iteration mit einem unteren Limit. Liegt der Wert einer Besetzungszahl mehrmals unter diesem von außen vorgegebenen Limit, so wird die Ionisationsstufe im Ratengleichungssystem ausgeschaltet. Dies kann nicht durch auf Null setzen der entsprechenden Gleichung erfolgen. Es ist notwendig die Ionisationsstufe komplett aus der Ratenmatrix zu entfernen, da sich diese sonst nicht mehr invertieren läßt. Das Entfernen eines Teils der Ratenmatrix ist aber mit erheblichem Aufwand verbunden, da sicher gestellt werden muss, dafl die Zuordnung von Übergangen und Atomniveaus im weiteren Iterationsprozess stimmt. DEPTOM ist in der Lage, Ionisationsstufen, die numerische Probleme erzeugen, zu erkennen und aus der Berechnung zu nehmen. Dadurch ist AcDc stabiler und etwas leistungsfähiger geworden. Es ist nun erstmals möglich ein Modell einer bestrahlten Akkretionsscheibe, die von 6 500 km bis 100 000 km reicht, mit ein und demselben Modellatom zu rechnen. Dies entspricht einem Temperaturbereich von 14 000K bis 500 000 K. Die Struktur und Temperaturschichtung der Scheibe hat nun ein glatten Verlauf und zeigt keine Schwankungen mehr, die auf numerische Instabilitäten hindeuten.

Abstract:

Most of the known accretion discs in binary systems are composed of hydrogen rich material. Only a very small number of systems is known, in which a helium rich accretion disc is found. Today eleven of this cataclysmic variables, named after the prototype AMCVn, are known. A few years ago, a new type of binary system was predicted, in which the disc is supposed to be free of hydrogen and helium. Nowadays six systems are known, with an accretion disc principally composed of C, O, Ne and Mg (Juett, Psaltis & Chakrabarty 2001; Schulz et al. 2001). This fact leads to the assumption, that the donor star is a white dwarf with a C/O core. The over abundance of Ne, found in these systems, is due to diffusion processes in the history of the white dwarf. To determine the abundances of the chemical species, using quantitative spectral analysis is very useful to learn more about the evolution of binary systems and the constitution of white dwarfs. The most prominent representative of these ultra compact binary systems is the LMXB (low mass X-ray binary) 4U 1626-67. The system is consisting of a 7.66 s X-ray pulsar as primary component and a 0.02 solar mass white dwarf as secondary (Chakrabarty 1998). The fairly high magnetic field strength of the neutron star (3E+12 G) was measured by Orlandini et al. (1998) using cyclotron resonance lines. The rather short binary period of about 42 min suggests that the companion must be very hydrogen poor (Paczynski & Sienkiewicz 1981). Chakrabarty (1998) calculated a X-ray luminosity between 7E+36 and 5E+37 erg/s. Therewith one gets an accretion rate of about 2E-10 solar masses per year. Having a binary separation of only 300 0000 km (Schulz et al. 2001), the system has a quite huge accretion disc. The disc is extended from the corotation radius at 6 500 km, where the strong magnetic field of the X-ray pulsar destroys the disc, to the tidal radius at 200 000 km. In such compact binary systems one can not neglect the external irradiation of the accretion disc by the neutron star. The external irradiation dominates the disc pectrum over a wide range. One may find emission lines of highly ionised elements, like OV or CIV. One of the biggest numerical problems, occurring during the calculation of an irradiated accretion disc model, is the non-common temperature stratification and structure. This causes severe problems with the occupation numbers of some atomic levels. These problems should be solved by implementing the depth dependent model atom, during my thesis work. During his PhD thesis, Nagel (2003) has developed the numeric code package AcDc. Therewith one can calculate the vertical structure and the emergent synthetic spectrum of an accretion disc. By assuming a radial disc structure corresponding to the a-disc approach of Shakura & Sunyaev (1973), the disc is separated in a set of concentric disc rings. Afterwards, assuming plane-parallel geometry, the equation of hydrostatic and radiative equilibrium are solved consistently together with non-LTE rate equations and the radiation transfer equation. Therefore the ALI method (accelerated lambda iteration, Werner & Dreizler (1999)) is used. The kinematic viscosity is para eterised by the Reynolds number. The vertical run of the viscosity is taken from Hubeny & Hubeny (1998). During the work presented here the depth dependent model atom (DEPTOM) was developed and implemented in ACDC as a code module. DEPTOM is comparing the occupation numbers of every ionisation stage, during the ALI iteration, to a lower limit. Ionisation stages which are several times below the limit, are deactivated. This could not be done by setting the corresponding rate equation to zero, because then the rate matrix could not be inverted any more. The associated parts of the rate matrix system have to be cut out, which is difficult. One has to assure that the corelation between the atomic levels and the transitions is still true. DEPTOM identifies the trouble making ionisation stages and deactivates them. Thus the calculation is numerical more stable and ACDC is more powerful. Now, it is possible to calculate a disc model from 6 500 km to 100 000 km, using the same model atom. Thereby a temperature range from 14 000K up to 500 000 K is covered by the disc model. The structure and the temperature stratification of the disc model shows no longer features, which are connected to numerical instabilities.

Das Dokument erscheint in: