The Shrinking Doughnut Around a Black Hole


Homer Simpson would be sad: recent observations of the binary system of a black hole and its companion star have shown the retreat of the doughnut-shaped accretion disk around the black hole. This shrinking ‘doughnut’ was seen in observations of the binary system GX 339-4, a system composed of a star similar in mass to the Sun, and a black hole of ten solar masses.

As the black hole feeds on gas flowing out from the orbiting star, the change in flow of the gas produces a varying size in the disk of matter that piles up around the black hole in a torus shape. For the first time, the changes in the size of this disk have been measured, showing just how much smaller the doughnut becomes.

GX-339-4 lies 26,000 light-years away in the constellation Ara. Every 1.7 days in the system, a star orbits around the more massive black hole. This system, and others like it, show periodic flares of X-ray activity when gas that is being stolen from the star by the black hole gets heated up in the accretion disk that piles up around the black hole. Over the last seven years, the system has had four energetic outbursts in the last seven years, making it a quite active black hole/stellar binary system.

The material falling into the hole forms jets of highly energized photons and gas, one of which is pointed in the direction of the Earth. It is these jets that a team of international astronomers observed using the Suzaku X-ray observatory, operated jointly by the Japan Aerospace Exploration Agency and NASA, and NASA’s X-ray Timing Explorer satellite. The results of their observations were published in the Dec. 10 issue of The Astrophysical Journal Letters.

Though the system was faint when they took their measurements with the telescopes, it was producing steady jets of X-rays. The team was looking for the signature of X-ray spectral lines produced by the fluorescence of iron atoms in the disk. The strong gravity of the black hole shifts the energy of the X-rays produced by the iron, leaving a characteristic spectral line. By measuring these spectral lines, they were able to determine with rather high confidence the size of the shrinking disk.

Here’s how the shrinking occurs: the part of the disk that is closer to the black hole is denser when there is more gas flowing out from the star that accompanies it. But when this flow is reduced, the inner part of the disk heats up and evaporates. During the brightest periods of the black hole’s output, the disk was calculated to be within about 30 km (20 miles) of the black hole’s event horizon, while during lower periods of luminosity the disk retreats to greater than 27 times further, or to 1,000 km (600 miles) from the edge of the black hole.

This has an important implication in the study of how black holes form their jets; even though the accretion disk evaporates close to the black hole, these jets remain at a steady output.

John Tomsick of the Space Sciences Laboratory at the University of California, Berkeley said in a NASA press-release, “This doesn’t tell us how jets form, but it does tell us that jets can be launched even when the high-density accretion flow is far from the black hole. This means that the low-density accretion flow is the most essential ingredient for the formation of a steady jet in a black hole system.”

Read the pre-print version of the teams’ letter. If you want more information on how the X-rays from the disks around black holes can help determine their shape and spin, check out an article from Universe Today from 2003, Iron Can Help Determine if a Black Hole is Spinning.

Source: NASA/Suzaku press release

7 Replies to “The Shrinking Doughnut Around a Black Hole”

  1. “…even though the accretion disk evaporates close to the black hole, these jets remain at a steady output.”

    “… jets can be launched even when the high-density accretion flow is far from the black hole. This means that the low-density accretion flow is the most essential ingredient for the formation of a steady jet in a black hole system.”

    That’s a rather remarkable finding. Wonder if this relationship may also apply to supermassive black holes?

  2. To follow up Jon Hanford’s last statement:

    What we see in jets of AGN are flares and especially in M87 there are also many knots visible. So this wouldn’t be what I call a steady outflow. On the other hand do we see the jets over large scales (the overall structure not the knots), so this could be a steady “ground”-flow that is superposed by some knots/flares.

    So maybe there could be something alike in jets of SMBHs. Then the knots/flares would be some special events of some extreme extend. Maybe a sudden loss of angular momentum in the disk with accompanied great influx into the BH and outflow through the jet. Could be due to some instabilities in the disk.

    Jets are fascinating objects. Hopefully I’ll be able to work on them in the future 🙂

  3. Dr Flimmer,

    Thanks for your clarification regarding my previous post. I was under the (naive) impression that a high density accretion flow near the event horizon was necessary to launch a jet, so Dr. Tomsick’s remarks seemed counterintuitive to me. And, as you note, jets from SMBHs don’t produce steady, energetic outflows in general.

    Agreed, jets are fascinating objects. Especially when they’re pointing at you (as in 3C 454.3) 🙂

  4. The accretion disk, and this doughnut which is a sort of pre-accretion disk, produces jets by the generation of magnetic fields. These magnetic fields are generated by the different transport properties of nuclei and electrons. In the large gravity field of the BH the differential flow of positive and negative charges are what set up the magnetic field in the ionized gases.

    I can’t state much with certainty, but this appears similar to the general law for accelerator physics, For E the beam energy ahd L the luminosity the rule EL = constant might be applying here. The energetic accretion disk might have a lower luminosity than the “doughnut,” but higher energy. There might then be some reciprocal relationship which maintains a more or less constant “beam” as the MHD of the accretion disk and pre-accretion disk (doughnut) evolves in time.


  5. Well, as far as I know, the “launch mechanism” of jets is rather poorly understood. So what is really needed to launch a jet is quite unclear.
    But I am not an expert (yet ;)) in this field. I’d like to work on it, as I said. But at first I have to finish my studies. Maybe I can work on them for my PhD thesis. That would be great 🙂 .

    Interestingly, jets are found around any object that has an accretion disk (AGN, Herbig-Haro objects). So the mechanism that drives the jets should be similar in all cases. Maybe that is a point to start. Probably it is easier to start with the less energetic jets of the Herbig-Haro objects. On the other hand it could be that jets are a “relativistic” effect, similar to the spin of particles which is also a relativistic effect.
    Mysterious 🙂

  6. I’m familiar with a body of work that relies on studies of ‘microquasars’ and other extreme objects in our galaxy as a proxy for more distant SMBHs. This led to my earlier question about scaling effects with regards to stellar-mass black holes.

    Hopefully direct imaging of the inner accretion disk of our local SMBH, Sgr A*, will provide theorists some hard data for them to accurately define the accretion disk-jet phenomenon (and possibly employ Dr. Flimmer 🙂 ).

Comments are closed.