1: %\documentstyle[apjpt4]{article}
2: %\documentstyle[aas2pp4]{article}
3:
4: \documentclass[12pt,preprint]{aastex}
5: %\documentstyle[aj_pt4]{article}
6:
7: \begin{document}
8:
9: \title{Evidence of a Massive Black Hole Companion in the Massive Eclipsing Binary V Puppis}
10:
11: \author{Qian S.-B.\altaffilmark{1,2,3}, Liao W.-P.\altaffilmark{1,2,3}, and
12: Fern\'{a}ndez Laj\'{u}s, E.\altaffilmark{4}}
13:
14: \altaffiltext{1}{National Astronomical Observatories/Yunnan
15: Observatory, Chinese Academy of Sciences (CAS), P.O. Box 110, 650011
16: Kunming, P.R. China (e-mail: qsb@netease.com)}
17:
18: \altaffiltext{2}{United Laboratory of Optical Astronomy, Chinese
19: Academy of Sciences (ULOAC), 100012 Beijing, P. R. China}
20:
21: \altaffiltext{3}{Graduate School of the CAS, Beijing, P.R. China}
22:
23: \altaffiltext{4}{Facultad de Ciencias Astron\'{o}micas y
24: Geof\'{i}sicas, Universidad Nacional de La Plata, Paseo del Bosque
25: s/n, 1900, La Plata, Pcia. Bs. As., Argentina}
26:
27: \begin{abstract}
28: Up to now, most stellar-mass black holes were discovered in X-ray
29: emitting binaries, in which the black holes are formed through a
30: common-envelope evolution. Here we give evidence for the presence of
31: a massive black hole candidate as a tertiary companion in the
32: massive eclipsing binary V Puppis. We found that the orbital period
33: of this short-period binary (P=1.45 days) shows a periodic variation
34: while it undergoes a long-term increase. The cyclic period
35: oscillation can be interpreted by the light-travel time effect via
36: the presence of a third body with a mass no less than 10.4 solar
37: mass. However, no spectral lines of the third body were discovered
38: indicating that it is a massive black hole candidate. The black hole
39: candidate may correspond to the weak X-ray source close to V Puppis
40: discovered by Uhuru, Copernicus, and ROSAT satellites produced by
41: accreting materials from the massive binary via stellar wind. The
42: circumstellar matter with many heavy elements around this binary may
43: be formed by the supernova explosion of the progenitor of the
44: massive black hole. All of the observations suggest that a massive
45: black hole is orbiting the massive close binary V Puppis with a
46: period of 5.47 years. Meanwhile, we found the central close binary
47: is undergoing slow mass transfer from the secondary to the primary
48: star on a nuclear time scale of the secondary component, revealing
49: that the system has passed through a rapid mass-transfer stage.
50: \end{abstract}
51:
52: \keywords{Stars: binaries : close --
53: Stars: binaries : eclipsing ---
54: Stars: individuals (V Puppis) ---
55: Stars: evolution ---
56: Stars: black hole candidate}
57:
58: \section{Introduction}
59:
60: With an orbital period of 1.4545 days and two B-type components
61: stars (spectral types B1 and B3), V Puppis (HR\,3129 and HD 65818)
62: is one of a few massive semi-detached eclipsing binaries where the
63: less massive components are filling their Roche lobes (e.g.,
64: Andersen et al. (1983); Bell et al. (1987a, b); Terrell et al.
65: (2005)). According to the theory of binary evolution (e.g., Sybesma
66: 1986), theses systems were formed through a case A mass-transfer
67: (mass transfer is occurring when hydrogen is burning in the core).
68: Since the discovery of V Puppis more than one century ago, it was
69: extensively investigated photometrically and spectroscopically (e.g.
70: Andersen et al. (1983) and references therein). A weak X-ray source
71: in the field of V Puppis (e.g., Giacconi et al. (1974); Groote et
72: al. 1978); Bahcall et al. (1975)) and a significant amount of
73: circumstellar matter around the binary (e.g., York et al. 1976 and
74: Koch et al. 1981) make it a very interesting binary to study.
75:
76: Because of its stable light curve and the deep and sharp eclipsing
77: minima (e.g., Andersen et al. (1983)), times of light minimum of V
78: Puppis can be determined precisely. Thus, the variations of the
79: orbital period can in principle be derived with high precision by
80: analyzing the observed-calculated (O-C) diagram. In the present
81: letter, orbital period changes of V Puppis are investigated. Then,
82: based on the period variations, the presence of a massive black hole
83: companion and the evolutionary state of the system are discussed.
84:
85: \section{Analysis of the orbital period changes}
86:
87: Epoches and orbital periods of V Puppis have been given by several
88: authors (e.g., Andersen et al. 1983). Kreiner \& Ziolkowski (1978)
89: did not detect any period change for the binary, while Andersen et
90: al. (1983) pointed out that the period of V Puppis is variable.
91: However, since the early eclipse times were mainly observed
92: photographically or visually, conclusions on the period change are
93: not reliable. For the present analysis, we only use the
94: photoelectric (Pe) and charge-coupled device (CCD) observations.
95: Most of the data were collected by Kreiner et al. (2001). Two new
96: CCD times of light minimum were obtained on November 2, 2007 and on
97: March 24, 2008 with the Virpi S. Niemela 0.8-m telescope at La Plata
98: Observatory (see Table 1). During the observation, the V filter was
99: used. The $(O-C)_1$ values of all available Pe and CCD times of
100: light minimum were computed with the linear ephemeris given by
101: Kreiner et al. (2001),
102: \begin{equation}
103: Min.I=JD(Hel.)2428648.2837+1^{d}.45448686\times{E},
104: \end{equation}
105: \noindent where $2428648.2837$ is the time of conjunction,
106: $1^{d}.45448686$ the constant ephemeris period $P_{e}$, and $E$ the
107: cycle. The $(O-C)_1$ values are listed in the sixth column of Table
108: 1 and are represented graphically against $E$ in the upper panel of
109: Figure 1.
110:
111:
112: If the orbital period of a binary varies linearly in time, the (O-C)
113: diagram should have a parabolic shape, as observed in other
114: eclipsing binaries (e.g., Qian \& Zhu 2006). As for V Puppis, a
115: simple second-order polynomial does not give a satisfactory fit (the
116: dashed line in the upper panel of Figure 1). Therefore, a sinusoidal
117: term was added to a quadratic ephemeris to give a good fit to the
118: $(O-C)_1$ curve. A least-squares solution leads to the following
119: ephemeris,
120: \begin{eqnarray}
121: Min.I&=& 2428648.2960(\pm0.0045)+
122: 1^{d}.45448114(\pm0.00000091)\times{E}\nonumber\\
123: & &+4.18(\pm0.44)\times{10^{-10}}\times{E^{2}}\nonumber\\
124: & &+0.0163(\pm0.0016)\sin[0.^{\circ}2621(\pm0.0011)\times{E}+44.^{\circ}3(\pm3.^{\circ}9)].
125: \end{eqnarray}
126: \noindent The sinusoidal term suggests an oscillation with a period
127: of about $T=5.47$\,year and an amplitude of about $A=0.^{d}0163$,
128: which suggests a cyclic period oscillation with a period of 5.47
129: years (the upper panel of Fig. 1). Though the $(O-C)_1$ data do not
130: cover a whole cycle, as shown in Figure 1, the minimum, the maximum,
131: and the period of the cyclic variation are all well constrained,
132: indicating that the period changes are reliable. The quadratic term
133: in Eq. (2) reveals a continuous period increase at a rate of
134: $dP/dt=+2.10\times{10^{-7}}$\,days/year corresponding to a period
135: increase of 1.8\,s per century.
136:
137: \section{Discussion of the orbital period variation}
138:
139: \subsection{The presence of a massive black hole companion}
140:
141: Both components of V Puppis are B-type stars containing a convective
142: core and a radiative envelope, which is different from the Sun.
143: Hence, the period oscillation can not be explained as solar-type
144: magnetic activity cycles (i.e. the Applegate mechanism) (Applegate
145: 1992). Therefore, the simplest explanation of the cyclic period
146: oscillation is a wobble of the binary's barycentre due to the
147: presence of a tertiary companion. The $(O-C)_2$ values calculated
148: from the quadratic ephemeris in Eq. (2) are shown in the seventh
149: column of Table 1. The corresponding $(O-C)_2$ diagram is plotted
150: against the epoch number E in the middle panel of Figure 1. By
151: considering a general case with an elliptical orbit, the following
152: equation was used to describe those $(O-C)_{2}$ residuals,
153: \begin{equation}
154: (O-C)_{2}=a_{0}
155: +\sum_{i=1}^{2}{[a_{i}\cos(i{\Omega}{E})+b_{i}\sin(i{\Omega}{E})]}.
156: \end{equation}
157: \noindent The results were obtained with the same value of $\Omega$
158: ($0.^{\circ}2621$) as that used in equation (2) (solid line in the
159: middle panel of Figure 1). The orbital parameters of the tertiary
160: companion were computed by using the formulae derived by Kopal
161: (1959),
162: \begin{equation}
163: a_{12}^{\prime}\sin{i}^{\prime}={\it c}\sqrt{a_{1}^{2}+b_{1}^{2}},
164: \end{equation}
165:
166: \begin{equation}
167: e^{\prime}=2\sqrt{\frac{a_{2}^{2}+b_{2}^{2}}{a_{1}^{2}+b_{1}^{2}}},
168: \end{equation}
169:
170: \begin{equation}
171: {\omega}^{\prime}=\arctan\frac{(b_{1}^{2}-a_{1}^{2})b_{2}+2a_{1}a_{2}b_{1}}
172: {(a_{1}^{2}-b_{1}^{2})a_{2}+2a_{1}b_{1}b_{2}},
173: \end{equation}
174:
175: \begin{equation}
176: {\tau}^{\prime}=t_{0}-\frac{T}{2\pi}
177: \arctan\frac{a_{1}b_{2}-b_{1}a_{2}}{a_{1}a_{2}+b_{1}b_{2}},
178: \end{equation}
179:
180: \noindent where {\it c} is the speed of light, $a_{12}^{\prime}$,
181: $i^{\prime}$, $e^{\prime}$, ${\omega}^{\prime}$, and
182: ${\tau}^{\prime}$ are the semi-major axis, the orbital inclination,
183: the eccentricity, the longitude of the periastron from the ascending
184: node, and the time of the periastron passage, respectively. The
185: solutions are displayed in Table 2. With the following well-known
186: equations,
187: \begin{equation}
188: f(m) =\frac{4\pi^{2}}{G{T}^{2}}\times(a_{12}\sin{i^{\prime}})^{3}
189: \end{equation}
190: and
191: \begin{equation}
192: f(m)=\frac{(M_{3}\sin{i^{\prime}})^{3}} {(M_{1}+M_{2}+M_{3})^{2}},
193: \end{equation}
194: \noindent the mass function, the masses and radii of the third body
195: were calculated using the absolute parameters determined by Andersen
196: et al. (1983). In the formula, $M_{1}$, $M_{2}$, and $M_{3}$ are the
197: masses of the eclipsing pair and the third body, respectively, G is
198: the gravitational constant, and T is the period of the (O-C)
199: oscillation. For different values of the orbital inclination, the
200: masses and orbital radii of the tertiary component are displayed in
201: Table 2. The corresponding relations between the mass and the radius
202: of the third body and the orbital inclination are shown in Figure 2.
203:
204: The lowest mass of the tertiary companion is $10.4$\,$M_{\odot}$,
205: i.e. larger than the mass of the secondary component. If the
206: tertiary component is a normal star, we would see its spectral lines
207: not changing with the orbital phase of the binary. However, this has
208: not been reported (e.g., York et al. 1976; Koch et al. 1981;
209: Andersen et al. 1983). Moreover, the third body is more massive than
210: the secondary component ($M_2=7.76$\,$M_{\odot}$) in the central
211: binary, it should be very luminous and should contribute a large
212: amount of third light to the total system unless it is not a normal
213: main-sequence star. Therefore, we suspect that the tertiary
214: component is a compact object and a probable black hole candidate.
215: This situation resembles that in the triple system HR 2876 where a
216: compact object (possibly a white dwarf star) with a mass excess of
217: 1.0\,$M_{\odot}$ was found to be orbiting a B-type close binary star
218: (B3.5V+B6V; P=15 days) (Burleigh \& Barstow 1998; Vennes 2000). The
219: main differences between the two systems are that the tertiary
220: companion in the V Puppis is much more massive and that the central
221: eclipsing binary in V Puppis system has a much more tight orbit and
222: shows strong interaction between both components.
223:
224: A weak X-ray source (3U 0750-49) with $9.4\pm2.3$ count $s^{-1}$ was
225: detected by the Uhuru satellite (Giacconi et al. 1974). The
226: counterpart of the X-ray source was a puzzling problem for
227: astronomers since the 70s of last century. V Puppis was initially
228: listed in the 3U catalogue as a possible candidate by Giacconi et
229: al. (1974). However, another bright star HD\,64740, close to the
230: source (within the 90 percent confidence error box) was also
231: considered as a possible candidate (Groote et al. 1978). By
232: analyzing the X-ray data of V Puppis and HD 64740 from the
233: Copernicus satellite, Bahcall et al. (1975) later concluded that the
234: weak X-ray source must almost certainly be closer to V Puppis than
235: to HD 64740, but most probably neither V Puppis nor HD 64740 is the
236: candidate. More recently, V Puppis was also discovered to be a X-ray
237: source by ROSAT satellite with $0.08$ count $s^{-1}$ (X-ray
238: 50\,{\AA}, 250\,eV) (e.g., Lampton et al. 1996). Our study indicates
239: that the tertiary companion may correspond to the weak X-ray source.
240: There are two mechanisms that can produce the X-ray emission in the
241: V Puppis system. If the massive tertiary companion is a normal star,
242: the X-ray emission could be caused by the colliding winds from the
243: eclipsing pair and the tertiary companion, resembling those observed
244: in other massive binary systems (e.g., Stevens 1992; Pittard \&
245: Stevens 2002). If the third body is a massive black hole, the X-ray
246: source is reproduced by accreting material from the massive binary
247: via stellar wind as observed in X-ray emitting binaries (e.g.,
248: Prestwich et al. 2007). However, the former mechanism can be ruled
249: out, because no spectral lines of the tertiary component were found.
250:
251: Another interesting feature of V Puppis is the existence of a
252: significant amount of circumstellar matter, which was deduced from
253: the optical spectra and ultraviolet observations obtained by the IUE
254: spacecraft and the Copernicus satellite (e.g., York et al. 1976;
255: Koch et al. 1981; Andersen et al. 1983). Numerous interstellar
256: spectral lines, especially those of many types of heavy elements
257: (e.g., Mg II, Si I, Si II, Si III, C I, O I, Fe II and Fe III) were
258: found in the H II region around V Puppis, in which strong lines of
259: Si III and Fe III are remarkable (e.g., York et al. 1976 and Koch et
260: al. 1981). The formation of the circumstellar matter around V Puppis
261: is an unsolved problem. Considering that V Puppis has undergone mass
262: exchange, one may think that the circumstellar matter results from
263: from the binary evolution. However, we may question why the other
264: systems (e.g. AI Cru) that resemble V Puppis do not have the same
265: significant amount of circumstellar matter (Bell et al. 1987). The
266: most probable reason for the formation of the circumstellar matter
267: around V Puppis is that it is formed by the supernova explosion of
268: the progenitor of the tertiary black hole companion (e.g., Woosley
269: \& Weaver (1995) and Zhang et al. (2007)). This may produce the
270: heavy elements in the circumstellar matter and results in the
271: formation of the H II region around the eclipsing binary.
272:
273: \subsection{The slow mass transfer of V Puppis}
274:
275: V Puppis is a semi-detached binary system with the secondary
276: component filling the critical Roche lobe (e.g., Andersen et al.
277: 1983). The long-term period increase can be explained by the mass
278: transfer from the secondary to the primary component. By using the
279: equation,
280: \begin{equation}
281: {\tau}_{N}=10^{10}M_2/L_2,
282: \end{equation}
283: \noindent where $M_2$ and $L_2$ are the masses and the luminosity of
284: the less massive component, the nuclear time-scale of the secondary
285: is calculated to be ${\tau}_{N}=7.1\times10^{6}$\,years. The value
286: is close to the timescale of the period change
287: ${\tau}_{P}=\frac{P}{\dot{P}}=6.9\times10^{6}$\,years, which reveals
288: that V Puppis is now undergoing a slow mass-transfer evolutionary
289: stage on the nuclear time-scale of the less massive component. This
290: suggests that V Puppis is formed via a case A evolution and has
291: passed through a rapid mass-transfer stage. The results are in
292: agreement with the prediction from the theory of massive binary
293: evolution (e.g., Sybesma 1986). By assuming a conservative mass
294: transfer, the well-known equation
295: \begin{equation}
296: \frac{\dot{P}}{P}=3\dot{M}(\frac{1}{M_{2}}-\frac{1}{M_1}),
297: \end{equation}
298: \noindent yields a rate of mass transfer
299: $dM/dt=7.82\times{10^{-7}}$\,$M_{\odot}/year$. However, since the
300: nuclear time-scale of the secondary is slightly longer than the time
301: scale of period change, the slow mass transfer is insufficient to
302: cause the observed period increase. This suggests that the stellar
303: wind from the massive component of the binary should contribute to
304: the period increase. The situation resembles that of AI Cru (Qian et
305: al. 2008).
306:
307: By considering conservative mass and angular momentum, the
308: calculation by Plavec (1968) suggests that, to produce a case A mass
309: transfer of massive binaries, the initial orbital period should less
310: than 1.8\,days. One may ask a question what is the reason that
311: causes the origin of the initially short-period detached system? It
312: is possible that the massive tertiary companion has played an
313: important role for the origin and evolution of inner binary by
314: removing angular momentum from the central system via Kozai
315: oscillation (Kozai 1962) or a combination of Kozai cycle and tidal
316: friction (e.g., Fabrycky \& Tremaine 2007). This makes the inner
317: eclipsing pair to have a lower angular momentum and a shorter
318: initial orbital period.
319:
320: \section{Conclusions}
321:
322: The cyclic change of the orbital period, the presence of a weak
323: X-ray source near V Puppis, and the significant amount of
324: circumstellar matter around the binary star all support the
325: conclusion that a massive black hole is orbiting the massive close
326: binary V Puppis with a period of 5.47 years. Several massive stellar
327: black hole candidates have been discovered recently (e.g., Greiner
328: et al. (2001) and Orosz (2007)), but all candidates are found to be
329: a component in a binary system. We conclude here that we have
330: provided observational evidence for the first massive stellar black
331: hole candidate as a tertiary companion of a massive close binary
332: star. Theoretical investigations have shown that the formation of
333: black hole binaries should experience a common-envelope phase with
334: the spiral-in of the companion in the envelope causing the ejection
335: of the envelope (e.g., Taam \& Sandquist (2000) and Podsiadlowski et
336: al. (2002)). The mean distance between the black hole candidate and
337: the eclipsing pair is about 5.5 astronomical units (AU). During the
338: supergiant phase, the radius of the precursor of the black hole
339: companion in V Puppis may reach to several AUs. The survival of V
340: Puppis in this evolutionary stage suggests that the tertiary black
341: hole companion must be formed through an evolutionary process that
342: is different from that of the black holes in X-ray binaries (e.g.,
343: Woosley \& Weaver (1995) and Zhang et al. (2007)) and it could give
344: some constraints to the evolution of single massive stars and the
345: formation of isolated black holes.
346:
347: From the present investigation, we deduce that the progenitor of V
348: Puppis is a triple system that is composed of a short-period
349: detached B-type eclipsing binary and a much more massive tertiary
350: companion. By drawing angular momentum from the inner binary through
351: Kozai oscillation (Kozai 1962) or a combination of Kozai cycle and
352: tidal friction (e.g., Fabrycky \& Tremaine 2007), the massive
353: tertiary companion may play an important role for the origin the
354: central system, which causes the inner eclipsing binary to have a
355: very short initial orbital period. In the triple system, the massive
356: tertiary companion evolves faster, and finally into a massive black
357: hole via a supernova explosion (e.g., Woosley \& Weaver (1995) and
358: Zhang et al. (2008)). At the same time, the original more massive
359: component of the eclipsing pair evolves to fill its critical Roche
360: lobe. Then, the binary was undergoing a rapid mass-transfer stage,
361: and after the mass ratio was revered, the central binary finally
362: reached the present semi-detached configuration with a slow
363: secondary-to-primary mass transfer on the nuclear time scale of the
364: present less massive star. Supergiant stars have recently been found
365: to be tertiary companions of OB-type close binaries in the two
366: triple systems HD 167971 and FR Sct (e.g., Leitherer et al. (1987);
367: Davidge \& Forbes (1988), and Pigulski \& Michalska (2007)). They
368: resemble the progenitor of V Puppis with the tertiary supergiant
369: companion evolving into a compact object. As for V Puppis, the
370: existence of the black hole in the outer orbit poses strong
371: constrain on the formation of the black hole, the mass loss in the
372: supernova that lead to the formation of the black hole, and the
373: magnitude of the kick received by the black hole, all make the
374: object a much more interesting system for future study.
375:
376: \acknowledgments{This work is partly supported by the Yunnan Natural
377: Science Foundation (No. 2005A0059M), Chinese Natural Science
378: Foundation (No.10573032 and No.10433030), and the National Key
379: Fundamental Research Project through grant 2007CB815406. New CCD
380: photometric observations of V Puppis were obtained with the Virpi S.
381: Niemela 0.8-m telescope at La Plata Observatory. We are grateful to
382: Prof. Kreiner for sending us the times of light minimum of the
383: binary and to Cecilia Fari\~{n}a and Carolina von Essen for their
384: helps during the observations and the data processing. The authors
385: thank Prof. J.-P. De Greve for improving the English writing of the
386: original manuscript and the referee for those useful comments and
387: suggestions that help to improve the paper.}
388:
389: \begin{references}
390: %
391: \reference{}Andersen, J., et al., 1983, A\&A 128, 17
392: %
393: \reference{}Applegate, J. H. A, 1992, ApJ 385, 621
394: %
395: \reference{}Bahcall, J. N., et al., 1975, MNRAS 171, 41
396: %
397: \reference{}Bell, S. A., Adamson, A. J., Hilditch, R. W., 1987a,
398: MNRAS 224, 649
399: %
400: \reference{}Bell, S. A., Kilkenny, D., Malcolm, G. J., 1987b, MNRAS
401: 226, 879
402: %
403: \reference{}Burleigh, M. R. \& Barstow, M. A., 1998, MNRAS 295,
404: L15
405: %
406: \reference{}Charles, P. A. \& Coe, M. J., 2006, In: Compact
407: stellar X-ray sources (Eds. by W. Lewin \& M. van der Klis), P.
408: 215 (Cambridge University Press, Cambridge)
409: %
410: \reference{}Davidge, T. J. \& Forbes, D., 1988, MNRAS 235, 797
411: %
412: \reference{}Fabrycky, D. \& Tremaine, S., 2007, ApJ 669, 1298
413: %
414: \reference{}Giacconi, R. et al., 1974, ApJS 27, 37
415: %
416: \reference{}Greiner, J., Cuby, J. G., McCaughrean, M. J., 2001,
417: Nature 414, 522
418: %
419: \reference{}Groote, D., Kaufmann, J. P., 1978, A\&A 63, L9
420: %
421: \reference{}Koch, R. H. et al., 1981, PASP 93, 621
422: %
423: \reference{}Kopal, Z., 1959, Close Binary systems. (London: Chapman
424: \& Hall). P.112
425: %
426: \reference{}Kozai, Y., 1962, AJ 67, 591
427: %
428: \reference{}Kreiner, M. J., 2006, private communication.
429: %
430: \reference{}Kreiner, M. J., Kim, C.-H., Nha, I.-S, 2001, An Atlas
431: of O-C Diagrams of Eclipsing Binaries, Krakov, P.1914-1915
432: %
433: \reference{}Kreiner, M. J. \& Ziolkowski, J., 1978, Acta Astron. 28,
434: 497
435: %
436: \reference{}Lampton, M., Lieu, R., Schmitt, J. H. M. M., Bowyer, S.,
437: Voges, W., Lewis, J., and Wu, X., 1997, ApJS 108, 545
438: %
439: \reference{}Leitherer, C. et al., 1987, A\&A 185, 121
440: %
441: \reference{}Orosz, J. A. A, 2007, Nature 449, 872
442: %
443: \reference{}Pigulski, A. \& Michalska, G., 2007, Inf. Bull. Var.
444: Stars No. 5757
445: %
446: \reference{}Pittard, J. M. \& Stevens, I. R., 2002, A\&A 388, L20
447: %
448: \reference{}Plavec, M., 1968, Adv. Astr. Astrophys. 6, 202
449: %
450: \reference{}Podsiadlowski, Ph., Rappaport, S., Han, Z., 2002,
451: MNRAS 341, 385
452: %
453: \reference{}Prestwich, A. H., et al., 2007, ApJ 669, L21
454: %
455: \reference{}Qian S.-B., Zhao, E.-G., Fern\'{a}ndez Laj\'{u}s, E.,
456: Zhu L.-Y., Liu L., Liao W.-P., Carolina von E., 2008, in preparation
457: %
458: \reference{}Qian, S.-B. \& Zhu, L.-Y., 2006, AJ 131, 1032
459: %
460: \reference{}Remillard, R. A. \& McClintock, J. E., 2006, Annu.
461: Rev. Astron. Astrophys. 44, 49
462: %
463: \reference{}Stevens, I. R., Blondin, J. M., Pollock, A. M. T.,
464: 1992, ApJ 386, 265
465: %
466: \reference{}Stickland, D. J., Lloyd, C., Pachoulakis, I., Koch, R.
467: H., 1998, Observatory 118, 356
468: %
469: \reference{}Sybesma, C. H. B., 1986, A\&A 159, 108
470: %
471: \reference{}Taam, R. E. \& Sandquist, E. L., 2000, Annu. Rev.
472: Astron. Astrophys. 38, 113
473: %
474: \reference{}Terrell, D., Munari, U., Zwitter, T., Wolf, G., 2005,
475: MNRAS 360, 583
476: %
477: \reference{}Vennes, S., 2000, A\&A 354, 995
478: %
479: \reference{}Woosley, S. E. \& Weaver, T. A., 1995, ApJS 101, 181
480: %
481: \reference{}York, D. G., Flannery, B., Bahcall, J., 1976, ApJ 210,
482: 143
483: %
484: \reference{}Zhang, W.-Q., Woosley, S. E., Heger, A., 2008, ApJ 679,
485: 639
486: %
487: \end{references}
488:
489: \clearpage
490:
491: \begin{table}
492: \caption{All available photoelectric and CCD times of light minimum
493: for V Puppis.}
494: \begin{tabular}{lllllllll}\tableline
495: JD.Hel. & Errors & Meth. & Min. & E & $(O-C)_1$ & $(O-C)_2$ &
496: Residuals & Ref. \\
497: 2400000+ & days & & & & days & days & days &\\\tableline
498: 31917.957 & $\pm0.0010$ & Pe & I & 2248 & -0.0132 & -0.0147 & -0.0004 & (1)\\
499: 31965.952 & $\pm0.0010$ & Pe & I & 2281 & -0.0162 & -0.0176 & -0.0046 & (1)\\
500: 39869.6264& $\pm0.0010$ & Pe & I & 7715 & -0.0234 & -0.0164 & -0.0011 & (1)\\
501: 41791.0113& $\pm0.0002$ & Pe & I & 9036 & -0.0157 & -0.0104 & +0.0065 & (2)\\
502: 42537.9043& $\pm0.0001$ & Pe & II&9549.5& -0.0017 & +0.0026 & -0.0042 & (2)\\
503: 42561.9044& $\pm0.0005$ & Pe & I &9566 & -0.0006 & +0.0036 & -0.0044 & (2)\\
504: 43124.8032& $\pm0.0002$ & Pe & I &9953 & +0.0118 & +0.0151 & -0.0001 & (2)\\
505: 43148.8024& $\pm0.0002$ & Pe & II&9969.5& +0.0119 & +0.0151 & +0.0014 & (2)\\
506: 43151.7111& $\pm0.0002$ & Pe & II&9971.5& +0.0117 & +0.0149 & +0.0014 & (2)\\
507: 45367.6063& $\pm0.0002$ & Pe & I &11495 & -0.0039 & -0.0056 & -0.0022 & (2)\\
508: 45383.6052& $\pm0.0001$ & Pe & I &11506 & -0.0043 & -0.0061 & -0.0014 & (2)\\
509: 48391.5025& $\pm0.0010$ & CCD& I &13574 & +0.0142 & +0.0026 & +0.0021 & (3)\\
510: 48425.6868& $\pm0.0007$ & CCD& II&13597.5&+0.0180 & +0.0063 & +0.0044 & (3)\\
511: 48500.5947& $\pm0.0050$ & Pe & I &13649 &+0.0198 & +0.0078 & +0.0025 & (4)\\
512: 49821.2500& $\pm0.0050$ & Pe & I &14557 &+0.0011 & -0.0164 & -0.0003 & (4)\\
513: 54456.7420& $\pm0.0010$ & CCD& I &17744 &+0.0435 & +0.0012 & -0.0024 & (5)\\
514: 54550.5667& $\pm0.0010$ & CCD& II&17808.5&+0.0538 & +0.0109 & +0.0028 & (5)\\
515: \tableline
516: \end{tabular}\\
517: {References in Table 1:\\
518: (1) Kreiner \& Ziolkowski (1978); (2) Andersen et al. (1983); (3)
519: Kreiner (2006); (4) Stickland (1998); (5)The present authors.}
520: \end{table}
521:
522: \begin{table}
523: \caption{Calculated orbital parameters of the tertiary companion in
524: V Puppis.}
525: \begin{tabular}{lll}\tableline\tableline
526: Parameters & Values & Errors \\\tableline
527: $a_{0}$ & -0.0005& $\pm0.0001$\\
528: $a_{1}$ & +0.0142& $\pm0.0012$\\
529: $b_{1}$ & +0.0113& $\pm0.0012$\\
530: $a_{2}$ & -0.0031& $\pm0.0019$\\
531: $b_{2}$ & +0.0026& $\pm0.0004$\\\tableline
532: $a_{12}^{\prime}\sin{i}^{\prime}$ (AU) & 3.14 & $\pm0.21$\\
533: $e^{\prime}$ & 0.45 & $\pm0.16$\\
534: ${\omega}^{\prime}$ & $-63.^{\circ}0$ & $\pm18.^{\circ}9$\\
535: ${\tau}^{\prime}$ (days) & 2428213 & $\pm101$\\
536: f(m) ($M_{\odot}$) & $1.03$ & $\pm0.21$\\
537: $M_{3}(i^{\prime}=90^{\circ})$ & $10.40$ & $\pm0.87$\\
538: $M_{3}(i^{\prime}=70^{\circ})$ & $11.25$ & $\pm0.96$\\
539: $M_{3}(i^{\prime}=50^{\circ})$ & $14.74$ & $\pm1.33$\\
540: $M_{3}(i^{\prime}=30^{\circ})$ & $27.44$ & $\pm2.87$\\
541: $d_{3}(i^{\prime}=90^{\circ})$ & $6.82$ & $\pm0.73$\\
542: $d_{3}(i^{\prime}=70^{\circ})$ & $6.80$ & $\pm0.73$\\
543: $d_{3}(i^{\prime}=50^{\circ})$ & $6.28$ & $\pm0.70$\\
544: $d_{3}(i^{\prime}=30^{\circ})$ & $5.56$ & $\pm0.66$\\
545: \tableline\tableline
546: \end{tabular}
547: \end{table}
548:
549: \clearpage
550:
551: \begin{figure}
552: \begin{center}
553: \includegraphics[angle=0,scale=1.2]{f1.eps}
554: \caption{A plot of the $(O-C)_1$ curve of V Puppis with respect to
555: the linear ephemeris given by Kreiner et al. (2001). The solid line
556: in the upper panel suggests a combination of a long-term period
557: increase and a cyclic change, while the dashed line refers to the
558: continuous increase of the orbital period. The middle panel shows
559: the plot of the $(O-C)_2$ residuals of the eclipse timings based on
560: the quadratic ephemeris in Eq. (1). The solid line in the middle
561: panel is the theoretical orbit of the massive black hole candidate
562: with an eccentricity of 0.45. Residuals after subtracting all
563: effects of period change are displayed in the lower panel, and no
564: variations can be traced there.}
565: \end{center}
566: \end{figure}
567:
568: \begin{figure}
569: \begin{center}
570: \includegraphics[angle=0,scale=1.2]{f2.eps}
571: \caption{The relations between the mass $M_3$\,($M_{\odot}$) and the
572: orbital radius $d_3$ (AU) of the tertiary component and its orbital
573: inclination $i^{\prime}$ in the V Puppis system. The tertiary
574: companion should be more massive than the secondary component and
575: thus may be a black hole candidate.}
576: \end{center}
577: \end{figure}
578:
579: \end{document}
580: