Black Holes | Home | Astronomy | Astrophotography | Science | Space Science | Astrobiology | Science Teaching | Related Links | Contact Me

Astrophysical Black Hole Candidates

Jeff Stanger

One of the most exciting predictions of Einstein’s theory of gravitation (general relativity) is that of the mysterious “black hole”. The extreme curvature of spacetime caused by these objects does not permit radiation to escape from within their event horizon; therefore their existence cannot be confirmed by direct observation. Despite this complication, numerous candidates have been identified and can be divided into four categories: primordial, stellar, intermediate and supermassive black holes3,9. Through the observation of one or more characteristics indicative of black holes, their existence can be inferred to a high degree of confidence.

 The indirect observation of these objects is possible through analysis of spectroscopic, dynamic and temporal data. These observations include: gravitational lensing events, indicative dynamics of stellar black hole binary companions, inner galactic bulge and accretion disk dynamics, characteristic spectroscopic signatures and relativistic jets3,15,20. The key factors that must be established in these observations are: confirmation that the object exceeds the Oppenheimer-Volkoff upper limit for the formation of a neutron star (3 solar masses (M¤)) and that this mass is contained in a sufficiently small region of space7. Studies have indicated that no more than ~1% of the mass currently visible in our galaxy should be comprised of black holes but this implies there may be as many as 109 stellar mass black holes7. Many of the current candidates were initially located via their persistent or transient X-ray emission within a class of objects known as X-ray binaries.

Figure 1: An illustration of the structure of an X-ray binary such as Cygnus X-1. The identification of black hole X-ray binaries is achieved primarily by the determination of a characteristically large mass from the orbital parameters of the stellar companion5. The most noted examples of these objects are Cygnus X-1, LMC X-3 and A0620-004 and these objects fall into the class of stellar mass black holes (3 to 30 M¤). Other supporting evidence is often observed in the form of high luminosities, energies and temperatures that can only be caused by the gravitational potentials surrounding a black hole7. The observed properties of these sources vary greatly and are thought to be determined by the amount of fuel available to the accretion disk around the compact object. These disks are thought to be the source of the X-rays observed in these systems and an illustration of such a system is shown in Figure 1.

 Stable X-ray emitting accretion disks in suitably spaced binaries are formed when the stellar companion expands beyond the size of its Roche lobe. Material from the companion star escapes through the inner Lagrangian point (L1) and spirals inwards towards the black hole1,3. Viscosity plays a crucial role in accretion by removing angular momentum from matter and allowing it to spiral inwards. In this process the matter is heated to millions of degrees and Wein’s law indicates that matter at these temperatures should emit X-rays. Up to 30% of the rest energy may be emitted as X-rays and most strongly in the final moments before material crosses the event horizon3.

 The properties of an accreting disk are determined by the rate of gas accretion. The X-ray emission of both X-ray binaries and active galactic nuclei (AGN) are commonly defined by the states: quiescent, low, intermediate, high and very high. These states are thought to correspond to increasing mass accretion rates and these rates are inferred by the Eddington critical luminosity. Models that combine the following two basic modes of accretion can explain the observed properties of X-ray binaries11. The thin accretion disk model allows gas to radiate efficiently and at relatively low temperatures, producing a predominantly blackbody spectrum. Advection-dominated accretion flow (ADAF) accretes gas that is optically thin, radiates inefficiently, and is quasi-spherical. Because of the negligible loss of energy through radiation, this gas is extremely hot and the resulting spectrum consists of Comptonized synchrotron and bremsstrahlung emission7,11.

 The quiescent state of black hole X-ray binaries has been explained by a model in which the accreting gas is in the form of a thin disk at large radii and an ADAF at smaller radii. The interpretation of the other spectral states of black hole X-ray binaries is less certain, but the following scenario explains most of the observations. The low state is similar to the quiescent state in geometry, but has a higher mass accretion rate. Comptonization is more important and the spectrum is a very hard power law extending to high frequencies. In the intermediate state, the transition radius moves in and the ADAF is limited to a smaller region. In the high state, the thin disk extends all the way down to the marginally stable orbit, and the spectrum is primarily an ultrasoft (long wavelength) X-ray blackbody. Finally, in the very high state, the thin disk appears to develop an active corona, which radiates a significant amount of hard (short wavelength) radiation11.

 The small spatial distribution of high X-ray emitting regions from accretion disks provides further evidence supporting the existence of black holes in these systems. Fast flickering quasi-periodic oscillations (QPO’s) are observed that are not considered to be associated with neutron stars on time scales as short as 1/1000sec. This flickering is thought to come from the inner regions of the accretion disk and analysis indicates an origin of small angular extent. There are currently about 25 X-ray binary black hole candidates that are considered as such with a high degree of confidence7. Several isolated observations of stellar black holes have also been inferred from gravitational lensing events5.

Figure 2: Sequential radio images of the bipolar jets in SS433, one of the brightest X-ray sources in the galaxy. In addition to accretion disks, two X-ray binary stellar mass black hole candidates are most notably associated with relativistic bipolar outflows or jets. These two examples are GRS1915+1059 and SS4333. Very long baseline interferometry (VLBI) radio images of SS433 are shown in Figure 2. Evolution is evident along its bipolar jets and similar images have been gained for GRS1915+105.

The relativistic jets observed in these black hole candidates are thought to be powered by accretion at greater than the Eddington limit. This gives a huge generation of X-rays and enough radiation pressure to expel a portion of the matter along the path of least resistance (the axis of spin). Jets of subatomic particles emit synchrotron radiation at radio wavelengths and undergo relativistic beaming at particle velocities approaching the speed of light. “Superluminal” jets are observed with GRS1915+105 and this effect is caused by relativistic speeds and the relative orientation9. Since these black holes weigh around 14M¤ they have posed many questions about black hole formation due to their relatively high mass for a binary black hole.

Figure 3: Radio and optical images showing bipolar jets and the dusty ring surrounding the ~ 1.2 x 10^9 M¤ supermassive black hole in NGC 426.Accretion disks and bipolar outflows are morphologies also associated with other black hole candidates. Supermassive black holes (106 to 109 M¤) are thought to be the “central engines” behind a broad range of objects known as Active Galactic Nuclei (AGN). A unified theory of AGN has been developed in recent years and encompasses many types of astrophysical objects previously classed separately, such as Quasars, Blasars, BLLacs and QSO’s. Investigating these objects is much more challenging than for nearby sources because of their small angular extent, large distances and intervening material. Despite these challenges there have been successful attempts to image the phenomenon associated with these objects and an example is shown in Figure 311.

Figure 4: A diagrammatic representation of the unified scheme of AGN. Observations from direction B are associated with type 1 AGN and observations from direction A are associated with type 2.

Approximately 1% of all galactic nuclei eject radio-emitting plasma and gas clouds, providing a “cosmic signpost” for the presence of their central engines. AGN are also associated with powerful sources of radio, IR, hard (short wavelength) UV, X-ray and gamma ray emission. As previously mentioned, inferring the existence of a black hole in these objects primarily involves measuring their mass and dimensions. Fortunately some AGN’s have been found to vary in brightness on very short time scales similarly to QPO’s in X-ray binaries. Through spectral analysis on these time scales it has been possible to estimate the dimensions of the central regions of these objects. The conclusion is that these regions are not more than a few light hours in diameter and therefore comparable in size to our solar system. Coupling this small size with the large masses inferred from dynamical studies of galactic nuclei indicates that the most probable object at the centre of these galaxies is a supermassive black hole5.

As the relationship between the observed features of AGN’s and their central engines is better defined we may be able to infer characteristics of black holes when more direct methods are not available. The first important step in this direction was in the formulation of the unified theory of AGN. This generally accepted theory classes these objects as type 1 or type 2 and these classes are broken down further into radio loud and radio quiet. These classes reflect similar objects in different states (mentioned previously for X-ray binaries) due to their accretion rates and at different relative orientations (see Figure 4).

 For exotic regions of the universe, such as in AGN, we rely on our knowledge of absorption and emission mechanisms to explain our observations and build theories. These mechanisms have been successfully attributed to the various regions within typical AGN shown in Figure 4. The observed continuum for a typical AGN is shown in Figure 5 and indicates that features are observed across a wide range of objects. The relation of the predominant features in Figure 5 to mechanisms and regions in AGN is summarised in Table 1.Figure 5: A depiction of the typical features in the continuum observed for many AGN.

Table 1

Continuum Features of AGN3,13


Mechanism and Feature

Dusty ring

Thermal emission = IR bump

Accretion disk

Unpolarised thermal emission = big blue bump

Hot Corona

Compton reflection

Base of Jets?

Inverse Compton scattering = soft X-ray excess


Synchrotron = ‘radio loud’ emission (dotted line)

Relativistic accretion disk

Broad FeKa emission line from fluorescence in the presence of a hard X-ray source = FeKa line

  The observed dynamics and energies specific to spectral features in Figure 5 support the view that AGN are powered by black holes3. In particular, the very broad Fe Ka emission lines are characteristic of a relativistic accretion disk and thereby infer a black holes existance17. In parallel with dynamical determinations, the masses of the black holes in AGN can also be inferred by using the limiting Eddington luminosity to infer the gravitational potential. In recent observations, characteristic jets and signatures of accretion disks have been shown to recur at different scales across the black hole mass range15. In microquasars (such as SS433)17 in our galaxy, superluminal radio features are seen to propagate along the jet shortly after a sudden dip in hard X-ray emission18. This X-ray dip is most likely caused by the in fall of matter into a black hole and remainder of the matter is sent into the jet giving radio bright spots15. These coupled accretion and radio jet outbursts are similarly observed in many AGN systems although on larger time scales due to their increased size6.

 Surveys show that AGN’s were much more numerous and active in the past and this indicates that all galaxies may contain central black holes3. It appears that AGN evolve and become “quiet”, presumably when fuel is less accesable11. This poses the challenge to detect these black holes in “normal galaxies” like our own. Several methods are used to infer the presence of these “dormant” AGN using indirect properties such as mass to light (M/L) ratios and stellar density distributions3. Generally the mass to light (M/L) ratio of elliptical galaxies and globular clusters is between 1 and 10. In the case of the spiral galaxy NGC 3115, it has a M/L ~ 4 which is almost constant throughout the disk but rapidly rises to ~ 40 very close to the central bulge3. This indicates the presence of a large dark mass of millions of solar masses and a black hole is the most plausible candidate. Black holes should also produce a cusp-like density distribution of stars at the centre of galaxies7. Brightness profiles of galactic centres do indeed reflect these distributions and this cannot be explained by K band IR observations3.

Figure 6: The 15.2 year period, orbital motion of S2 around Sagittarius A. A more direct method of “pinning down” a galactic supermassive black hole has been used recently in our own galaxy25. Observations of the star designated S2 around Sagittarius A have indicated that a mass of 3.7 ± 1.5 x 106 M¤ lies at Sagittarius A’s position (Figure 6).

 From the data collected, researchers have ruled out any explanation for this object other than a supermassive black hole. This evidence constitutes some of the strongest evidence to date for a supermassive black hole in our galaxy and within a volume that is thought to be too small to be explained by any other objects.

Intermediate mass black holes (400 – 105 M¤) form a recently discovered population so far thought to populate galactic cores and globular clusters8. These objects were inferred by dynamical analysis and gravitational lensing events. New theories regarding black hole evolution may flow from their discovery as they form a missing link in the universe’s black hole population. There are ~ 150 globular clusters around the Milky Way and from the sample studied so far it appears that these black holes may be common in globular clusters. Intermediate black holes appear to scale well compared to the supermassive black holes in galaxies since the ratio of their mass to that of the host cluster appears constant8. This fact may indicate links in formation processes to the mass of their host cluster. Two examples of clusters in which intermediate black holes have been detected are shown in Figure 7.Figure 7: Globular clusters M15 and M31 G1 that appear to contain intermediate mass black holes.

            Perhaps the most unusual of all black holes are those proposed by British astrophysicist Stephen Hawking. These primordial black holes (PBH’s) would have masses less than a typical asteroid and may be very numerous. There was a time in the early universe when the size of the universe was comparable to the deviations from an isotropic and homogenous universe18. It was in this epoch that suitable conditions existed for primordial black holes to form2.

 Black holes are thought to loose mass over time via hawking radiation5. This quantum evaporation is negligible for stellar and supermassive black holes but small Black holes would evaporate in an explosion of gamma rays in less than 1010 years2. Searches have been conducted to detect the background radiation from PBH’s or the final bursts of gamma rays associated with their quantum evaporations, but their existence remains theoretical. These objects have been indicated as candidates for dark matter that is implied by dynamical and lensing studies of galaxies and clusters of galaxies2.

 From the wide range of astrophysical black hole candidates many similarities have been observed. These similarities indicate that there is convincing evidence for a distinct class of astrophysical object. Using many methods researchers have come to similar conclusions that cannot be dismissed. Although the evidence to date is not absolutely definitive, due in part to the nature of these objects, their existence is assured to a high degree of confidence. Almost 30 years of research has yielded an enormous amount of knowledge regarding black holes. Considering future plans for gravity wave detectors and X-ray interferometers7,20, in conjunction with recent developments, the next 30 years are sure to be just as enlightening.



 1. Bromley, B.C., Miller, W.A. & Pariev, V.I. (1998). The inner edge of the accretion disk around a supermassive black hole. Nature, 391, 54-56.

 2. Carr, B. (1994). Baryonic Dark Matter. Annu. Rev. Astron. Astrophys. 32, 531-590.

 3. Carroll. B.W. & Ostlie. D.A. (1996). An introduction to Modern Astrophysics. Addison-Wesley Publishing Co., USA.

 4. Cowley, A.P. (1992). Evidence for Black Holes in Stellar Binary Systems. Annual Reviews of Astronomy and Astrophysics. 30, 287.

 5. Encyclopaedia of Astronomy and Astrophysics. (2001). Nature publishing group, UK.

 6. Fender, R.P. & Kulkers, E. (2001). On the peak radio and X-ray emission from neutron star and black hole candidate X-ray transients. Mon. Not. R. Astron. Soc. 324, 923-930.

 7. Frolov, V.P. & Novikov, I.D. (1998). Black Hole Physics: Basic Concepts and New Developments. Klewer Academic Publishers, Netherlands.

 8. Gerssen, J. & Van der Marel, R.P.(2002). Hubble Space Telescope Evidence for an Intermediate-Mass Black Hole in the Globular Cluster M15. Accessed at:

 9. Greiner, J., Cuby, J.G. & McCaughrean, M.J. (2001). An unusually massive stellar black hole in the Galaxy. Nature. 414, 522 – 525.

 10. Junor, W., Biretta, J.A. & Livio, M. (1999). Formation of the radio jet in M87 at 100 Schwarzschild radii from the central black hole. Nature, 401, 891-892.

 11. Kembhavi, A.K. & Narlikar, J.V. (1999). Quasars and Active Galactic Nuclei: an Introduction. Cambridge University Press, Cambridge, UK.

 12. Kanbach, G., Straubmeier, C., Spruit, H.C. & Belloni, T. (2001). Correlated fast X-ray and optical variability in the black hole candidate XTE J1118+480. Nature. 414, 180-182.

 13. Koratkar, A. & Blaes, O. (1999). The Ultraviolet and Optical Emission in Active Galactic Nuclei: The Status of Accretion Disks. PASP. 111, 755, 1-30.

 14. Lena. P., Lebrun. F. & Mignard. F. (1998). Observational Astrophysics. 2nd Ed. Springer-Verlag, Germany.

 15. Marcher, A.P. et al. (2002). Observational evidence for the accretion-disk origin for a radio jet in an active galaxy. Nature. 417, 625 – 627.

 16. Melina, F. (2001). X-rays from the edge of infinity. Nature. 413, 25-26.

 17. Mirabel, I.F. & Rodriguez, I.F. (1994). A superluminal source in the galaxy. Nature. 371, 46-48.

 18. Mirabel, I.F. & Rodriguez, I.F. (1998). Microquasars in our galaxy. Nature. 392, 673-676.

 19. Schödel, R. et al. (2002). A star in a 15.2-year orbit around the supermassive black hole at the centre of the Milky Way. Nature. 419, 694 - 696.

 20. White, N. (2000). X-ray astronomy: Imaging black holes. Nature. 407, 146 – 147.

Image Credits

Figure 1 Source unknown.

Figure 2 NRAO / AUI / NSF.

Figure 3 Goddard Space Flight Center (HqL-362)

Figure 4 Adapted from: Kembhavi, A.K. & Narlikar, J.V. (1999). Quasars and Active Galactic Nuclei: an Introduction. Cambridge University Press, Cambridge, UK.

Figure 5 Koratkar, A. & Blaes, O. (1999). The Ultraviolet and Optical Emission in Active Galactic Nuclei: The Status of Accretion Disks. PASP. 111, 755, 1-30.

Figure 6 Schödel, R. et al. (2002). A star in a 15.2-year orbit around the supermassive black hole at the centre of the Milky Way. Nature. 419, 694 - 696.

Figure 7 Gerssen, J. & Van der Marel, R.P.(2002). Hubble Space Telescope Evidence for an Intermediate-Mass Black Hole in the Globular Cluster M15. Accessed at:

Black Holes | Home | Astronomy | Astrophotography | Science | Space Science | Astrobiology | Science Teaching | Related Links | Contact Me