Astronomy 162: Professor Barbara Ryden

Thursday, February 13


``When you have excluded the impossible, whatever remains, however improbable, must be the truth.''
- Arthur Conan Doyle

Key Concepts

(1) Black holes can be detected by their gravitational effect on luminous matter.

A black hole, by definition, doesn't emit light; photons can enter the event horizon of a black hole, but cannot emerge from it. If you can't see a black hole, then how do you know it's there?? In order to detect a black hole, you have to be devious and use indirect methods.

The method of detecting dark objects by noting their gravitational effect on luminous objects is a very old one.

(2) Some binary star systems contain a black hole, the remnant of a massive star.

The explosion of an extremely massive star leaves behind a core which is more massive than 3 solar masses. Too massive to become a white dwarf, or even a neutron star, the core collapses to become a black hole. Some extremely massive stars, we know, are in close binary systems with a second star. Thus, some black holes will be in close binary systems as well. (The existence of neutron stars in close binary systems indicates that a supernova explosion will not necessary break up the binary.)

If a black hole exists in a close binary system, it can be detected indirectly. Gas ripped from the companion star will spiral in toward the black hole. As it does so, it is heated up (by tidal heating, and by friction as it rubs against neighboring gas). The hot gas emits X-rays as it undergoes its death spiral to the event horizon. (Once it's inside the event horizon, we don't see the gas any more, but it's highly visible as long as it's outside.) If you are hunting for black holes, it's good to start by looking for binary star systems which are also strong X-ray sources.

To demonstrate typical techniques for deducing the existence of black holes, I will use the story of Cygnus X-1 as a case study. Cygnus X-1 is the brightest X-ray source in the constellation Cygnus. (Also known as the ``Northern Cross'', Cygnus is visible in the summer sky.) X-rays from Cygnus X-1 are seen to vary irregularly - in strong contrast to the regular pulses emitted by X-ray pulsars. The X-rays from Cygnus X-1 can grow substantially brighter or dimmer over time scales as short as 0.01 second. The natural deduction from the rapid variability of Cygnus X-1 is that the X-rays are coming from a region less than 0.01 light-second across. (0.01 light-second is 3000 kilometers, or roughly the diameter of the Earth's Moon.)

The position of the X-ray source Cygnus X-1 coincides with that of a star called HDE 226868. (The fact that it has a boring catalog number instead of a name like ``Alpha Cygni'' will tell you that the apparent brightness of HDE 226868 is not very high.) The spectrum of HDE 226868 reveals that it is a hot blue supergiant. Its mass, if it's a supergiant star, must be substantial; it's estimated to be about 30 solar masses. The distance to HDE 226868 is estimated to be 2500 parsecs. Stars (even hot supergiant stars) don't produce a significant amount of X-rays; their photospheres just aren't hot enough. However, when the spectrum of HDE 226868 is examined, it has the characteristic Doppler shifts of a star in a binary system. The star HDE 226868 is part of a binary star system with an orbital period of 5.6 days. The mass of the unseen companion to HDE 226868 must be, from an application of Kepler's Third Law, at least 7 solar masses.

Astronomers deduce that the unseen companion to HDE 226868 must be the X-ray source Cygnus X-1. Cygnus X-1 is too compact to be a star. It is too massive to be a white dwarf or a neutron star. Anything so compact and so massive must be a black hole. (For more about Cygnus X-1 and other black hole candidates, try the ``Imagine the Universe'' site, sponsored by Goddard Space Flight Center.

(3) Many galaxies, including our own, have a supermassive black hole at their center.

About a dozen other black hole candidates, with masses up to 10 solar masses or so, have been found in binary star systems within our galaxy. (We don't expect too many black hole stellar remnants, given the scarcity of the extremely massive stars which are their precursors.)

An interesting recent development in astronomy is that supermassive black holes (that is, black holes more than a million times the mass of the Sun) are found at the centers of many galaxies. Our own galaxy, for instance, has a black hole with mass 2.6 million Msun at its center, 8000 parsecs away from us. Our own galactic black hole is actually fairly wimpy. The Andromeda Galaxy, 700,000 parsecs away, has a black hole with a mass of 30 million Msun.

How do you go about hunting such supermassive black holes? As usual, you go about it indirectly, by looking at the orbital motion of gas and stars near the centers of galaxies. For instance, the observation of stars near the center of our own galaxy has revealed that they are orbiting a massive object with M = 2.6 million Msun. The closest star to this massive object comes within 17 light-hours (120 A.U.) of its center. The only plausible way to cram 2.6 million solar masses of stuff into a sphere 120 A.U. in radius is to make it all into a single big black hole. (For instance, if you tried to make a cluster of a million neutron stars 120 A.U. in radius, the collisions between neutron stars would make them all merge into a single big black hole within a few hundred thousand years.)

Note that the Schwarzschild radius of a 2.6 million Msun black hole is 7.8 million kilometers, or 0.05 A.U. (that's about 1/8 the size of Mercury's orbit). Thus, even a supermassive black hole would be a small and inconspicuous object, were it not for its gravitational effect on surrounding material.

Galaxies other than our own cannot be viewed in such great detail. However, the Hubble Space Telescope, with its high angular resolution, has been very useful in searching for black holes in other galaxies, as well as in our own.

Prof. Barbara Ryden (

Updated: 2003 Feb 13

Copyright 2003, Barbara Ryden