Final Report Summary - X-RAY BINARIES (Determining the system parameters of low mass X-ray binaries)
Soon after the discovery of the first extra-solar X-ray sources in the 1960s, it became clear that the brightest sources in the sky are the so-called low mass X-ray binaries (named X-ray binaries from now on). In these exotic objects a 'normal' star with a mass similar to the sun is gravitationally coupled to a so-called compact object, i.e. a neutron star or black hole, and orbiting each other in a period ranging from minutes up to a few days. The normal star is losing its outer layer that will settle into a disk around the compact object, the so-called accretion disk. The material in the accretion disk is slowly accreted onto the compact object, and the energy released in the inner-accretion disk (due to friction) gives rise to the observed X-rays.
The detection of these sources opened the opportunity to address a whole range of interesting fundamental questions. First of all, X-ray binaries are the only way to study stellar mass black holes, and by far the best way to determine masses of black holes in general (important to test several predictions of general relativistic effects). The formation of black holes is still not understood very well, and understanding these objects, and in particular their exact masses, is of great importance in our understanding of the evolutionary path that lead to the black hole. Furthermore, X-ray binaries are also excellently suited to study the physics of accretion, i.e. the most efficient form of energy production known (over 10 times more efficient than energy production in the Sun). Accretion is thought to be of great importance in the Universe, playing an important role in the formation of planets and stars, X-ray binaries, and it is even the energy production mechanism for the super-massive black holes observed in some of the most distant galaxies known. Finally, the neutron star X-ray binaries are an excellent way to study the behavior of matter at extreme densities than cannot be produced on earth. Many models exist that predict how matter should behave in neutron stars, but none of these theories can be ruled out yet. Observations of radio pulsars have shown that all neutron star masses cluster around 1.4 Msun (the reason for this is not well understood) and is therefore, the only point that constrain these models. However, since the neutron stars in X-ray binaries have been accreting a significant amount of material for a long time they should be more massive than 1.4 Msun are therefore excellent candidates to further constrain the behaviour of matter at extreme densities.
All these theories have one key parameter in common, the mass of the compact object and/or the companion star, making accurate mass measurement for X-ray binaries extremely important. However, measuring masses in X-ray binaries (but also in general in astronomy) has been notoriously difficult! Only for a very small sub-set of X-ray binaries has this been possible, namely the transient sources that are still relatively bright in optical while in quiescence.
Only about three dozen of the known low mass X-ray binaries are persistently active X-ray sources, but the large majority show transient outburst. These transients suddenly become bright in X-rays (i.e. they reach X-ray luminosities of 10^(36-38) erg/s) for periods of days to years before they start to fade again to quiescent luminosities (<10^33 erg/s). During such an outburst, the optical flux is completely dominated by the outer accretion disk that is many magnitudes brighter than the companion star (note that this is always the case for the persistently active ones), and no signature of the normal star is visible. It is only when an X-ray binary has returned to the quiescent state, and the dominant source in the optical is the normal star again, that it becomes possible to constrain the masses of the compact object and companion star. However, since the X-ray binary becomes also much fainter in the optical while in quiescence these observations are only possible for a very few relatively bright systems.
In the last few years, we developed a technique that allows us to obtain an estimate of the mass while it is X-ray active. It was already known that the reprocessing of X-rays in the inner accretion disk produces ultraviolet He II photons that can act as seeds to excite fluorescence lines, in particular in the Bowen region (around 4620 - 4660 Angstrom) that is visible in the optical. What was not realised was that this mechanism should also excite Bowen lines coming from the surface of the companion star, and show up as narrow emission lines (superimposed on the broad emission lines coming from the outer-accretion disk) in medium resolution spectroscopy. Using large (8 m class and larger) telescopes, we showed that it was possible to detect these narrow emission lines and that they move as a function of orbital period (i.e. they trace the orbital motion of the companion). In combination with emission lines coming from close to the compact object, we were able to constrain for the first times the masses of both the compact object and the companion in persistently bright (in both X-rays and optical) X-ray binaries.
During the course of the Marie Curie European Reintegration Grant (ERG), we mainly focused on determining the system parameters of transient X-ray binaries that are too faint to observe when they are in quiescence. One such object is the neutron star X-ray binary SAX J1808.4-3658 a transient that shows an outburst about two years. From previous outbursts, it was already known that this system is one of the few X-ray binaries that show a pulsar signal, thereby already gives strong constraints on the orbital motion of the compact object. Our observations showed a clear presence of these narrow emission lines in the Bowen region that arise on the surface of the normal star, and from that obtained orbital motion of normal star. This allowed us to constrain the mass of the neutron star for the first time, and despite previous claims that the mass of this X-ray binary must be >>1.4 Msun we showed that it must be very close to this value.
Another X-ray binary transient for which we successfully observed the narrow components in the Bowen region is MAXI J0556-332. This was the very first time this X-ray binary showed an outburst and the first X-ray observations suggested that this target was a black hole binary with an orbital period of about 9.5 hrs. However, our observations show that the orbital movement of the narrow components points toward a true orbital period of about 16.5 hrs. Furthermore, making some very reasonable assumptions our observations also imply that the compact object must be a neutron star and again must have a mass very close to 1.4 Msun.
Two other X-ray binary transients that we are currently still in the process of analysing are Aquila X-1 and MAXI J1305-704. The first one we observed previously during an outburst several years ago but only for a small fraction of the orbital period. Despite the fact that the narrow components were clearly visible, the short observation did not allow us to strongly constrain the orbital motion of the normal star. However, the results did suggest that the compact object (which is already known to be a neutron star) must have a mass >> 1.4 Msun. Now with the new data set covering many orbital phases we have finally have been able to constrain the motion of the normal star. Despite some intriguing changes (apparently the accretion disk has changed between the outbursts), we can show that the neutron star is truly >>1.4Msun. The second system that we are currently still analyzing is claimed to be a black hole candidate with a relatively short orbital period (90 minutes). However, our spectroscopic observations have shown that the orbital period must be >>90 minutes. We think that in combination with our photometric observations that we are currently analyzing, we will be able to obtain the true period and thereby confirm the black hole nature of this system.
Finally, within the framework of the Marie Curie ERG we also observed the persistently bright X-ray binary 4U 1957+11. This system is claimed to be the only black hole binary that is always persistently active, but this has never been truly proved. Although this system is faint in optical, our observations did show the presence of the narrow lines. However, the orbital motion of the these lines suggest that the mass of the compact object is 2 - 4 Msun. Unfortunately, this mass range includes the maximum limit for a neutron star (3.2 Msun) and we can therefore still not prove the true nature of this object. However, these observations did suggest new avenues that we will follow to further limit the mass of this X-ray binary and hopefully in the future we can truly constrain its nature.
The detection of these sources opened the opportunity to address a whole range of interesting fundamental questions. First of all, X-ray binaries are the only way to study stellar mass black holes, and by far the best way to determine masses of black holes in general (important to test several predictions of general relativistic effects). The formation of black holes is still not understood very well, and understanding these objects, and in particular their exact masses, is of great importance in our understanding of the evolutionary path that lead to the black hole. Furthermore, X-ray binaries are also excellently suited to study the physics of accretion, i.e. the most efficient form of energy production known (over 10 times more efficient than energy production in the Sun). Accretion is thought to be of great importance in the Universe, playing an important role in the formation of planets and stars, X-ray binaries, and it is even the energy production mechanism for the super-massive black holes observed in some of the most distant galaxies known. Finally, the neutron star X-ray binaries are an excellent way to study the behavior of matter at extreme densities than cannot be produced on earth. Many models exist that predict how matter should behave in neutron stars, but none of these theories can be ruled out yet. Observations of radio pulsars have shown that all neutron star masses cluster around 1.4 Msun (the reason for this is not well understood) and is therefore, the only point that constrain these models. However, since the neutron stars in X-ray binaries have been accreting a significant amount of material for a long time they should be more massive than 1.4 Msun are therefore excellent candidates to further constrain the behaviour of matter at extreme densities.
All these theories have one key parameter in common, the mass of the compact object and/or the companion star, making accurate mass measurement for X-ray binaries extremely important. However, measuring masses in X-ray binaries (but also in general in astronomy) has been notoriously difficult! Only for a very small sub-set of X-ray binaries has this been possible, namely the transient sources that are still relatively bright in optical while in quiescence.
Only about three dozen of the known low mass X-ray binaries are persistently active X-ray sources, but the large majority show transient outburst. These transients suddenly become bright in X-rays (i.e. they reach X-ray luminosities of 10^(36-38) erg/s) for periods of days to years before they start to fade again to quiescent luminosities (<10^33 erg/s). During such an outburst, the optical flux is completely dominated by the outer accretion disk that is many magnitudes brighter than the companion star (note that this is always the case for the persistently active ones), and no signature of the normal star is visible. It is only when an X-ray binary has returned to the quiescent state, and the dominant source in the optical is the normal star again, that it becomes possible to constrain the masses of the compact object and companion star. However, since the X-ray binary becomes also much fainter in the optical while in quiescence these observations are only possible for a very few relatively bright systems.
In the last few years, we developed a technique that allows us to obtain an estimate of the mass while it is X-ray active. It was already known that the reprocessing of X-rays in the inner accretion disk produces ultraviolet He II photons that can act as seeds to excite fluorescence lines, in particular in the Bowen region (around 4620 - 4660 Angstrom) that is visible in the optical. What was not realised was that this mechanism should also excite Bowen lines coming from the surface of the companion star, and show up as narrow emission lines (superimposed on the broad emission lines coming from the outer-accretion disk) in medium resolution spectroscopy. Using large (8 m class and larger) telescopes, we showed that it was possible to detect these narrow emission lines and that they move as a function of orbital period (i.e. they trace the orbital motion of the companion). In combination with emission lines coming from close to the compact object, we were able to constrain for the first times the masses of both the compact object and the companion in persistently bright (in both X-rays and optical) X-ray binaries.
During the course of the Marie Curie European Reintegration Grant (ERG), we mainly focused on determining the system parameters of transient X-ray binaries that are too faint to observe when they are in quiescence. One such object is the neutron star X-ray binary SAX J1808.4-3658 a transient that shows an outburst about two years. From previous outbursts, it was already known that this system is one of the few X-ray binaries that show a pulsar signal, thereby already gives strong constraints on the orbital motion of the compact object. Our observations showed a clear presence of these narrow emission lines in the Bowen region that arise on the surface of the normal star, and from that obtained orbital motion of normal star. This allowed us to constrain the mass of the neutron star for the first time, and despite previous claims that the mass of this X-ray binary must be >>1.4 Msun we showed that it must be very close to this value.
Another X-ray binary transient for which we successfully observed the narrow components in the Bowen region is MAXI J0556-332. This was the very first time this X-ray binary showed an outburst and the first X-ray observations suggested that this target was a black hole binary with an orbital period of about 9.5 hrs. However, our observations show that the orbital movement of the narrow components points toward a true orbital period of about 16.5 hrs. Furthermore, making some very reasonable assumptions our observations also imply that the compact object must be a neutron star and again must have a mass very close to 1.4 Msun.
Two other X-ray binary transients that we are currently still in the process of analysing are Aquila X-1 and MAXI J1305-704. The first one we observed previously during an outburst several years ago but only for a small fraction of the orbital period. Despite the fact that the narrow components were clearly visible, the short observation did not allow us to strongly constrain the orbital motion of the normal star. However, the results did suggest that the compact object (which is already known to be a neutron star) must have a mass >> 1.4 Msun. Now with the new data set covering many orbital phases we have finally have been able to constrain the motion of the normal star. Despite some intriguing changes (apparently the accretion disk has changed between the outbursts), we can show that the neutron star is truly >>1.4Msun. The second system that we are currently still analyzing is claimed to be a black hole candidate with a relatively short orbital period (90 minutes). However, our spectroscopic observations have shown that the orbital period must be >>90 minutes. We think that in combination with our photometric observations that we are currently analyzing, we will be able to obtain the true period and thereby confirm the black hole nature of this system.
Finally, within the framework of the Marie Curie ERG we also observed the persistently bright X-ray binary 4U 1957+11. This system is claimed to be the only black hole binary that is always persistently active, but this has never been truly proved. Although this system is faint in optical, our observations did show the presence of the narrow lines. However, the orbital motion of the these lines suggest that the mass of the compact object is 2 - 4 Msun. Unfortunately, this mass range includes the maximum limit for a neutron star (3.2 Msun) and we can therefore still not prove the true nature of this object. However, these observations did suggest new avenues that we will follow to further limit the mass of this X-ray binary and hopefully in the future we can truly constrain its nature.