Highlight
Free Access
Issue
A&A
Volume 587, March 2016
Article Number A61
Number of page(s) 20
Section Catalogs and data
DOI https://doi.org/10.1051/0004-6361/201527130
Published online 17 February 2016

© ESO, 2016

1. Introduction

X-ray binaries (XRBs) are systems formed by either a neutron star (NS) or a black hole (BH), which is accreting mass from a companion donor star. Their detection began thanks to the development of space-based instrumentation in the 1960s, with the number of detections rising substantially since the implementation of all-sky monitors on board X-ray satellites, e.g. Ginga (1987–1991), RXTE (19962012) and, more recently, Swift (2004) and MAXI (2009). They are broadly divided into high-mass X-ray binaries (HMXBs) and low-mass X-ray binaries (LMXBs) according to the mass of the donor star. In the former, the early spectral type (O–B) massive star (10 M) mainly transfers material to the compact object through strong stellar winds. In LMXBs, on the other hand, the K–M spectral type star (M2 ≤ 1M) fills the Roche lobe and transfers mass by Roche lobe overflow through the inner Lagrangian point (Charles & Coe 2006). In this latter case, the material forms an accretion disc around the compact object whereby material is accreted. A few XRBs are also found with intermediate mass companions of spectral types in the range B–F. It has been proposed that these so-called intermediate-mass XRBs (IMXBs) may be the progenitors of some LMXBs through an episode of enhanced mass-transfer rate (Podsiadlowski et al. 2002).

The mass transfer rate () largely determines the observational properties and gives rise to a subclassification within the LMXB class. Persistent sources are those with high accretion rates ( ~ 10-8M yr-1; Tanaka & Shibazaki 1996) and X-ray luminosities close to the Eddington limit. This high luminosity ensures that the outer parts of the accretion disc dominate the optical spectrum owing to reprocessing, and effectively hide the companion star. By contrast, transient sources are systems with low accretion rates ( ≤ 10-9M yr-1; Tanaka & Shibazaki 1996) which exhibit long quiescent states and sporadic outburst episodes that are produced by thermal-viscous instabilities in the accretion disc (see Frank et al. 2002 for further explanation). During these outbursts, the brightness of the system rises to luminosities similar to those found in the persistent sources. After the outburst, transients decay back to quiescent states where they remain for most of their lifetimes. Typical recurrence times between outbursts span from years to centuries, depending on (Ritter & King 2002; McClintock & Remillard 2006).

Observations have revealed that ~25% of the transient LMXBs contain bursting NS (King et al. 1996a) while the rest (~75%) display X-ray spectral and/or timing properties that are characteristic of accreting black holes (hereafter we will refer to transient black hole systems as black hole transients or BHTs). In this paper we focus on the properties of Galactic BHTs since they represent the vast majority of the population of BHs. However, there are a few persistent, non-active, and extragalactic XRBs that harbour or may contain BHs.

Regarding the extragalactic population of BHs, dynamical evidence has been presented for LMC X-1 (a 11 ± 1M BH with an O7III companion; Orosz et al. 2009) LMC X-3 (a 7.0 ± 0.6M BH with a B3–5V star; Orosz et al. 2014 and Val-Baker et al. 2007) and M33 X-7 (a 16 ± 1M BH with a O7–8III star; Orosz et al. 2007) the first eclipsing stellar-mass BH ever detected.

On the other hand, indirect evidence for the presence of BHs in the HMXBs NGC 300 X-1 (12−24M BH with a Wolf-Rayet star, Crowther et al. 2010) and IC 10 X-1 (a 23−34M BH with a Wolf-Rayet star, Silverman & Filippenko 2008) has been postulated. Owing to the unique challenges of observing the wind-dominated Wolf-Rayet companions, the masses of the compact objects in these systems are rather uncertain and, indeed, the presence of a neutron star cannot be ruled out (Binder et al. 2015; Laycock et al. 2015). Finally, we should also mention ultraluminous X-ray sources (ULXs), systems with X-ray luminosities that are greater than the Eddington limit for a 10M BH. The source of these luminosities is still uncertain, and it has been proposed that they may be produced by intermediate mass BHs (~103M, see e.g. Miller et al. 2003) or stellar-mass BHs (e.g., Poutanen et al. 2007; Kawashima et al. 2012). More recently, for one case, it has been found that the compact object is a NS (M82 X-2; Bachetti et al. 2014) which confounds our understanding of ULXs even further.

Hereafter, we will only focus on the Galactic population of BHs. We have performed a thorough search of all X-ray systems published in literature since 1962, when the first extrasolar X-ray source was detected, (the NS system Sco X-1, Giacconi et al. 1962) up until 2015. In Sect. 2 we motivate the generation of the catalogue with a historical view of the sample of BHTs and present the catalogue itself. In Sect. 3 we study the population of BHTs, where we analyse the vertical distribution of BHTs and constrain the expected number of systems in the Milky Way. In Sect. 4 we focus on the population of dynamically confirmed BHs, presenting their distributions of periods, magnitudes, and masses.

thumbnail Fig. 1

Cumulative histogram of discovered (red) and dynamically confirmed (blue) BHTs as a function of time. Here, we also count Swift J1357.20933 as a dynamical BH. The lifetimes of the main X-ray satellites with all-sky monitor capabilities are shown with black lines.

Open with DEXTER

2. BlackCAT: the catalogue of Galactic BHs

Since 1966 up to 2015, 59 BHTs have been discovered and are represented in Fig. 1 as a cumulative histogram of red bars, the slope of which represents a detection rate of ~1.2 targets per year. However, only 17 of these BHTs have been dynamically confirmed to harbour accreting BHs (i.e. mass function 3M, see, e.g., Casares & Jonker 2014) and are represented by the blue bars in Fig. 1. Here, we should also add Swift J1357.20933, where the dynamical confirmation is indirect because it is not based on the detection of the secondary star. However, there is robust evidence that it contains a BH (see Corral-Santana et al. 2013 and Mata Sánchez et al. 2015 for more details).

Therefore, the 17+1 dynamical BHs represent ~ 30% of the total number of BHTs discovered so far. This low fraction is due to most BHTs becoming too faint in quiescence for radial velocity studies using current instrumentation because of their intrinsically faint companions, high extinction, large distance, or a combination of the above. This problem could be alleviated with the exquisite sensitivity of future facilities like the European Extremely Large Telescope (E-ELT), especially with better IR instrumentation. The rest of the population of BHTs are named BH candidates because they share similar X-ray characteristics in outburst to the confirmed ones, but they lack a final dynamical confirmation (see, e.g. McClintock et al. 2006 and Belloni et al. 2011 for reviews on the X-ray observational properties of the BH candidates). Indeed, NS and BH sometimes display qualitatively similar phenomenology in outburst (see, e.g. Muñoz-Darias et al. 2014) and we cannot ignore that some of the objects included in this catalogue might harbour NS.

BlackCAT is a complete catalogue containing the astrometric, photometric (near-infrared -NIR-/optical magnitudes in outburst and quiescence), number of outbursts, the peak X-ray flux, distance, finding charts, and dynamical parameters of all the BHTs discovered so far1. In this paper we present the most relevant properties in the tables, as explained below.

We divide the catalogue into three main types of Galactic BHs: transients, persistent, and non-active, depending on their X-ray activity.

Persistent

Cyg X-1 (a 15 ± 1M BH with a O9.7 Iab donor star, Orosz et al. 2011a) is the only confirmed Galactic BH in a persistent X-ray binary. On the other hand, 4U 1957+11 (Wijnands et al. 2002; Nowak et al. 2008, 2012; Hakala et al. 2014) and 1E 1740.7-2942 (Sunyaev et al. 1991) are persistent BH candidates, but they have not been dynamically confirmed. GRS 1758-258 (Mandrou 1990) is a quasi-persistent microquasar with a large extinction Av ~ 8.4 (Mereghetti et al. 1997) that is located near the Galactic centre region. Finally, SS 443 (Stephenson & Sanduleak 1977) is a non-transient source with a supercritical accretion regime onto a relativistic star (see Fabrika 2004 for a detailed review on the system). It is very likely a BH candidate with indirect arguments that support a compact object of 10−20M with a high inclination (Eikenberry et al. 2001). However, despite it being intensively studied for almost 30 years, the nature of this system is still uncertain and it could be a Galactic ULX.

Non-active BHs

MWC 656 has recently been proposed as the first BH HMXB with a Be-type companion star (3.8−6.9M BH with a B1.5-2 III star, Casares et al. 2014). This is based on radial velocity curves of gas encircling the companion star and the spectroscopic mass of the Be star. This system has not shown any type of outbursting activity, so we label it as a non-active BH. Three other black hole candidates have been discovered in globular clusters: the flat-spectrum radio sources M22-VLA1 and M22-VLA2 (Strader et al. 2012) and M62-VLA1 (Chomiuk et al. 2013).

Transients

The remaining systems are transients, although some of them could be considered semi-persistents (e.g. GRS 1915+105 or 4U 1630-472) because of their long stays in the outburst state. Below, we detail the content of the tables presented in this paper. The systems included are either dynamically confirmed BHs or have shown spectra and/or timing features that are typically found in BHs (Belloni et al. 2011). We also note the existence of Cyg X-3 (Giacconi et al. 1967), a transient source showing strong radio outbursts. However, the nature of the primary, which accompanies the Wolf-Rayet donor, is unclear and it could be either an NS or a BH. Thus, we have excluded Cyg X-3 from our list of BHTs.

Table A.1 presents the basic astrometric properties of the BHTs sorted chronologically by year of detection in X-rays. The dynamically confirmed BHs are highlighted in grey. The column distribution is:

  • (1)

    Year of discovery of the BHT;

  • (2)

    Name of the system and optical counterpart when known;

  • (35)

    Right ascension (RA) and declination (DEC) coordinates in equinox J2000. The accuracy in the astrometry and the source of the coordinates are also shown;

  • (67)

    Galactic longitude () and latitude (b) in degrees;

  • (89)

    Estimated distance (d) and height above the Galactic plane (z) in kpc;

  • (10)

    Number of outbursts detected after discovery in X-rays;

  • (11)

    References for the detection, best coordinates, and distance determinations.

Table A.2 shows the main properties, both in outburst and in quiescence, for all the BHTs presented in Table A.1. We also list the measured or predicted orbital period. Again, the dynamically confirmed BHs are highlighted in grey. The column distribution is:
  • (1)

    ID number used for cross-reference with the web version ofthis catalogue;

  • (2)

    Name of the system and optical counterpart when known;

  • (3)

    Peak X-ray flux in erg s-1 cm2, standardized to the 210 keV band. To do so, we begin with the X-ray flux that has been published in the literature (or in archive). We assume a power-law spectrum with a photon index Γ = 2 (Belloni et al. 2011) and the total neutral Galactic Hydrogen column density (NH), published by Kalberla et al. (2005). If there is a measured NH published in literature derived from direct X-ray spectral analysis, we use this instead of the radio-derived interstellar one;

  • (45)

    Optical or IR magnitude in the peak of the outburst and quiescence, respectively, in the AB system. To document the original observed band, we provide the name of the band in its original system;

  • (6)

    Optical Galactic extinction [E(BV)] reported in the literature. If unknown, we list the total Galactic line-of-sight absorption given by the Schlafly & Finkbeiner (2011) dust maps;

  • (7)

    Reported or estimated orbital period of the binary in hours;

  • (8)

    References for all the parameters above.

Table A.3 lists the optical/NIR photometry of the dynamical BHTs in quiescence:
  • (1)

    Preferred name of the dynamically confirmed BHT;

  • (29)

    Quiescent magnitudes of the dynamically confirmed BHTs. All the magnitudes were transformed to the AB system using the transformation coefficients taken from Table 2 in Frei & Gunn (1994) and Eqs. (5) in Blanton et al. (2005). To document the original observed band, we provide the name in its original system;

  • (10)

    References of the magnitudes.

Table A.4 provides the dynamical parameters of the BHTs:
  • (1)

    Preferred name of the dynamically confirmed BHT;

  • (2)

    Spectral type of the companion star;

  • (3)

    The orbital period of the binary in hours;

  • (4)

    The radial velocity of the companion star (K2) in  km s-1;

  • (5)

    The mass function of the BH f(M1) in  M;

  • (6)

    The mass of the BH (M1) in  M. If there is an uncertain value, we prefer to show a range of masses;

  • (7)

    The binary mass ratio q = M2/M1;

  • (8)

    The inclination of the system (i) in degrees;

  • (9)

    The rotational broadening (vrotsini) in  km s-1;

  • (10)

    References of all the parameters.

At the bottom of each table, we present a detailed list with the particularities marked in Table A.1 and Table A.2.

This census is the most updated available and, compared to previous catalogues (e.g. Gottwald et al. 1991; Chen et al. 1997; Remillard & McClintock 2006; Ritter & Kolb 2003; Liu et al. 2006, 2007), represents a substantial improvement in both statistics and observational parameters. While it only focuses on the population of BHs, it provides a thorough and more complete coverage of optical and NIR data and dynamical parameters2.

3. Analysis of the spatial distribution of BHTs

Table A.1 allows us to perform a statistical study of the distribution of BHTs in the Galaxy. In Fig. 2, we plot the 35 objects with estimated distances as viewed from the pole of the Milky Way. Filled orange circles represent the dynamically confirmed BHs while yellow stars mark the BH candidates. About 50% of the confirmed BHs are located within 4.5 kpc of the Sun. This is a clear indication that interstellar extinction is a severe limitation to dynamical mass determinations. Moreover, from Fig. 2 it seems that almost all BHs lie within a spiral arm. Thus, for BHTs with uncertain distances, likely values could be estimated or constrained using the distance to the spiral arm.

Figure 3 shows the position of all the BHTs in Galactic coordinates overlaid onto a projected image of the Milky Way, as well as histograms of their distribution in Galactic longitude and latitude. There is a clear concentration of objects towards the direction of the bulge (340°<< 20° and | b | < 10°, Dwek et al. 1995) and disc. Only two objects are located at high Galactic latitudes (XTE J1118+480 at b = + 62° and Swift J1357.20933 at b = + 50°), while three BHTs lie between 150°<< 210°. Moreover, 31 out of the 59 BHTs have no reported quiescent optical counterparts. In addition, 18 of them were detected in outburst, but they are either too faint in quiescence or the crowding of the field prevented the detection of a quiescent counterpart. On the other hand, 28 radio counterparts have been found (mostly during outburst). In Col. 5 of Table A.1, we indicate the range of the source of the coordinates together with the accuracy.

Stellar evolution predicts a population of 108−109 BHs in our Galaxy (van den Heuvel 2001; Remillard & McClintock 2006). However, the total number of Galactic BHTs is very uncertain and depends on several considerations. Thus, van den Heuvel (1992, 2001) estimated a population between of 500 and 1000 BHTs, based on only three to four BHTs being known at the time. Current estimates predict 103−104 Galactic BHTs (Romani 1998; Kiel & Hurley 2006; Yungelson et al. 2006) based on models, although these numbers are probably an underestimate because of the existence of systems with extremely long outburst duty cycles (Ritter & King 2002) or very faint peak X-ray luminosities (King & Wijnands 2006).

thumbnail Fig. 2

Galactic distribution of 35 BHTs with distance estimates (Table A.1), as seen from the Galactic pole. Dynamically confirmed black holes are marked in orange circles, whereas BH candidates are indicated by yellow stars. Distance ranges are represented with bars while the systems with lower limits are indicated by arrows that follow the same colour index. Some systems are overlapped with other. Each faint white ring in the background image represents an additional ~1.5 kpc from the Sun. Objects seem to concentrate along the direction towards the Galactic centre. The dynamical BHT in the upper right of the figure is BW Cir. Background image credit: NASA/JPL-Caltech/R. Hurt (SSC/Caltech).

Open with DEXTER

thumbnail Fig. 3

Top: the distribution of dynamically confirmed BHs (circles) and BH candidates (stars) in the Galactic plane, using the Hammer projection. (Background image credit: Mellinger 2009). Some of the symbols are overlapped by others, especially in the Galactic centre region. Bottom: histogram of the distribution of BH transients in Galactic longitude (left) and latitude (right). A 10° bin size in longitude and 5° bin in latitude were used respectively.

Open with DEXTER

Here we revisit the problem of the size of the Galactic population of BHs, based on the currently observed sample, which is heavily affected by extinction. Duerbeck (1984) modelled the density distribution of several interacting binaries in the Galaxy. In a similar way, we want to obtain the density of Galactic BHTs ρ(z) and analyse the height distribution (z; Col. 8 in Table A.1). Only 35 out of the 59 detected BHTs have estimated distances. Because of their transient nature, we have only detected those systems that went into outburst. Therefore, in a given period of time, we have only discovered a fraction of them (e.g. the brightest and/or closest BHTs – with the shortest outburst cycles). Thus, to derive the true z distribution, we first need to find the maximum radial distance at which completeness of the sample is guaranteed, i.e. such that none of the X-ray outbursts are missed.

The completeness of the sample can be obtained from the analysis of the density of objects as a function of the radial distance that is projected onto the Galactic plane (r). The latter is derived from the Galactic latitude and the distance listed in Cols. 5 and 6 of Table A.1.

Here, we have used 31 out of the 35 BHTs with a reported distance (we exclude the four systems with lower or upper limits). In addition, we consider a conservative 50% uncertainty in systems with only a rough estimate of distance. To derive the density of objects (Σ) up to a radii r, we count the number of BHTs lying in cylinders or radius r centred at the Sun and infinite height (h). However, given the large uncertainties with distance, in some cases an object could be placed in more than a single cylinder within errors. To account for this effect, for each object we randomly generate 10 000 values that assume a Gaussian distribution. Here, we naively assume the distance errors follow this sort of distribution, based on the assumption that all of the errors that went into the distance errors were Gaussian. In reality, we have no easy way of knowing this, as the values are from the literature, as determined by many groups, and in many cases are not fully documented. Thus, the final density of objects obtained in a cylinder is given by the median value of all the 10 000 Σ that were obtained in the process, with the error given by the standard deviation. The result is shown in Fig. 4 where the density is represented as a function of the radial distance and normalized to its maximum value. In this figure we find a plateau in Σ up to r = 4 kpc (within errors) and a decrease for r> 4 kpc. We interpret this as an indication that the density of objects is approximately constant with increasing radii up to ~4 kpc. We note here that it is difficult to establish a limit, but the density decreases clearly at r> 4 kpc, whereas it is not clear before this limit. Moreover, using the counting method proposed by Duerbeck (1984), we obtain a more abrupt decrease in Σ for r> 4 kpc. However, we believe this technique is unsuitable for systems with large uncertainties in the distance.

We also perform a complementary analysis based on the X-ray luminosity. The least luminous BHT detected is XTE J1118+480 which reached an X-ray luminosity of 3.6 × 1035erg s-1 in the 210 keV range (Dunn et al. 2010). Therefore, we can assume that no BHT peaked at a lower luminosity. The sensitivity limits by the all-sky monitor (ASM) on RXTE in the 210 keV range have been above 2.4 × 10-10 erg s-1 cm-2 since 1996, which means that any BHT that reached 3.6 × 1035erg s-1 within 3.5 kpc would have been detected (without considering absorption). This is consistent with the method described above and, therefore, we assume that the sample of BHTs is complete out to ~4 kpc.

thumbnail Fig. 4

Variation of the density of BHTs (Σ) in a cylindrical volume of infinite height h with increasing radii r. Above r> 4 kpc the density decreases and, therefore, we infer the sample of BHTs is complete out to r ~ 4 kpc (grey bar).

Open with DEXTER

There are ten objects in this volume (listed in Table 1). However, for the sake of the analysis we should only consider those BHTs detected since the rate of discoveries has become more or less constant. We assume that this has occurred since 1988 (see Fig. 1) when the sky started to be intensively scrutinized by all-sky monitors on-board X-ray satellites. Hence, we eliminate A0620003 from this analysis because it was discovered in 1975 – before the beginning of the assumed constant rate of discoveries. This yields nine BHTs in the solar neighbourhood (r ≤ 4 kpc), discovered at a rate of ~0.3 BHTs yr-1 since 1988. For comparison, in the same time interval, ~46 BHTs have been discovered in the whole Galaxy (which implies a rate of 1.7 BHTs yr-1).

It is expected that the vertical distribution of BHTs will follow the same exponential function as the stellar distribution (Duerbeck 1984). In addition, it is reasonable to assume that the population of BHTs is representative of the parent Galactic population of BHTs, since there is no strong bias against their detection in X-rays (Özel et al. 2010).

Therefore, the space/time density distribution can be approximated by (1)where is the space/time density of objects in the Galactic plane and z0 the scale height of the distribution that is perpendicular to the plane of the Galaxy, both measured in the solar neighbourhood.

Table 1

List of BHTs within r = 4 kpc of completeness.

thumbnail Fig. 5

Variation of the space/time density (ρ) of BHTs with the scale above the Galactic plane z. The solid line represents the Levenberg-Marquardt least-squares fit (Markwardt 2009) to the exponential function shown in Eq. (1).

Open with DEXTER

Using a Levenberg-Marquardt non-linear least-squares fit (Markwardt 2009) of ρ(z), we derive  kpc-3 yr-1 and z0 = 0.123 ± 0.008 kpc (see Fig. 5). This small scale height z0 indicates a clear concentration of systems in the plane.

To obtain the final space density of BHTs ρ(z), we must assume a mean outburst recurrence period (ORP). We have only observed transients for ~50 years and the observed ORPs for scrutinising the X-ray sky are thus biased by this limited time frame. In Fig. 6 we show the frequency of outbursts of all the BHTs over the past 50 years. The majority of systems have shown only one outburst, with only a small fraction showing multiple outbursts. The most extreme cases are H 1743-322, GX 339-4, and 4U 1630-472, which have triggered 10, ~19, and ~20 outbursts in 50 years (some of them were not considered “full outbursts”), respectively. In particular, 4U 1630-472 shows an ORP of 600700 d, which lasted from 100200 d up to 2.4 years (see, e.g. Kuulkers et al. 1997). There is also evidence for additional outbursts produced by A 0620+003 (in 1917, Eachus et al. 1976) and V404 Cyg (in 1938 and 1956, Wachmann 1948; Richter 1989) detected by analysing photographic plates that were made previous to their discoveries. Nevertheless, we included only those outbursts discovered by X-ray satellites (i.e. since 1966) to be consistent and consequently, the observed ORP is biased by the 50 yr time lapse of X-ray astronomy observations. On the other hand, based on an analysis of the mass-transfer rate needed to reach the critical surface density that produces instabilities in the accretion disc, White & van Paradijs (1996) consider it that BHTs have average ORPs above 100 yr. Therefore, the local density at the Galactic plane becomes  kpc-3.

thumbnail Fig. 6

Histogram of the frequency of outbursts detected over 50-yr period for all BHTs.

Open with DEXTER

A rough extrapolation of this local density distribution to the entire Galaxy, assuming no radial dependence, enables us to obtain the total number of BHTs by integrating Eq. (1) in cylindrical coordinates. Using R = 14 kpc as the truncation radius for the Galactic disc and H = 2.5 kpc as the maximum height above the Galactic plane given by MAXI J1659152 (Kuulkers et al. 2013), we find a total number of BHTs in the Galaxy with comparable properties to the systems detected so far, i.e. with peak X-ray luminosities a few tenths of the Eddington luminosity. This is consistent with the results obtained using other techniques (van den Heuvel 1992; Tanaka 1992; White & van Paradijs 1996; Romani 1998) and implies that only ~4% of Galactic BHTs have been discovered.

Our empirical estimate is an order of magnitude lower than the 104 BHTs predicted by Kiel & Hurley (2006) or Yungelson et al. (2006) using population synthesis models. However, we note that our analysis is based on the study of observed systems but limited only to the nine BHTs with reliable distance estimates, which are located in a cylinder of 4 kpc radius centred on the Sun. In addition, we have assumed that the solar vertical distribution (ρ0,z0) can be extrapolated to other regions of the Galaxy. However, the bulge contains ~30% of the stellar mass of the Galaxy, which is confined in a reduced spheroid and it is expected to host a higher concentration of BHTs (Muno et al. 2005). Furthermore, we considered a cylinder with a height defined by MAXI J1659152 (the object with the highest h) but there could be objects located at higher distances over the plane. Finally, we normalized our estimated value to an average recurrence period of 100 yr, which explicitly does not take into consideration any systems with lower accretion rates or longer recurrence periods, nor does it account for a likely population of intrinsically faint X-ray BHTs. Taking in all of the above, we conclude that our crude calculation of the number of BHTs expected in the Galaxy is very conservative and sets a lower limit to the hidden population.

4. Physical properties of dynamical BHTs

thumbnail Fig. 7

Histograms of the 17 dynamically confirmed BHs. Left: extinction-corrected absolute R-band magnitudes in bins of 2 mag. The black histogram denotes confirmed IMXBs. Right: orbital periods in a logarithmic scale using bins of 0.2.

Open with DEXTER

With the information on the apparent quiescent magnitudes, distances, and reddening listed in Tables A.1A.4, we can recover the R-band absolute magnitudes (corrected for extinction and in the AB system) of the BHTs in quiescence. The magnitude distribution (Fig. 7, left) peaks strongly at MR ≃ 4−6, where ~40% of the systems lie. This result is expected since the quiescent spectra are mainly dominated by the light from the donor (mostly K-type stars) with some contribution from the accretion flow (typically 50% in this waveband). From Table A.4, we note a small fraction of IMXBs with A–F spectral-type companions (V4641 Sgr, 4U J1543–475, and GRO J1655–40), which have been indicated in black in Fig. 7.

In addition, from Table A.2 we can revisit the observed distribution of orbital periods that shows a bimodal shape with a gap at ~15 h, known as the bifurcation period (Fig. 7, right). This is driven by different evolutionary paths, with systems above the gap evolving towards longer orbital periods through nuclear evolution of the donor stars, while angular momentum losses shrink the orbit of the binaries below the gap (Pylyser & Savonije 1988; Menou et al. 1999). In this case, the IMXBs are located at long orbital periods, consistent with the presence of giant/subgiant donor stars. The majority of systems have periods between 610 h, which corresponds to main-sequence or slightly evolved K-type donors filling their Roche lobes.

It is noticeable that none of the 59 BHTs listed in Tables A.1 and A.2 present X-ray or optical eclipses (the extragalactic BH-HMXB M33 X-7 is the only one known as showing eclipses). In fact, all the dynamical BHTs found so far have binary inclinations 75° (see Table A.4). However, it is expected that at least 20% of the 17 dynamically confirmed BHs will have i> 75°, taking an isotropic distribution of inclinations into consideration. Therefore, our estimate of the expected distribution of BHTs may be underestimated by 20% and should be , although this percentage is probably smaller than the uncertainty associated with our systematic errors. Narayan & McClintock (2005) propose that the lack of high inclination systems is due to a selection effect, which is produced by the inner accretion disc hiding or obscuring the central BH, making these systems very faint in X-rays and preventing their detection during outbursts. For typical disc-flaring angles of ~12° (de Jong et al. 1996), the outer disc rim will obscure the central X-ray source permanently when viewed at inclinations 78°.

Swift J1357.20933 may be the first object to be seen edge-on according to the optical properties that were displayed during the decay from its 2011 outburst (Corral-Santana et al. 2013), but it did not show either X-ray or optical eclipses. Armas Padilla et al. (2014) and Torres et al. (2015) present alternative explanations to the high inclination that is based on the quiescent X-ray properties and the Hydrogen column density obtained from the Na doublet in outburst, respectively. However, Mata Sánchez et al. (2015) present new evidencet that supports the edge-on configuration.

thumbnail Fig. 8

Distribution of observed compact object masses. The vertical dashed line represents the maximum mass allowed for NS (Fryer & Kalogera 2001). Open circles below that limit represent the masses of the NS compiled by Lattimer & Prakash (2005), extended with updated data from Özel et al. (2012) and Antoniadis et al. (2013). The solid circles indicate reliable BH masses (adopting the values favoured by Casares & Jonker 2014), while arrows indicate lower limits based on mass functions and upper limits to the inclination.

Open with DEXTER

An accurate determination of the binary inclination is crucial for obtaining the mass of the compact object. The mass distribution of black holes has a significant impact on the physics of supernova explosions, the survival of interacting close binaries and the equation of state of nuclear matter. The distribution of masses of compact objects is expected to be smooth because of its correlation with the distribution of their progenitor masses (Fryer & Kalogera 2001). However, the observed distribution shows a gap between neutron stars and BHs between 2−5M (Fig. 8). Unlike Özel et al. (2010, 2012), here we only show with solid circles those systems with the most reliable mass determinations following Casares & Jonker (2014). On the contrary, for those systems with inaccurate masses, we prefer to display large bars which encompass both the minimum and maximum published masses.

The existence of this mass gap is under debate. Özel et al. (2010) argue that selection effects may bias the observed distribution at high BH masses (10 M) but cannot explain the gap that may be related to the physics of supernova explosions. Farr et al. (2011) perform a Bayesian analysis of the BH mass distribution and conclude that their larger sample provides strong evidence for the existence of the gap, reinforcing the results by Bailyn et al. (1998) and Özel et al. (2010). However, Kreidberg et al. (2012) warn about a large source of systematic errors that arise from a possible underestimation of the inclination angle i. This is produced by assuming that the emission from the accretion disc (i.e. the disc veiling) is negligible in the infrared when performing ellipsoidal modelling. They found that, in the case of A0620003, this led to an underestimate of the inclination angle of at least 10°. Because of the cubic dependence of the mass function with sini, this could lead to a considerable overestimate of the BH mass. Correcting the BH mass in GRO J0422+32 from the estimated bias in the inclination, Kreidberg et al. (2012) find that the BH would lie in the gap. However, they also note that if this object is excluded from the analysis, previous conclusions remain intact.

On the other hand, Belczynski et al. (2012) and Fryer et al. (2012) state that the mass gap may be real and it could reveal new insights into the supernovae explosion models. More recently, Kochanek (2014) suggest an alternative explanation that is based on the absence of red supergiants in the range 16.5−25M as progenitors of type IIp supernova. As a result of the weakly bound hydrogen envelopes on these massive stars, they eject the outer layers, which leaves a BH with the mass of the star’s helium core (5−8M). This would explain (i) the lack of supernova progenitors in the 16−25M mass range and (ii) the existence of the mass gap and the typical masses of BHs.

5. Conclusions

We have presented the main properties of a large catalogue of BHTs. Of the 59 BHTs detected in outburst so far, only ~30% have been confirmed as dynamical BHTs. A number of them (35) have distance estimates that allow us to study the Galactic distribution. This in turn results in a population of ~ such systems in the Milky Way. This value agrees with previous estimates that use other techniques but is an order of magnitude lower than theoretical predictions, which are based on population synthesis models. We argue that this value must be considered as a lower limit since it is based on the extrapolation of a small number of systems (nine out of the 59 systems detected so far).

We provide several tables listing the astrometric parameters, distances, number of eruptions, X-ray fluxes at the peak of the outburst and an outburst and a quiescence magnitude, reddening, and orbital periods. For dynamically confirmed systems, we have also detailed the magnitudes in quiescence in every available optical and NIR band, together with their dynamical parameters. In the online version of the catalogue, we have also added finding charts, links to the references and relevant information for forthcoming observations of these systems. We plan to include more information on other wavelengths in the near future (e.g. radio or X-ray states) and update the online catalogue with new targets once discovered.


1

The electronic and most complete version of this catalogue is available on www.astro.puc.cl/BlackCAT although all data will also be available through the Virtual Observatory.

2

It is available at www.astro.puc.cl/BlackCAT and will be continuously updated with more systems and information in other spectral bands.

Acknowledgments

We thank the anonymous referee for useful comments. We acknowledge financial support from CONICYT-Chile grants FONDECYT Postdoctoral Fellowship 3140310 (JMC-S), FONDECYT 1141218 (FEB), Basal-CATA PFB-06/2007 (JMS-C, FEB), “EMBIGGEN” Anillo ACT1101 (FEB), the Ministry of Economy, Development, and Tourism’s Millennium Science Initiative through grant IC120009, awarded to The Millennium Institute of Astrophysics, MAS (FEB) and the Spanish Ministerio de Economía y Competitividad (MINECO) under grant AYA 2013-42627 (JC, TMD, IGMP). T.M.D. acknowledges hospitality during his 2015 visit to IA-PUC. This work makes use of observations from the LCOGT network, the CNTAC programs ID CN2014B-44 and CN2015A-88, the ESO Science Archive Facility under request number jcorral-160882, the GTC Public Archive at CAB (INTA-CSIC), and the Isaac Newton Group archive, which is maintained as part as the CASU Astronomical Data Centre at the Institute of Astronomy, Cambridge and the LCOGT Archive, which is operated by the California Institute of Technology, under contract with the Las Cumbres Observatory. We are thankful to Danny Steeghs and Manuel A. P. Torres for providing us with some of the finding charts and Jorge Andrés Perez Prieto for his help with the creation of the web. We have used the web applications of the FTOOLS (Blackburn 1995) and PIMMS (Mukai 1993) software to create the transformation of the X-ray fluxes. Some peak X-ray fluxes were provided by the ASM/RXTE teams at MIT and at the RXTE SOF and GOF at NASA’s GSFC. This research made use of the MAXI data provided by RIKEN, JAXA and the MAXI team (Matsuoka et al. 2009).

References

Appendix A: Tables of BHTs

Table A.1

Astrometric properties.

Table A.2

Peak X-ray flux and optical/NIR photometric parameters.

Table A.3

The dynamical BHs: photometric parameters in quiescence.

Table A.4

The dynamical BHs: binary parameters.

All Tables

Table 1

List of BHTs within r = 4 kpc of completeness.

Table A.1

Astrometric properties.

Table A.2

Peak X-ray flux and optical/NIR photometric parameters.

Table A.3

The dynamical BHs: photometric parameters in quiescence.

Table A.4

The dynamical BHs: binary parameters.

All Figures

thumbnail Fig. 1

Cumulative histogram of discovered (red) and dynamically confirmed (blue) BHTs as a function of time. Here, we also count Swift J1357.20933 as a dynamical BH. The lifetimes of the main X-ray satellites with all-sky monitor capabilities are shown with black lines.

Open with DEXTER
In the text
thumbnail Fig. 2

Galactic distribution of 35 BHTs with distance estimates (Table A.1), as seen from the Galactic pole. Dynamically confirmed black holes are marked in orange circles, whereas BH candidates are indicated by yellow stars. Distance ranges are represented with bars while the systems with lower limits are indicated by arrows that follow the same colour index. Some systems are overlapped with other. Each faint white ring in the background image represents an additional ~1.5 kpc from the Sun. Objects seem to concentrate along the direction towards the Galactic centre. The dynamical BHT in the upper right of the figure is BW Cir. Background image credit: NASA/JPL-Caltech/R. Hurt (SSC/Caltech).

Open with DEXTER
In the text
thumbnail Fig. 3

Top: the distribution of dynamically confirmed BHs (circles) and BH candidates (stars) in the Galactic plane, using the Hammer projection. (Background image credit: Mellinger 2009). Some of the symbols are overlapped by others, especially in the Galactic centre region. Bottom: histogram of the distribution of BH transients in Galactic longitude (left) and latitude (right). A 10° bin size in longitude and 5° bin in latitude were used respectively.

Open with DEXTER
In the text
thumbnail Fig. 4

Variation of the density of BHTs (Σ) in a cylindrical volume of infinite height h with increasing radii r. Above r> 4 kpc the density decreases and, therefore, we infer the sample of BHTs is complete out to r ~ 4 kpc (grey bar).

Open with DEXTER
In the text
thumbnail Fig. 5

Variation of the space/time density (ρ) of BHTs with the scale above the Galactic plane z. The solid line represents the Levenberg-Marquardt least-squares fit (Markwardt 2009) to the exponential function shown in Eq. (1).

Open with DEXTER
In the text
thumbnail Fig. 6

Histogram of the frequency of outbursts detected over 50-yr period for all BHTs.

Open with DEXTER
In the text
thumbnail Fig. 7

Histograms of the 17 dynamically confirmed BHs. Left: extinction-corrected absolute R-band magnitudes in bins of 2 mag. The black histogram denotes confirmed IMXBs. Right: orbital periods in a logarithmic scale using bins of 0.2.

Open with DEXTER
In the text
thumbnail Fig. 8

Distribution of observed compact object masses. The vertical dashed line represents the maximum mass allowed for NS (Fryer & Kalogera 2001). Open circles below that limit represent the masses of the NS compiled by Lattimer & Prakash (2005), extended with updated data from Özel et al. (2012) and Antoniadis et al. (2013). The solid circles indicate reliable BH masses (adopting the values favoured by Casares & Jonker 2014), while arrows indicate lower limits based on mass functions and upper limits to the inclination.

Open with DEXTER
In the text

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.