INIS-mf—8631 FRANK VERBUNT ERRATA p.3O: 1.36: p(:)R25 should be p(:)R"2'5 p.55: add: Space Science Reviews, 1982, in press. p.57: 1.8: lOkeV should be 1 keV p.64: eq.(9) left hand side should be ph p.66: table 2: p should be in units g/cm3 p.68: 1.17: gas pressure is dominant at radii r > 1000 km, p. 73: the cooling formula of Bisnovatyi-Kogan and Blinnikov has a term VAT this should be AVT. p.76: 1.13: M should be M. p.95: unnumbered formula: 2(a-x ) should be 2(a-2x) Men moet gelijkelijk laohen en wetenschap beoefenen. Epikouros il MASS TRANSFER IN STELLAR X-RAY SOURCES PROEFSCHRIFT TER VERKRIJGING VAN DE GRAAD VAN DOCTOR IN DE WISKUNDE EN NATUURWETENSCHAPPEN AAN DE RIJKSUNIVERSITEIT TE UTRECHT, OP GEZAG VAN DE RECTOR MAGNIFICUS PROP.DR. M.A. BOUMAN, VOLGENS BESLUIT VAN HET COLLEGE VAN DECANEN IN HET OPENBAAR TE VERDEDIGEN OP WOENSDAG 7 JULI 1982 DES NAMIDDAGS TE 2.45 UUR DOOR FRANCISCUS WILHELMUS MARIA VERBUNT GEBOREN OP 14 MEI 1953 TE GOIRLE I PROMOTOREN : PROF. DR. H.G. VAN BUEREN PROF. DR. E.P.J. VAN DEN HEUVEL (UNIVERSITEIT VAN AMSTERDAM) Dankbetuiging Het Is een plezierige plicht al die personen te bedanken zonder wie dit proefschrift niet tot stand zou zijn gekomen. In de eerste plaats zijn dat uiteraard mijn beide promotores: Henk van Bueren en Ed van den Heuvel. Zowel binnen als buiten de sterrenkunde was het me een genoegen met hen te mogen omgaan. Ze lieten me veel vrijheid, maar waren steeds aanwezig als ik hun hulp nodig had: ik ben hen hier zeer erkentelijk voor. Roeif Takens wil ik bedanken voor de tijd en de moeite die hij heeft genomen om me op weg te helpen in het onderzoek naar groei8chijven. In het tweede deel van dit proefschrift heeft menige opmerking van hem een plaats gevonden. Op het moment dat het onderzoek naar groeischijven in een luwte raakte, bracht Jan van Paradijs uit de V.S. de vaart en de allure mee waar het eerste deel van dit proefschrift een gevolg van is. Mijn dank! Karin Kieboom wil ik bedanken voor het doorzettingsvermogen waarmee zij de literatuur over massaoverdracht in dubbelsterren met een lage massa heeft bestudeerd. Het derde deel van dit proefschrift profiteert daarvan. Naast deze mensen en de overige co-auteurs wil ik ook nog al diegenen bedanken met wie ik op zo'n plezierige manier over allerlei fysische problemen heb kunnen discussieren: met name noem ik Bram Achterberg, Aad van Ballegooijen, John Heise, en Leo van der Horn. Ed Faverey, Maria Hoesman, en Evert Landré dank ik voor de nauwgezetheid waarmee ze de figuren uit dit proefschrift hebben verzorgd. De Nederlandse Organisatie voor Zuiver Wetenschappelijk Onderzoek, Z.W.O., heeft gedurende de eerste drie jaar van dit onderzoek in mijn levensonderhoud voorzien, de Rijksuniversiteit van Utrecht in het laatste jaar. Dat ik me al die tijd redelijk wel ben blijven voelen dank ik niet in de laatste plaats aan de goede vriendschap met Frits Paerels en met Astrid de Leeuw. Summary ' In this thesis some aspects of binaries that emit X-rays, are discussed. The brightest X-ray sources in our galaxy as observed by satellites like UHURU all consist of a compact star - i.e. a star with a similar mass as the sun, but with a much smaller radius - and an accompanying ordinary star. The ordinary star loses matter that falls in the deep gravitational potential well of the compact star, and emits its released energy as X-rays. In chapters VII and VIII we investigate which mechanism is responsible for the mass transfer in systems where the mass-losing star is less massive than the sun. According to the theory of general relativity the binary loses energy in the form of gravitational radiation, which causes the stars to fall towards one another. According to the current theory, proposed by Paczynski in 1967, this effect sufficiently explains the mass transfer. In chapter VII it is shown that this is not the case, and that especially in the somewhat wider binaries much more matter is transferred than can be explained in terms of this theory, even when we take some hitherto neglected effects into account. In chapter VIII we describe another mechanism of mass transfer. It is well known that a star like the sun loses angular momentum with its stellar wind. When such a star loses angular momentum in a binary, this will lead to loss of angular momentum from the system, as rotation and revolution are tldally coupled in a narrow binary. By a rough estimate we show that it is possible to explain the observed mass-transfer rates with this mechanism. The transferred matter will often enter a gaseous disk around the compact star, and spiral inwards slowly through this disk. In chapter V we investigate the conditions for the formation of such a disk, and write down the equations governing its structure. A classification scheme enables us to discuss the different models that have been proposed» It is concluded that different models lead to very similar results for those regions of the disk where gas pressure is more important than radiative pressure, and that these results agree fairly well with observations. No consistent model has been constructed as yet for the region where radiative pressure is dominant, and that occurs only around non-magnetic neutron stars or black holes. Theoretically one predicts that the optical light emitted by a disk around a neutron star is mainly caused by X-ray photons from the immediate surroundings of the neutron star that hit the outer disk surface, are absorbed, theraalised, and re-eaitted in the optical and ultraviolet region* of the spectrum* Thia expectation it verified by comparison with the collected observational data of low-mass X-ray binaries. In chapter II-IV optical observations on two sources are analysed. 2A0311-227 consists of a strongly magnetised white dwarf and a low-mass companion. Matter is falling from the companion towards the surface of the white dwarf along the magnetic field lines. The lightcurves are very intricate, and difficult to explain in a detailed model.' But this much can be concluded that the striking change of the Ha-lines from emission into absorption is caused by the appearance of a bright red source of light behind the gae that emits the emission lines. In chapter IV an MgH and Mgb triplet depression are identified in the spectrum of the transient X-ray source Cen X-4. This depression must be explained by the presence of a K-star, and the existence of a companion to the neutron star is therefore proved observationally. Apart from this we find that the emission lines are still rather strong, although the X-ray flux has decreased with a factor 101*. A possible explanation for this is given as heating of the gas by thermal radiation from the hot neutron star. In the introduction I try to improve the legibility of the articles for a more general public than the specialists for who the articles were written by giving some information on the general background of this research. Also I use the opportunity to discuss the recent X-ray and infrared observations on 2AO311-227. CONTENTS Dankbetuiging 5 Summary 6 I Introduction 9 II Simultaneous Optical Photometric and Spectroscopie Observations of 2A0311-227 23 (co-authora: E.P.J. van den Heuvel, Th.J. van der Linden, J. Brand, F. van Leeuwen, and J. van Paradijs) III Simultaneous Spectroscopie and Photometric Observations of 2A0311-227 33 (co-authors: as chapter II) IV Spectroscopie Observations of the Optical Counterpart of Centaurus X-4 45 (co-authors: J. van Paradijs, Th.J. van der Linden, H. Pedersen, and W. Warnsteker) V Accretion Disks in Stellar X-ray Sources 55 VI Optical Properties of Low-mass X-ray Binaries 85 (co-author: J. van Paradijs) VII Some Aspects of Low-mass Close Binary Models for Bright Galactic Bulge X-ray Sources and 4U1626-67 91 (co-author: Karin H. Kieboom) VIII Magnetic Braking in X-ray Binaries 103 (co-author: C. Zwaan) Samenvatting 110 Levensloop 112 INTRODUCTION I 1. Historical overview and general background. When the launch of the UHURÜ satellite, December 12 1970, opened the era of systematic investigation of the X-ray sky, rocket experiments had already revealed the existence of a surprisingly large number of strong celestial X-ray sources. From the concentration of the sources towards the galactic plane it was clear that a substantial fraction of the objects in the second UHURU catalogue of X-ray sources (1) is situated in our own galaxy. A model for the X-ray sources in our own galaxy had already been proposed by Shklovsky (2), following the identification of one of them with a nova-like optical star (3). In this model a neutron star captures matter lost by an accompanying main-sequence star of low (<1 Me) mass. As the captured matter falls into the deep gravitational potential well of the neutron star, an enormous amount of energy is liberated and emitted as X-rays (4). Several serious objections to this model were raised almost immediately. The supernova explosion in which the neutron star is formed will tear a low-ma88 companion star to pieces (5). And even granted that the low-mass star survives the explosion, it would still be thrown out of the neighbourhood of the neutron star, as the explosion dissolves the binary (6). It was therefore a great surprise when the analysis of one year of data from the pulsating source Cen X-3 (i.e.