1: \documentstyle[11pt,IAUS215,twoside,epsf]{article}
2:
3: \markboth{N. Langer et al.}{Binary evolution models with rotation}
4: \pagestyle{myheadings}
5: \setcounter{page}{1}
6:
7: \nofiles
8:
9: % Some definitions used in these instructions.
10:
11: \def\msun{\,\mathrm{M}_\odot}
12: \def\mso{\,\mathrm{M}_\odot}
13: \def\msoy{\, \mso~{\rm yr}^{-1}}
14: \def\simle{\mathrel{\hbox{\rlap{\hbox{\lower4pt\hbox{$\sim$}}}\hbox{$<$}}}}
15: \def\simgr{\mathrel{\hbox{\rlap{\hbox{\lower4pt\hbox{$\sim$}}}\hbox{$>$}}}}
16: \def\msoy{\, \mso~{\rm yr}^{-1}}
17:
18: \def\emphasize#1{{\sl#1\/}}
19: \def\arg#1{{\it#1\/}}
20: \let\prog=\arg
21:
22: \def\edcomment#1{\iffalse\marginpar{\raggedright\sl#1\/}\else\relax\fi}
23: \marginparwidth 1.25in
24: \marginparsep .125in
25: \marginparpush .25in
26: \reversemarginpar
27:
28: \begin{document}
29: \vspace*{1cm}
30: \title{Binary evolution models with rotation}
31: \author{N. Langer, S.-C. Yoon, J. Petrovic}
32: \affil{Astronomical Institute, P.O. Box 80000, NL-3508 TA Utrecht,
33: The Netherlands}
34: \author{A. Heger}
35: \affil{Department of Astronomy and Astrophysics, Enrico Fermi
36: Institute, The University of Chicago, 5640 S. Ellis Avenue, Chicago
37: IL 60637, USA}
38:
39: \begin{abstract}
40: We discuss the first available binary evolution models which include
41: up-to-date rotational physics for both components, as well as angular momentum
42: accretion and spin-orbit coupling. These models allow a self-consistent
43: computation of the mass transfer efficiency during Roche-lobe overflow
44: phases, and a determination of the transition from quasi-conservative to
45: non-conservative evolution. Applications to massive binary systems lead to
46: predictions for the spin rates of compact objects in binaries, and for
47: the occurrence of gamma-ray bursts from collapsars in binaries.
48: Rotational effects in
49: accreting white dwarfs are found to stabilise the shell burning and
50: decrease the carbon abundance in progenitor models for Chandrasekhar-mass
51: Type~Ia supernovae, and to potentially avoid a detonation of the white
52: dwarf within the sub-Chandrasekhar mass scenario.
53: \end{abstract}
54:
55: \section{Introduction}
56: The evolution of a single star can be strongly influenced by
57: its rotation (e.g., Heger \& Langer 2000; Meynet \& Maeder
58: 2000), and evolutionary models of rotating stars are
59: now available for many masses and metallicities.
60: While the treatment of the rotational processes in these models
61: is not yet in a final stage (e.g., magnetic dynamo processes are just
62: about to be included; Heger et al., this volume), they provide
63: first ideas of what rotation can really do to a star.
64:
65: Effects of rotation, as important they are in single stars,
66: can be much stronger in the components of close binary systems:
67: Estimates of the angular momentum gain of the accreting star
68: in mass transferring binaries show that critical rotation
69: may be reached quickly (Packet 1981; Langer et al. 2000).
70: Therefore, we need binary evolution models
71: which include a detailed treatment of rotation in the stellar interior,
72: as in recent single star models. However, in binaries, tidal
73: processes as well as angular momentum accretion need to be considered
74: at the same time. Some first such models are now available and
75: are discussed below.
76:
77: These models provide evidence for rotational processes in binaries
78: being essential for some of the most exciting cosmic
79: phenomena, which may occur exclusively in binaries: Type~Ia supernovae,
80: the main producers of iron and cosmic yardstick to measure the
81: accelerated expansion of the universe. Gamma-ray bursts from collapsars
82: --- which current stellar models with rotation
83: preclude to occur in single stars (Heger et al., this volume;
84: Woosley \& Heger, this volume)
85: --- may provide the most powerful explosions
86: in the universe and trace star formation to its edge. And black holes,
87: even though also formed by single stars, need a companion to become
88: ``visible''.
89:
90: \section{How much matter can stars accrete from a binary companion?}
91:
92: As mentioned above, non-magnetic accretion, i.e. accretion via a viscous disk or
93: via ballistic impact, transports angular momentum and can lead to a strong
94: spin-up of the mass gaining star. For disk accretion, it appears plausible that
95: the specific angular momentum of the accreted matter corresponds to
96: Kepler-rotation at the stellar equator; this leads to a spin-up of
97: the whole star to critical rotation when its initial mass is increased
98: by about 20\% (e.g., Packet 1981). Can accretion continue beyond this?
99:
100: Theoretically, this appears possible, as viscous processes may
101: transport angular momentum
102: outwards through the star, the boundary layer, and the accretion
103: disc (e.g., Paczynski 1991).
104: However, as the star may be strongly rotating, its wind mass
105: loss may be dramatically increased (Langer 1997, 1998), which may render
106: the mass transfer process inefficient.
107:
108:
109: Observations of massive post-mass transfer binary systems constrain this
110: effect.
111: Table~1 lists parameters of four different kinds of massive close
112: binary systems which give opposite answers. The two O~stars in the
113: Case~A binary (mass transfer starts while both stars undergo core
114: hydrogen burning)
115: V$\,$729~Cyg have a mass ratio of 3.5 but the same
116: spectral type and visual flux. Clearly, an initial mass ratio close
117: to~1 is required to get close to the observed current mass ratio.
118: However, as during Case~A the primary star (as we designate the
119: initially more massive star in a binary) loses just about half of
120: its mass, a mass ratio of at most~2 could be produced were the
121: secondary (the initially less massive star in a binary)
122: not allowed to accrete. Another system showing strong evidence
123: for accretion is the massive X-ray binary Wray~977; it would require
124: that stars of $\ga 40\mso$ form neutron stars to explain this system
125: without accretion (Wellstein \& Langer 1999).
126:
127: \begin{table}[t]
128: \caption{Clues on accretions efficiencies from observed binaries}
129: \begin{tabular}{c c c c c}
130: \tableline
131: object & sp. types & orb. period & masses/ratio & accretion? \\
132: \tableline
133: V729~Cyg$^1$ & O7+O7$^2$ & 6.6$\,$d & $q=3.5$ & YES \\
134: Wray 977$^3$ & BI+NS & 44$\,$d & $40\msun + 1.4\msun$ & YES \\
135: 3 systems$^4$& WNE+O & $\sim 8\,$d& $q\simeq 0.5$ & NO$^5$ \\
136: 4U~1700-37$^6$& O6I+NS/BH& 3.4$\,$d & $58\msun + 2.4\msun$ & NO$^5$ \\
137: \tableline
138: \tableline
139: \end{tabular}\\
140: $^1$ Bohannan \& Conti (1976) \\
141: $^2$ both components have the same visual magnitude \\
142: $^3$ Kaper et al. (1995), Wellstein \& Langer (1999) \\
143: $^4$ Petrovic \& Langer (2002) \\
144: $^5$ meaning: 10\% or less \\
145: $^6$ Clark et al. (2002) \\
146: \end{table}
147: \index{V729~Cyg}\index{Wray 977}\index{4U~1700-37}
148:
149:
150: Several Galactic short period WNE+O binaries, on the other hand, can
151: not be understood had the O~star accreted substantial amounts
152: from the WNE progenitor
153: (Petrovic \& Langer 2002). While those might have formed through
154: common envelope evolution --- for which little accretion is expected --- the key
155: X-ray binary 4U$\,$1700-37 has such a short period that a major
156: accretion phase can be excluded. However, as Case~C evolution
157: (mass transfer starts after core helium burning of the primary
158: star) would
159: lead to a compact object much more massive than 2.4$\mso$, Case~B
160: evolution
161: (mass transfer starts just after core hydrogen burning of the primary
162: star)
163: is most likely here (Clark et al. 2002).
164:
165: We conclude from these observed binary systems
166: that in some massive close (Case~A or~B)
167: binaries, the mass
168: transfer process is nearly conservative, while in others it is strongly
169: non-conservative. In the following, we show that including rotational
170: physics into binary evolution calculations allows to recover both features
171: within the same physical model.
172:
173: \section{Binary models with rotating components}
174:
175: We have constructed binary evolution models using the code of Wellstein
176: et al. (2001), but including the physics of rotation as in the single
177: star models of Heger et al. (2000) for both components.
178: In addition, spin-orbit coupling according to Zahn (1977) has been
179: added, and rotationally enhanced winds are implemented as in Langer
180: (1998). The specific angular momentum of the accreted matter is assumed
181: to be that of Kepler rotation at the stellar equator in the case of
182: disk accretion, and determined by integrating the equation of motion
183: of a test particle in the Roche potential in case the accretion stream
184: impacts directly on the secondary star (Wellstein 2001).
185:
186: \begin{figure}[th!]
187: %\plotfiddle{figure1.ps}{12 truecm}{0}{59}{59}{-180}{-80}
188: \centering
189: \epsfxsize=0.88\hsize
190: \centerline{\epsffile{figure1.ps}}
191: \caption{Equatorial rotation velocity for primary
192: (dashed line) and se\-condary (solid line) component
193: of a $16\mso +15\mso$ system with an initial orbital period of
194: 3~d as function of time, starting at an age of 7~Myr, i.e., well before
195: the onset of mass transfer, until the end of Case~ABB mass transfer,
196: which corresponds to the time of the supernova explosion of the primary.
197: The four different mass transfer phases which occur in this system
198: are indicated; except for
199: the slow Case~A mass transfer they occur on the thermal time scale
200: of the primary star (see also Fig.~2).
201: }
202: \end{figure}
203:
204: Fig.~1 shows the evolution of the equatorial rotation velocity in a
205: system starting out with a $16\mso$ and a $15\mso$ star in a 3~day
206: orbit. The initial rotational velocity of both stars is unimportant
207: since they evolve quickly into rotation which is synchronous with the orbital
208: revolution, due to spin-orbit coupling.
209: Each of the three thermal time scale mass transfer phases through
210: which this system evolves (rapid Case~A, Case~AB, and Case~ABB;
211: see Wellstein et al. 2001) leads to a strong spin-up of the secondary
212: star and an equally drastic spin-down of the primary (see Langer 1998,
213: for the purely mechanical spin-down effect).
214:
215: \begin{figure}[t]
216: %\plotfiddle{figure2.ps}{10 truecm}{0}{59}{51}{-180}{-70}
217: \centering
218: \epsfxsize=0.70\hsize
219: \centerline{\epsffile{figure2.ps}}
220: \caption{For the $16\mso +15\mso$ system with an initial orbital period of
221: 3~days: mass transfer rate ($-\dot M_{\rm 1}$; full drawn line)
222: and mass accretion
223: rate of the secondary ($\dot M_{\rm 2}$; dotted line) as function of the
224: total amount of mass which has already been transferred.
225: Four discrete mass transfer phases can be distinguished:
226: rapid Case~A ($-\dot M_{\rm 1}\ga 10^{-4}\msoy$,
227: $\Delta M_{\rm 1}\simeq 8.7\msun$),
228: slow Case~A ($-\dot M_{\rm 1}\ga 10^{-7}\msoy$,
229: $\Delta M_{\rm 1}\simeq 0.8\msun$),
230: Case~AB ($-\dot M_{\rm 1}\ga 10^{-5}\msoy$,
231: $\Delta M_{\rm 1}\simeq 3.8\msun$), and
232: Case~ABB ($-\dot M_{\rm 1}\ga 10^{-4}\msoy$,
233: $\Delta M_{\rm 1}\simeq 1.2\msun$).
234: }
235: \end{figure}
236:
237: The accretion induced spin-up can bring the secondary close to critical
238: rotation and thus strong rotationally enhanced mass loss sets in. The
239: difference between the mass overflow rate $-\dot M_{\rm 1}$ and the
240: wind mass loss rate of the secondary is the net accretion rate
241: $\dot M_{\rm 2}$ of the secondary star. Fig.~2 shows that during much
242: of the rapid Case~A mass transfer --- during which the accreting star
243: increases its mass from 15$\mso$ to $\sim 23\mso$ ---,
244: it is $-\dot M_{\rm 1} = \dot M_{\rm
245: 2}$. This is possible due to two factors: 1)~During direct impact
246: accretion, the specific angular momentum of the accreted material is
247: only a fraction of the respective Kepler angular momentum. 2)~As the
248: secondary star fills a significant fraction of its Roche volume, tidal
249: coupling removes part of the accreted angular momentum and feeds it
250: into the orbit.
251:
252: As the orbit widens significantly during the later
253: evolution, the specific angular momentum of the accreted matter
254: increases and tidal forces weaken. Thus, while the accretion efficiency
255: is close to~1 during Case~A, it is less than~0.1 later, resulting in a
256: time average over the whole evolution of~0.67. For systems which
257: start out with a wider orbit, we find a low accretion efficiency
258: throughout (Fig.~3). For $16\mso +15\mso$ systems, we find a critical
259: initial period of $\sim 8\,$d beyond which the secondary accretes only
260: little.
261:
262: \begin{figure}[t]
263: %\plotone{figure3.ps}
264: \centering
265: \epsfxsize=0.92\hsize
266: \centerline{\epsffile{figure3.ps}}
267: \caption{Time averaged mass accretion efficiency $\beta$ for various
268: binary evolution sequences (symbols) and different physical assumptions,
269: as function of the initial orbital period for $16\mso +15\mso$ systems.
270: Triangles mark systems which have been calculated using rotational
271: physics and the tidal model as proposed by Zahn (1977), which in systems
272: marked by squares has been assumed to be one order of magnitude more
273: efficient. Star symbols designate models by Wellstein et al. (2001)
274: which have been computed without rotational physics and assumed to
275: evolve conservatively unless a contact situation occurs. The dotted
276: vertical line separates Case~A systems (to the left) from Case~B
277: systems. }
278: \end{figure}
279:
280: These models imply that, despite the angular momentum problem
281: for the accretion star (Packet 1981), quasi-conservative evolution of
282: massive close binaries is possible. However, already in early Case~B
283: systems, the accretion efficiency may be strongly reduced compared to
284: binary models without rotation.
285:
286:
287: \section{Black hole formation in binaries}
288:
289: While single stars more massive than $\sim 25\mso$ are supposed to form
290: black holes (Fryer 2002), it is a challenge to binary
291: evolution modellers to find a way to make black hole binaries ---
292: binary systems consisting of a non-degenerate star and a black hole.
293: The reason is that the initially more massive star, which is supposed
294: to end up as black hole, undergoes tremendous mass loss (Fig.~2,
295: and see Wellstein \& Langer 1999).
296:
297: Brown, Lee \& Bethe (1999) suggested the Case~C channel to form
298: the Galactic low mass BH binaries --- a low mass star
299: in close orbit with a black hole of typically 6$\mso$ (Orosz
300: 2002), i.e., starting with a wide system consisting of a (say)
301: 25$\mso$ and a $5\mso$ star. This suggestion is strongly supported by
302: the massive post-Case~B X-ray binary 4U$\,$1700-37 (Table~1):
303: it shows that in Case~B
304: systems, even a 60$\mso$ primary forms a compact object of only
305: 2.4$\mso$ (Clark et al. 2002).
306:
307: The problem of how wide the Case~C
308: channel is for such massive stars is difficult. Current single star
309: models suggest no significant radius increase late during or after core
310: helium burning at solar metallicity. On the other hand,
311: Smartt's (2002) abundance analysis of Galactic supergiants
312: suggests that they do not undergo ``blue loop'' evolution --- which
313: contradicts the current models. The problem is less severe at low
314: metallicity, where many models predict a gradual radius increase during
315: core helium burning.
316:
317: It is even harder to produce Cyg~X-1 type BH binaries, with black hole
318: masses as high as 14$\mso$. Brown et al. (2001) argue that this
319: requires Case~C evolution and an upturn of the initial-final mass
320: relation for single stars above $\sim 60\mso$, as predicted by
321: Langer (1987).
322:
323: \section{Spins of compact objects and gamma-ray bursts}
324:
325: The inclusion of rotationally enhanced magnetic fields in the evolution
326: of massive single stars (Heger et al., this volume) has improved the
327: agreement between observed ($15 ... 150\,$ms) and predicted ($\simgr 5\,$ms)
328: spin periods of young neutron stars --- with the consequence that
329: obtaining gamma-ray bursts from black hole formation in single stars
330: (collapsars; Woosley \& Heger, this volume) seems difficult at
331: present.
332:
333: From Sections~3 and~4 above, we conclude that the initially more
334: massive stars in massive close binaries are even less likely to produce a
335: gamma-ray burst. First of all, they lose so much mass that even stars
336: with a very large initial mass may not even form a black hole but rather
337: a neutron star (see also Wellstein \& Langer 1999). And secondly, Figure~1
338: shows how drastic these stars are spun down as a consequence of their
339: heavy mass loss. The 16$\mso$ star in the computed binary system is
340: expected to produce a neutron star with an initial spin period of more
341: than one second!
342:
343: The only way to avoid both drawbacks is to employ Case~C evolution,
344: which leads to a core evolution as in single stars. However, even then
345: the CO-core of the star needs to be spun-up significantly to produce
346: a collapsar and a gamma-ray burst --- a possibility suggested by Brown et
347: al. (1999, 2000) in the context of common envelope evolution and
348: spiral-in. No detailed models for this scenario exist at present.
349:
350: The initially less massive star in a massive binary, on the other hand,
351: accretes large amounts of angular momentum and will thus acquire a larger
352: core spin than a corresponding single star (see Fig.~1).
353: It is thus conceivable that accretion stars are the progenitors of
354: asymmetric supernova explosions and rapidly spinning compact objects.
355: Those sufficiently massive to transform into a Wolf-Rayet star during core
356: helium burning, or those which lose their envelope in a reverse Case~C
357: mass transfer,
358: may possess all required ingredients to produce a
359: gamma-ray burst within the collapsar model.
360:
361: \section{Progenitors of Type~Ia supernovae}
362:
363: Supernovae of Type~Ia are supposed to occur in close binary systems where
364: the accreting component is a CO-white dwarf. The currently favoured
365: scenario assumes that a white dwarf of initially less than $1\mso$
366: grows in mass due to quiescent nuclear shell burning of hydrogen and/or
367: helium, which is accreted from a non-degenerate companion.
368: In order to explode, it must reach the Chandrasekhar-mass, i.e., it needs
369: to accrete at least $0.4\mso$. While single white dwarfs are slow
370: rotators (Kawaler, this volume), the idea that Type~Ia supernova
371: progenitors are significantly spun-up due to accretion appears to be
372: supported by recent measurements of the rotational velocity of the white
373: dwarf in cataclysmic variables (Starrfield, this volume).
374:
375: In order for the degenerate CO-core to grow in mass, accreting white
376: dwarfs need to possess a helium burning shell source. However, like in
377: AGB stars, which have a degenerate CO-core in their center, the helium
378: shell source in accreting white dwarfs tends to turn thermally unstable,
379: causing large oscillations in nuclear energy production, as well as in
380: the white dwarf radius (Fig.~4). It is unclear whether the binary system can
381: survive these oscillations intact.
382: %
383: \begin{figure}[t]
384: \epsfxsize=0.88\hsize
385: \centerline{\epsffile{figure4.ps}}
386: \caption{Stellar and nuclear luminosity as function of time for a
387: C/O-white dwarf model starting at $0.8\mso$, which accretes helium
388: at a constant rate of $5\times 10^{-7}\msoy$ (Scheithauer 2000). }
389: \end{figure}
390: %
391: \begin{figure}[t]
392: \epsfxsize=0.8\hsize
393: \centerline{\epsffile{figure5.ps}}
394: \caption{Helium and carbon mass fraction as
395: function of the mass coordinate in a CO white dwarf which starts
396: at $1\mso$ and accretes helium-rich matter at a rate of $10^{-6}\msoy$,
397: at a white dwarf mass of $1.0286\mso$, computed with and without
398: rotation (Yoon \& Langer 2002). Note that the carbon mass
399: fraction $X_{\rm C}$ behind the shell source ($M_{\rm r} < 1.025\mso$)
400: is lower in the rotating
401: model ($X_{\rm C}\simeq 0.3$) than in the non-rotating one
402: ($X_{\rm C}\simeq 0.8$), corresponding to an
403: oxygen mass fraction of 0.7 rather than 0.2.}
404: \end{figure}
405: %
406: However, Yoon et al. (2003) found that the consideration that angular
407: momentum is accreted along with the matter by the white dwarf results in
408: different shell source properties. The spin-up of the outer part of the
409: white dwarf leads to a lower effective gravity and to a more relaxed
410: helium shell source. Additionally, differential rotation leads to
411: shear mixing at the location of the shell source, increasing its
412: geometrical thickness and often turning it stable. This effect increases
413: the likelihood of a given system to reach the Chandrasekhar-mass at all.
414:
415: However, not only does the thickness of the helium shell source
416: increase, but the chemical profiles through it become shallower (Fig.~5).
417: As thus most of the helium burning occurs at a lower helium
418: concentration, oxygen is produced at a much higher abundance than in
419: corresponding non-rotating models, and carbon is produced
420: correspondingly less.
421: As the combustion of oxygen yields less energy than that of carbon,
422: rotational effects influence the brightness of Type~Ia
423: supernovae. As the rotation rate depends on the amount of accreted
424: material, and the average amount of accreted matter increases with the
425: initial metallicity of the binary system (Langer et al. 2000), a drop of
426: the brightness of Type~Ia supernovae with higher metallicity ---
427: or an increase with redshift --- might occur.
428:
429: Finally, rotation also affects the evolution
430: of white dwarfs in the so called sub-Chandrasekhar mass scenario, as
431: outlined by Yoon \& Langer (2003; also this volume).
432:
433: %\section{Conclusions}
434:
435: %\acknowledgments
436:
437: \begin{references}
438: \reference Bohannan~B., Conti~P.S., 1976, ApJ 204, 797
439: \reference Brown G.E., Lee C.-H., Bethe H.A., 1999, New Astron. 4, 313
440: \reference Brown G.E., Lee C.-H., Wijers R.A.M.J., Lee H.K.,
441: Israelian G., Bethe H.A., 2000, New Astron. 5, 191
442: \reference Brown G.E., Heger A., Langer N., Lee C.-H., Wellstein
443: S., Bethe H.A., 2001, New Astron. 6, 457
444: \reference Clark J.S., Goodwin S.P., Crowther P.A., Kaper L.,
445: Fairbairn M., Langer N., Brocksopp C., 2002, A\&A 392, 909
446: \reference Fryer C.L.,
447: in {\it A massive star odyssey: from main sequence to supernova},
448: proc. IAU-Symp. 212, (San Francisco: ASP),
449: K.A. van der Hucht, A. Herrero \& C. Esteban, eds., in press
450: \reference Heger A., Langer N., 2000, ApJ 544, 1016
451: \reference Heger A., Langer N., Woosley S.E., 2000, ApJ 528, 368
452: \reference Kaper L., Lamers H.J.G.L.M., Ruymaekers E., van den Heuvel
453: E.P.J., Zuidervijk E.J., 1995, A\&A 300, 446
454: %\reference MacFadyen A.I., Woosley S.E., 1999, ApJ 524, 262
455: \reference Langer N., 1987, A\&A 171, L1
456: \reference Langer N., 1997,
457: in {\it Luminous Blue Variables: Massive Stars in Transition},
458: A. Nota, H.J.G.L.M. Lamers, eds, ASP Conf. Ser., Vol.~120, p.~332
459: \reference Langer N., 1998, A\&A 329, 551
460: \reference Langer N., Deutschmann A., Wellstein S., H\"oflich
461: P., 2000, A\&A 362, 1046
462: \reference Meynet G., Maeder A., 2000, A\&A 361, 101
463: \reference Orosz J.A., 2002,
464: in {\it A massive star odyssey: from main sequence to supernova},
465: proc. IAU-Symp. 212, (San Francisco: ASP),
466: K.A. van der Hucht, A. Herrero \& C. Esteban, eds., in press
467: \reference Packet W., 1981, A\&A 102, 17
468: \reference Paczynski B., 1991, ApJ 370, 597
469: \reference Petrovic J., Langer N., 2002,
470: in {\it A massive star odyssey: from main sequence to supernova},
471: proc. IAU-Symp. 212, (San Francisco: ASP),
472: K.A. van der Hucht, A. Herrero \& C. Esteban, eds., in press
473: \reference Scheithauer S., 2000, Diploma thesis, Potsdam University
474: \reference Smartt S.J., 2002,
475: in {\it A massive star odyssey: from main sequence to supernova},
476: proc. IAU-Symp. 212, (San Francisco: ASP),
477: K.A. van der Hucht, A. Herrero \& C. Esteban, eds., in press
478: \reference Wellstein S., 2001, PhD thesis, Potsdam University
479: \reference Wellstein S., Langer N., 1999, A\&A 350, 148
480: \reference Wellstein S., Langer N., Braun H., 2001, A\&A 369, 939
481: \reference Yoon S.-C., Langer N., 2003, A\&A, in preparation
482: \reference Yoon S.-C., Langer N., 2002,
483: in: {\it The Physics of Cataclysmic Variables and Related Objects},
484: ASP Conf. Ser. Vol 261, 2002, B.T. Gaensicke et al., eds., p~79
485: \reference Yoon S.-C., Langer N., Scheithauer S., 2003, A\&A, in preparation
486: \reference Zahn J.-P., 1977, A\&A 57, 383
487: \end{references}
488:
489: \end{document}
490:
491: