M33 X-7 is way too massive. Here’s why.
The most massive stars in the local universe often form in binaries. This binary challenges models of stellar evolution.
X-ray binaries are some of the most energetic star systems known. They consist of a normal star and a compact companion — either a neutron star or a black hole. This small but massive object, which would otherwise be dim and hard to detect, captures matter from the primary star and forms a disk of hot gas. Energy from the infalling matter leads to the emission of X-rays, often forming extremely energetic jets at the poles. These systems are massive, luminous, and important sources of X-rays.
One of the most massive X-ray binaries known is M33 X-7, located in the Triangulum galaxy (itself designated M33). It consists of a hot O-type star about 70 time as massive as the Sun, orbiting with a black hole of about 15.5 solar masses. These two objects are extraordinarily massive, even for the objects in X-ray binaries — so massive, in fact, that traditional theories of the evolution of such systems simply didn’t work.
Discovery and mass measurements
Although M33 X-7 had been known since the 1980s, precise measurements of the system weren’t made for several decades, and its true nature wasn’t known until the mid-2000s. In 2006, Pietsch et al. presented data from the Chandra X-ray Observatory that showed the expected eclipses of the system, where one body passes in front of the other. With tighter constraints on the properties of the O-type star, as well as better measurements of the eclipses, they found that the compact object must have had a mass of at least 9 solar masses, which meant it could only be a black hole. Several other facts supported their argument:
- There were no periodic X-ray pulsations observed in the light curves, as would be expected if the compact object was a neutron star.
- Fits of the X-ray luminosity indicated a high accretion rate, which seemed consistent with a black hole.
- The X-ray spectra, as well as X-ray variability observed in between eclipses, indicated the presence of accretion disk behaving as it would around a black hole.
In short, it seemed pretty certain that M33 X-7 contained a black hole. However, the minimum mass measurement of 9 solar masses wasn’t unusual for a black hole formed from the collapse of a star, and would make the system a fairly ordinary X-ray binary.
This changed the next year. Orosz et al. (2007) performed a more detailed spectroscopic analysis to determine the mass of the black hole. Spectroscopy involves measuring the strength and location of certain emission or absorption lines in an object’s spectrum. If the object is moving away or towards the observer, the lines should appear shift to higher or lower wavelengths. By studying how the lines change over time, astronomers can put together something called a radial velocity curve, explaining the motion of the system. Using the laws of orbital mechanics, they can then determine the mass of the other object.
Astronomical spectroscopy is used in a variety of subfields, including exoplanet research. In this instance, however, it was applied to the O-type star in the system, using the Gemini Observatory in Chile. The radial velocity curves produced by the group indicated that the mass of the black hole must be 15.65 +/-1.45 solar masses — greater than that of any previously known stellar-mass black hole in an X-ray binary, and known to greater precision, because M33 X-7 was the first such system exhibiting eclipses.
This alone would not be too unusual, were it not for that O-type star. Its mass was roughly 70 solar masses; furthermore, the two objects were found to be separated by only 42 solar radii. A black hole of 15.65 solar masses would have required a progenitor so massive that it would have grown to touch the companion star, forming a common envelope shared by both stars. The problem is, such a scenario would either have resulted in a merger between the stars, caused by gas drag in the envelope, or extreme mass loss, to the point that the massive star could not have formed such a black hole.
One way around this problem would be that the common envelope phase started only when the primary star finished fusing helium in its core. The main issue with this idea? It would have had an initial mass of about 100 solar masses. Stars this massive lose mass at a rate billions of times that of the Sun, though, and this progenitor star would, again, have lost far too much mass.
A Wolf-Rayet progenitor?
Within the next two years, a theory arose that would explain the mass discrepancy. Two separate groups proposed variants of it: Abubekerov et al. (2009) and Valsecchi et al. (2010). Both suggested that the system began with a high-mass primary star (perhaps 100 solar masses) and a somewhat lower-mass secondary star. As time went on, the high-mass star would expand as hydrogen burning continued, until it was able to transfer matter to the secondary; the two would then move closer together.
Once most of the primary’s outer envelope had been transferred — cutting its mass in half, and perhaps doubling the companion’s mass to 70 or 80 solar masses — it would evolve into what is called a Wolf-Rayet star. These are violent stars with strong stellar winds; they’ve moved off the main sequence and lost most of their hydrogen. Both groups suggested that the star could eventually be stripped to its core by the wind, with the orbit slowly increasing. Eventually, the core would collapse into a black hole, retaining most of its remaining mass. This object could indeed end up with about 15 solar masses of matter.
Furthermore, it was suggested that other known systems could display the same behavior in the future — or already are. Abubekerov et al. mentioned IC 10 X-1, a similar system with an even more massive black hole, as a possible evolutionary counterpart, while Valsecchi et al. suggested that a number of known binary stars could end their lives this way. Binaries of O-type main sequence stars and Wolf-Rayet stars are known, such as some in the famous massive star cluster R136.
These systems will evolve quickly by stellar timescales, but will still take millions of years to reach the same stage of life as M33 X-7. Nonetheless, if systems like IC 10 X-1 can be monitored in more detail, it’s possible that we can find more examples of these rare but extraordinary subclass of X-ray binaries.
