Chemical processes in the crust of a neutron star

Neutron stars are one of the possible end products of the stellar life cycle. These objects are the compact remnants of once massive stars that burned all their fuel and consequently collapsed under their own gravity and exploded in a supernova. One can thus say that neutron stars are ‘dead stars’: they have no internal resources to replenish the energy that is lost via photons emitted from their surface and neutrinos escaping from their interior. When left alone, neutron stars would therefore slowly fade away over the course of millions of years. However, a neutron star can escape this fate when it is a member of a binary system: it can then regain energy by pulling off and accreting matter from its accompanying star (it can suck life out of its companion, so to say).

Accretion causes a complex network of nuclear reactions inside the neutron star, at a few hundred meters below its surface. These reactions serve as an effective energy supply. When the neutron star is done accreting, however, it will immediately start to loose energy again. Using sensitive X-ray telescopes and dedicated observing techniques, the signatures of these processes can be detected: the temperature of a neutron star rises during accretion episodes but then gradually decreases afterward. In particular, a detailed mapping of the ‘cooling curve’ of a neutron star can give unique insight into its tantalizing interior structure, one of the basic quests of neutron star research.

These cooling studies cannot be carried out for just any neutron star: in order to create detectable temperature differences, a neutron star must be very cold to start with, and/or accrete (i.e., be heated) for a long time. EXO 0748-676 is one of the few sources for which such a study could be carried out. This neutron star started accreting some time between 1980 and 1984 and happily continued to cannibalize its companion star until it suddenly stopped in 2008. We seized this rare opportunity to study a cooling neutron star and embarked on a monitoring campaign using the sensitive Chandra, XMM-Newton and Swift X-ray satellites.

Our X-ray observations revealed that EXO 0748-676 harbors a relatively hot neutron star, as expected given its long history of feasting on its companion. During the first year after it stopped accreting, its temperature was gradually decreasing indicating that the hot neutron star was cooling. However, between 2009 and 2011, we did not detect any significant temperature change any more. To verify that the neutron star had indeed stopped cooling, we obtained a new Chandra observation in 2013. Surprisingly, this new data point showed that the temperature had dropped again, indicating that cooling is ongoing. Excitingly, the apparent plateau of stalled cooling in 2009-2011 could be the result of a chemical process in which light and heavy atoms separate out in different layers.

If this process indeed occurs, it would suggest that the crust of neutron stars has a very ordered, strong structure. That has important implications for several observational properties of neutron stars, including the emission of gravitational waves from deformations in their structure. This result on EXO 0748-676 nicely illustrates how astronomical observations can be used to probe micro-physical processes occurring in the dense interior of neutron stars.

Degenaar, Medin, Cumming et al. 2014, ApJ 791, 47: Probing the Crust of the Neutron Star in EXO 0748-676

Paper link: ADS

 Schematic overview of the interior of a neutron star.  Credit: Chamel & Haensel 2008, LRR 11, 10

Schematic overview of the interior of a neutron star.
Credit: Chamel & Haensel 2008, LRR 11, 10

The accretion flow around a black hole

Black holes are infamous for their relentless gravitational pull through which they drain matter and energy from their surroundings. However, with their enormous power, these tantalizing objects also blast matter back into space via ultra-fast collimated jets and dense winds. Understanding the exact connection between how black holes accrete from – and supply feedback to – their environment is one of the outstanding challenges of modern astrophysics.

X-ray binaries are excellent laboratories to study the eating habits of black holes. In these binary star systems a black hole orbits a Sun-like star close enough to pull off and accrete the outer layers of its unfortunate companion. This accretion process liberates enormous amounts of energy that is emitted across the electromagnetic spectrum. Studying the accretion flow in X-ray binaries thus warrants a multi-wavelength approach.

We recently performed such a study for the newly discovered X-ray binary Swift J1910.2-0546.  In 2012 May the Swift satellite suddenly discovered a new, bright X-ray point source in the sky and very soon it became clear that the X-ray emission was powered by accretion onto a black hole. Using the X-ray and UV telescopes onboard Swift, we continued to monitor this new X-ray binary for about three months. To complement these observations, the source was also closely followed at optical and infrared wavelengths (B, V, R, I, J, H, and K filters) using the 1.3-m SMARTS telescope located at the Cerro Tololo Inter-American Observatory (CTIO) in Chile. Finally, a high-resolution X-ray spectroscopic observation was obtained with the Chandra satellite.

This monitoring campaign allowed us to map out the accretion morphology around the black hole in Swift J1910.2-0546. Firstly, X-ray spectroscopy revealed two peculiarities: although disk winds appear to be ubiquitous in black hole X-ray binaries when they are at their brightest, our Chandra observations did not reveal any emission or absorption features that are the imprints of an accretion disk wind. Since such winds are thought to be concentrated in the equatorial plane, this may imply that we are viewing the binary at relatively low inclination. Moreover, even during the brightest stages of its outburst tracked by Swift, the temperature of the accretion disk did not reach above 0.5 keV (about 6 million degrees Kelvin), whereas most black hole disks are much hotter with temperatures above 1 keV. This could plausibly be a geometrical effect, again suggesting that the inclination angle of the binary is relatively low.

Comparing the overall light curves of the outburst in different wavebands revealed two other striking features. A sharp and prominent flux dip appeared in the X-rays almost one week later than at UV, optical and infrared wavelengths. The detailed properties of this flux dip appear to point to a global change in accretion flow geometry, possibly related to the formation of a collimated jet or the condensation of the inner part of the accretion disk. In addition, when the activity of the black hole started to cease, we found that the X-rays steadily decreased whereas the UV emission suddenly was rising again. The observed strong anti-correlation between the X-ray/UV flux also indicates a global change in accretion flow.

Degenaar, Maitra, Cackett et al. 2014, ApJ 784, 122: Multi-wavelength Coverage of State Transitions in the New Black Hole X-Ray Binary Swift J1910.2-0546

Paper link: ADS

Artist impression of the accretion flow around a black hole.  Credit: NASA/Dana Berry, SkyWorks Digital

Artist impression of the accretion flow around a black hole.
Credit: NASA/Dana Berry, SkyWorks Digital

Challenging theoretical models

Studying the thermal evolution of neutron stars is a promising avenue to gain insight into their structure and composition, and therefore of how matter behaves under extreme physical conditions. These compact stellar remnants are born hot in supernova explosions, but quickly cool as their thermal energy is drained via neutrino emission from their dense interior and thermal photons radiated from their surface. When residing in low-mass X-ray binaries, neutron stars pull off and accrete the outer layers of a companion star. This can re-heat the neutron star and drastically impact its thermal evolution.

The accretion of matter causes a series of nuclear reactions (electron-captures and fusion of atomic nuclei) that produce heat at a rate that is proportional to the rate at which matter is falling onto the neutron star. These processes can significantly raise the temperature of the outer layers of the neutron star, the crust, which then become hotter than the stellar interior, the core. How much heat is released, and at which depth, depends sensitively on the structure and composition of the crustal layers. When the neutron star stops accreting (during so-called quiescent episodes), the crust cools down until it reaches the same temperature as the core again. How fast the crust cools depends strongly on its ability to store and transport heat, and hence on its structure and composition. Studying temperature changes of a neutron star due to accretion episodes thus holds valuable information about the properties of its crust.

In 2010, a neutron star called IGR J17480-2446 started accreting matter from its companion star, temporarily making it the brightest X-ray source in the globular cluster Terzan 5. When it stopped 10 weeks later, we began to follow the neutron star every few months with the Chandra satellite to study any possible changes in its temperature. With our initial observations we discovered that the neutron star was about 300 000 degrees hotter than before it started accreting. This provided the first strong evidence that a neutron star can become significantly heated after just 2 months of accreting matter (rather than >1 year; read more here).

Since this neutron star was heated for a relatively short time, theory predicted that it should cool down rapidly. On the contrary, we have found that the neutron star is still considerably hotter than its pre-accretion temperature 2.5 years later! This indicates that there might be some important ingredients missing in our current understanding of heating and cooling of accreting neutron stars. To be able to distinguish between different possible explanations and hence fully grasp the implications of our findings of IGR J17480-2446, we aim to keep observing this neutron star until it has fully cooled. A new Chandra observation has been scheduled for this purpose in the next year (2014). We are anxious to find out how the temperature of the neutron star has changed by that time.

Degenaar, Wijnands, Brown et al. 2014, ApJ 775, 48: Continued Neutron Star Crust Cooling of the 11 Hz X-Ray Pulsar in Terzan 5: A Challenge to Heating and Cooling Models?

Paper link: ADS

terzan5_edser

Three-color image of the globular cluster Terzan 5, obtained with the Chandra X-ray satellite.

An ultra-magnetized neutron star awakes

These days, astronomers all over the world sleep with one eye open, keeping a close watch of of the supermassive black hole located in the center of our Milky Way Galaxy: Sagittarius A* (Sgr A*). A mysterious gas cloud called “G2” is on a collision course with our Galactic nucleus and may produce some fireworks in the near future (read all about it here).

Imagine the excitement when on April 24 (2013), our daily observations performed with Swift’s X-ray Telescope suddenly detected enhanced activity at the position of Sgr A*. An Astronomer’s Telegram was readily distributed to instantly notify the astronomical community. To everybody’s surprise, however, rapid follow-up observations at infrared and radio wavelengths did not detect anything out of the ordinary and in stead suggested the supermassive black hole remained quiet as always.

The mystery was resolved when right next day, Swift’s Burst Alert Telescope detected a very short (less than a second) and energetic burst of gamma-ray emission. Together with the detection of a pulsed X-ray signal using the brand-new high-energy telescope NuSTAR, this revealed that an otherwise dormant neutron star, located very close to the supermassive black hole, had been revived. This neutron star, named SGR J1745-29, has an extremely strong magnetic field and belongs to the rare class of “magnetars”. So far it is only the magnetar that continues to show fireworks, whereas Sgr A* remains as quiet as it has ever been.

Kennea et al. 2013, ApJ Letters 770, L24: Swift Discovery of a New Soft Gamma Repeater, SGR J1745-29, near Sagittarius A*

Paper link: ADS

Press item: Sky&Telescope feature

Artist’s concept of an explosion on the surface of a neutron star. Credit: NASA/Dana Berry.

Artist’s concept of an explosion on the surface of a neutron star.
Credit: NASA/Dana Berry.

The table manners of the Milky Way’s supermassive black hole

 

Understanding accretion onto supermassive black holes and the associated feedback to their environment lies at the basis of understanding their formation, growth and evolution, the chemical enrichment of the interstellar medium, galaxy evolution, and the formation of large scale structures in the universe. Sagittarius A* (a.k.a. Sgr A*) is a supermassive black hole that forms the dynamical center of our Milky Way Galaxy. Being the most nearby Galactic nucleus, it allows for an unparalleled study of the fueling process of supermassive black holes.

Surprisingly enough, the bolometric luminosity of Sgr A* is about 8-9 orders of magnitude lower than the maximum radiation (the Eddington limit) that can be emitted from the environment of a supermassive black hole with a mass of 4 million times that of our Sun. Its faintness is particularly puzzling because nearby dense star cluster are thought to supply enough matter to serve as a grant banquet for Sgr A*. However, it appears that our Galactic nucleus is on a diet.

Nevertheless, it appears to crave for an occasional snack; the relatively steady quiescent radiation of Sgr A* is, however, occasionally punctured by hours-long flares during which the X-ray emission increases by 1-2 orders of magnitude. These events are likely related to small accretion events or magnetic processes. Most excitingly, the time scale involved with these phenomena suggest that they must be originating very close to the black hole (within approximately 15 Schwarzschild radii). A few dozens of X-ray flares have been detected from Sgr A* by using the Chandra and XMM-Newton satellites. The far majority of these are relatively weak; only on 4 occasions was the emission observed to increase more than 100 times the steady base level.

We investigated nearly 800 observations of the center of our Galaxy that were obtained with the X-ray Telescope onboard the Swift telescope between 2006 and 2012. In these 6 years of monitoring data we discovered a total of 6 bright X-ray flares from Sgr A* during which the emission increased by a factor of 100. Owing to its uniquely dense sampling, the Swift campaign more than double the number of observed bright X-ray flares from our supermassive black hole. This allowed to constrain the recurrence rate of these events, and made an unbiased comparative study of their spectral properties possible for the first time. Having mapped out the long-term X-ray behavior of Sgr A* with Swift provides an important calibration point to assess whether the activity of our supermassive black hole is going to change as the result of its interaction with an approaching gas cloud (read more about this upcoming exciting event here).

Degenaar, Miller, Kennea, Gehrels, Reynolds, Wijnands 2013, ApJ 769, 155: The X-Ray Flaring Properties of Sgr A* during Six Years of Monitoring with Swift

Paper link: ADS

The Swift monitoring website: www.swift-sgra.com

Press: German radio interview

Three-color accumulated Swift X-ray Telescope Image of the Galactic center (2006-2014).

Three-color accumulated Swift X-ray Telescope Image of the Galactic center (2006-2014).

The effects of a violent thermonuclear burst

Thermonuclear X-ray bursts manifest themselves as intense flashes of X-ray emission that have a duration of seconds to hours, during which a total energy of approximately 1039 to 1042 erg is radiated. These events are caused by unstable thermonuclear burning that transforms hydrogen and/or helium that has falling onto the surface of a neutron star into heavier chemical elements. X-ray bursts are a unique signature of neutron stars in low-mass X-ray binaries. In such interacting binary star systems, a Roche-lobe overflowing late-type companion star feeds matter to the compact object via an accretion disk.

There is a delicate connection between the properties of X-ray bursts and that of the accretion flow. On the one hand, the rate at which mass is accreted onto the neutron star determines the duration, recurrence time, and radiated energy of the X-ray bursts, whereas the accretion geometry can strongly influence the observable properties. On the other hand, it has been proposed that particularly powerful X-ray bursts may be able to influence the accretion flow.

IGR J17062-6143 is an X-ray source that was discovered in 2006, but remained unclassified until the Burst Alert Telescope onboard the Swift satellite detected an X-ray burst in 2012. This unambiguously identified IGR J17062-6143 as a neutron star low-mass X-ray binary. But not just any ordinary neutron star. The X-ray burst was highly energetic (classifying as a so-called intermediately long X-ray burst), and displayed three very unique features that indicate that the explosion was violent enough to disrupt the accretion disk surrounding the neutron star.

Firstly, the 18-min long X-ray burst displayed dramatic, irregular intensity variations that were clearly visible in the X-ray light curve. Similar fluctuations have only been seen on a handful of occasions. They are likely caused by swept-up clouds of gas or puffed-up structures in the accretion disk. The time-scale of the fluctuations suggest this gas is located at a distance of approximately 103 km from the neutron star. Secondly, the X-ray spectrum of the X-ray burst showed a highly significant emission line around an energy of 1 keV (most likely a Fe-L shell line). This emission feature can be explained as irradiation of relatively cold gas. The width of the line suggests that this material is located at a distance of 103 km from the neutron star. Thirdly, significant absorption features near 8 keV were present in the X-ray spectrum (in the Fe-K band). These likely result from hot, ionized gas along the line of sight. Fitting these features with photo-ionization models points towards a similar radial distance as inferred from the emission line and the light curve fluctuations. Spectral emission and absorption features have never been (unambiguously) detected during an X-ray burst, making this a highly exciting discovery.

In conclusion, three independent observational features suggest that the energetic X-ray burst from IGR J17062-6143 swept up gas along our line of sight, out to a distance of roughly 1000 km from the neutron star (approximately 50 gravitational radii). This provides strong evidence  that powerful X-ray bursts can indeed disrupt the (inner) accretion disk.

Degenaar, Miller, Wijnands, Altamirano, Fabian 2013, ApJ 767, L37: X-Ray Emission and Absorption Features during an Energetic Thermonuclear X-Ray Burst from IGRJ17062-6143

Paper link: ADS

An artist impression of an interacting binary. Image credits: David. A. Hardy / STFC

An artist impression of an interacting binary.
Image credits: David. A. Hardy / STFC

Real-time cooling of a neutron star?

Neutron stars are the densest directly observable stellar objects in our universe and constitute ideal astrophysical laboratories to study matter under extreme physical conditions: immense gravitational fields, ultra-strong magnetic fields, vigorous radiation fields, and supra-nuclear densities. Many observable properties of neutron stars are set by the structure and composition of their crust. A promising way to investigate the properties of these outer stellar layers is to study neutron stars in low-mass X-ray binaries.

In these interacting binary star systems, the neutron star pulls off and accretes matter from a companion star that has a mass comparable to (or lower than) that of our Sun. Many of such X-ray binaries are transient and exhibit outbursts of accretion that last only a few weeks. These are interleaved by years-long episodes of quiescence during which little or no matter is being accreted onto the neutron star. During these quiescent episodes the thermal heat radiation from the glowing neutron star surface becomes visible. This effectively serves as a thermometer of the neutron star and provides a powerful probe of its interior properties.

Sensitive X-ray instruments aboard the Swift, Chandra and XMM-Newton satellites have revealed that outbursts of accretion can severely affect a neutron star’s temperature. Using sophisticated and tailored observing strategies, it has been shown that it causes the crust of a neutron star to be heated to millions of degrees Kelvin. This heat is produced in a cascade of nuclear reactions, including fusion of atomic nuclei (due to the high matter density), and other chemical processes. Once neutron stars stop swallowing matter, the crustal layers slowly cool until they return to their pre-outburst temperature after several years. Both the heating and the cooling encode unique information about the structure and composition of the neutron star’s crust.

With the aim to study the cooling-down of the neutron star in an X-ray binary called XTE J1709-267, we targeted this object in 2013 September shortly after it exhibited an accretion outburst, using the Swift and XMM-Newton satellites. We were expecting to see a gradual fading over the course of several years. Much to our surprise, however, we found that the thermal radiation from the neutron star was rapidly fading during our 8-hour long XMM-Newton observation. An intriguing explanation for this is that we were witnessing fast cooling of the very outer layers of the neutron star in real-time.

If this interpretation is correct, the time-scale of the observed decay places new, strong constraints on the amount of heat that was generated inside the neutron star, and at which depth. When taken at face value, the findings indicate that one particular process that causes atomic nuclei to separate out in different layers, is important in the heat generation. This has important implications for our understanding of the crust structure of accreting neutron stars. A plausible alternative explanation is that the rapid fading was caused by a rapid reduction of the matter supply onto neutron star, hence that our observations tracked the cessation of the accretion flow marking the end of the outburst.

Degenaar, Miller, Wijnands 2013, ApJ Letters 767, L31: A Direct Measurement of the Heat Release in the Outer Crust of the Transiently Accreting Neutron Star XTE J1709-267

Paper link: ADS

An artist impression of an X-ray binary.  Credit: Stuart Littlefair.

An artist impression of an X-ray binary.
Credit: Stuart Littlefair.

Swift: A neutron star finding machine

Many neutron stars reside in X-ray binaries, where they pull off and consume matter from a companion star. As the accreted material accumulates on the surface of the neutron star, the temperature and pressure rise, what can eventually lead to a gigantic explosion: a thermonuclear X-ray burst. These are very energetic, bright flashes of X-ray emission that can last from a few seconds to a few hours. These are a unique characteristic of neutron stars.

NASA’s Swift satellite carries instruments that can detect X-ray, ultra-violet and optical emission from astronomical objects. In addition, it is equipped with a Burst Alert Telescope (BAT). This instrument has a very wide field of view of about 2 steradians and monitors a large part of the sky with the aim to detect (rare) energetic events. The BAT has proven to be an very suitable instrument to detect thermonuclear X-ray bursts from accreting neutron stars.

Some neutron stars display X-ray bursts only very rarely (maybe only once every year). These events can be easily missed, so that the neutron star can remain hidden for a very long time. Indeed, Swift’s BAT has detected several thermonuclear X-ray bursts from previously unknown X-ray sources. In some cases, the BAT picked up an X-ray burst from an X-ray sources that had been discovered before, but was not known to harbor a neutron star. At present, approximately 100 X-ray bursting neutron stars are known (see this list of Galactic X-ray bursters and the MINBAR catalog for overviews). A significant fraction of these (about 10) have been discovered by the Swift satellite.

In 2011 and 2012, the BAT helped identify four new neutron stars via the detection of their thermonuclear X-ray bursts.

Degenaar, Altamirano, Wijnands 2012, Astronomer’s Telegram 4219: IGR J17062-6143 is likely a bursting neutron star low-mass X-ray binary

Paper link: ADS

Degenaar, Wijnands, Reynolds et al. 2014, ApJ 792, 109: The Peculiar Galactic Center Neutron Star X-Ray Binary XMM J174457-2850.3

Paper link: ADS

Degenaar, Linares, Altamirano, Wijnands 2012, ApJ 759, 8: Two New Bursting Neutron Star Low-mass X-Ray Binaries: Swift J185003.2-005627 and Swift J1922.7-1716

Paper link: ADS

 

Artist impression of the Swift satellite. Credit: NASA.

Artist impression of the Swift satellite.
Credit: NASA.

Staring at the center of the Milky Way

The region around Sagittarius A*, the supermassive black hole that represents the dynamical center of our Milky Way Galaxy, harbors a large number of accreting neutron stars and black holes. Between 2005 and 2008, we targeted this region every few months using the X-ray instruments onboard the Chandra and XMM-Newton satellites. The main objective of this monitoring campaign was to study the behavior of transient X-ray binaries. These spend most of their time in a dim quiescent state, during which they often can not be detected, but experience occasional outbursts of bright X-ray emission when the neutron star or black hole pulls off and accretes matter from its companion star.

Our observations covered a region of 1.2 square degree around Sagittarius A* that contains 17 known X-ray transients, 8 of which were active during our campaign. We performed a detailed study of the energy distribution and temporal variations of their X-ray emission. From one of the active neutron stars we detected two thermonuclear explosions, which occurred within a time interval of only 3.8 minutes. Such a short repetition time is only rarely seen and poses a challenge for theoretical models. In addition, we discovered a previously unknown X-ray source, which we tentatively classify as an accreting white dwarf.

Most remarkably, the majority of X-ray transients located near Sagittarius A* are considerably fainter during outburst than is usually seen for accreting neutron stars and black holes. One possible explanation for their sub-luminous character is that these X-ray binaries have very small orbits, in which the compact primary and their companion revolve around each other in less than two hours. Finding such binaries is of particular interest, because they are thought to be strong sources of gravitational waves. The existence of gravitational waves is one of the predictions of Einstein’s theory of General Relativity, which future space-missions hope to prove.

Degenaar, Wijnands, Cackett et al. 2012, A&A 545, 49: A four-year XMM-Newton/Chandra monitoring campaign of the Galactic centre: analysing the X-ray transients

Paper link: ADS

Chandra X-ray image of the center of our Milky Way Galaxy.  Credit: NASA/Wang et al. 2002.

Chandra X-ray image of the center of our Milky Way Galaxy.
Credit: NASA/Wang et al. 2002.

Quiescent but not quite?

The X-ray binary Swift J1749.4-2807 contains a neutron star that rotates around its own axis at a dazzling rate of 518 times per second. To date, only 14 of such fast spinning accreting X-ray pulsars are known. Amongst these, Swift J1749.4-2807 is the only one that shows eclipses: a temporary dramatic drop in the X-ray emission that lasts for approximately 36 minutes and repeats every 8.8 hours. These are caused by the companion star that periodically moves into our line of sight, thereby blocking the X-ray bright central part of the binary.

The unique combination of X-ray pulsations and eclipses makes Swift J1749.4-2807 a particularly promising target to precisely constrain the mass of the neutron star. This is one of the key objectives of modern astrophysics. We used the European satellite XMM-Newton to study the source in quiescence, when the accretion is thought to have switched off and the surface of the neutron star may become directly visible. Quiescent X-ray observations are an important aspect of the challenge to accurately constrain the mass of the neutron star.

Contrary to that seen for the majority of neutron stars, we found that the quiescent X-ray spectrum of Swift J1749.4-2807 consists primarily of high-energy (> 2 keV) photons and shows no evidence for heat radiation that comes from the surface of the neutron star. Its unusual properties can possibly be explained if matter continues to fall onto the neutron star in quiescence. This severely complicates the determination of its mass. It is of utmost importance to understand whether quiescent accretion is common amongst neutron star X-ray binaries.

Degenaar, Patruno, Wijnands 2012, ApJ 756, 148: The Quiescent X-Ray Properties of the Accreting Millisecond X-Ray Pulsar and Eclipsing binary Swift J1749.4-2807

Paper link: ADS

Discovery of eclipses in Swift J1749.4-2807 (2010): NASA press release

Schematic representation of the eclipsing binary Swift J1749.4-2807. Credit: NASA/GSFC.

Schematic representation of the eclipsing binary Swift J1749.4-2807.
Credit: NASA/GSFC.