Eta Carinae is very often referred to as one of the most remarkable stellar objects that are close enough to be observed in great detail. Although it is an object of high astronomical interest and there is a significant amount of data collected about it, multiple observational features remain poorly understood. This blog is my final project for an astrophysics class, and in it I want to outline some of the most characteristic features of Eta and explain how they are being made sense of, in a way that is accessible for people with little knowledge of physics.
Bednarek and Pabich. 2011. “High-energy radiation from the massive binary system Eta Carinae”. Astronomy & Astrophysics 530 A49.
Damineli. 1999. “Eta Carinae: Binarity Confirmed”
Davidson and Humphreys. 1997. “Eta Carinae and its environment”. Annual Review of Astronomy and Astrophysics 35: 1-32.
Davidson. 2002. “Chandra meets Eta Carinae”. Astronomical Society of the Pacific. Provided by the NASA Astrophysics Data System.
Farnier, Walter and Leyder. 2010. “Eta Carinae: a very large hadron collider”. Astronomy & Astrophysics 526 A57.
Madura, Gull, Owocki, Groh, Okazaki and Russel. 2011. “Constraining the Absolute Orientation of Eta Carinae’s Binary Orbit: A 3-D Dynamical Model for the Broad [Fe III] Emission”
Ryden and Peterson. 2010. “Foundations of Astrophysics” Pearson Education, Inc: San Francisco.
Smith and Gehrz. 1998. “Proper Motions in the Ejecta of Eta Carinae with a 50 year baseline”. The Astronomical Journal, 116: 823-828.
Thackeray. 1956. “The Distance and Absolute Magnitude of Eta Carinae”. Notes from observatories.
η Car is a good example of how some astrophysical objects are far from well understood. It is a remarkably well-studied star, yet as astronomers come up with explanations for modern observations, they open a comparable amount of new questions.
Eta is an extraordinary object. It is extreme in its parameters (among the most luminous, most massive stars we know of, it has the brightest know stellar wind at mm wavelengths, and an elegant bipolar nebula). It unites multiple points of interest in astrophysics: stellar physics (instability and the Eddington limit), stellar evolution (extreme case of LBV), gas dynamics (in the bipolar flows of the Nebula), and the prospect of going hypernova, among others
I hope this blog was useful in highlighting some of the particularities and surprising features of Eta Carinae. I hope it has at least done the star some justice in how interesting it is!
η Car can go supernova any minute now. Of course, any minute in astronomical terms means some time within the next few thousand years. But the neutrino detectors are all ready. Supernovae are rare events: about one per century in our galaxy (Ryden and Peterson, 2010). The most recent “naked eye” supernova was supernova 1987a in the Large Megallanic Cloud.
In fact, Eta is a candidate for a hypernova explosion, or type 1c supernova, which refers to the supernovae of the most massive stars, with masses between 100 and 300 times that of the Sun. Decaying 56Ni is believed to provide much of a hypernova’s light, with an energy output of over 1048J. The core of a hypernova collapses directly into a black hole. If Eta does go hypernova, it will be visible with the naked eye, and be of brightness comparable to that of the full moon.
For stars, to a first order approximation, mass is destiny. Given η’s current size and composition, we know that it started with a very large mass. Since it is extraordinarily luminous, hot, and energy-releasing, stars like Eta can only burn the fuel in their core at that rate for so long, implying a very short lifetime (of only a few million years).
The Hertzsprung-Russel diagram is a scatter of stars that shows the relationship between their luminosity and their spectral type (and thereby, their surface temperature). Stars change places in the HR diagram as they evolve. A very massive star like Eta would be likely to follow this evolutionary sequence (Davidson and Humphreys, 1997):
After leaving the main sequence, a Luminous Blue Variable would expand at a roughly constant luminosity, moving to the right in the HR diagram. Its luminosity to mass ratio (L/M) is of the order of the Eddington Limit, which makes the star potentially unstable. When it reaches some critical radius or surface temperature, there occurs a LBV eruption (whose causes are not yet understood) and the star loses mass. The mass loss causes the star to shrink and move back to the left of the HR diagram, and it also gives way to further instability as the L/M radius is increased. Eventually, Eta and similar stars might become a Wolf-Rayet star (characterized by large size, high mass loss rate through stellar wind, and a very high surface temperature of 25,000-50,000K). Its instability makes it unlikely that η will “experience the pleasures of supergianthood” (Davidson and Humphreys, 1997, p.8).
Eta is one of the very few known examples of LBV, along with P Cygni, the Pistol star, and less that twenty other. It is an extraordinarily rare object. LBV, in general, have features, such as rapid mass loss, that are empirically rather than theoretically understood. Moreover, Davidson and Humphreys raise the point that the Great Eruption was more dramatic than other well-studied LBV events, and open the possibility that η is an extreme case of LBV, or an “even more mysterious” object.
For a long time, astronomers were not quite sure on how to explain some of the properties of η Car. That it was a system of two stars was something astronomers had been considering for some decades, but it was not until as recently as 2005 that Eta was confirmed to be a binary system.
There is strong evidence that supports this insight. The models that predict an orbiting period of 2022±1.3 days accurately predict features of multiwavelength observations (radio, millimeter, optical, near-infrared, and X-ray data observed over the past few decades, Farmer, 2010). Moreover, the “line intensity and the radial-velocity curve display a phase-locked behaviour implying that the energy and dynamics of the event repeat from cycle to cycle” (Damineli, 1999, p.2).
The consensus is that the first component of the system, ηA, is a Luminous Blue Variable (LBV), and the second is probably a late-type nitrogen-rich O star. The semimajor axis of the orbit is 16.64 AU, with a very high eccentricity e~0.9. The primary star’s radius is estimated to be in the range 0.7-1 AU (Farnier, 2010). From spectroscopic observation it has been inferred that ηA has a current mass ≥ 90 M⊙, and a stellar wind with a mass-loss rate ~10-3 M⊙ yr-1 and terminal speed of ~500 km s-1 (Madurai et al., 2011). It has a luminosity 4.5×106 L⊙, and a surface temperature of 20,000K (Bednarek, 2011). We do not have access to such certainty in the parameters of its companion ηB, since ηA dwarfs its emission, but we know that its luminosity is of the order of 9×105 L⊙, its surfece temperature, ~40,000K, its radius ~1010m, and its mass-loss rate a more modest 10-5 M⊙ yr-1 (Bednarek, 2011).
Because both members of the system are massive stars producing stellar wind, a shock front is produced where the winds meet, causing the X-ray emission seen in Chandra’s photographs. A binary star system of these characteristics is called a colliding wind binary.
In astronomy, luminosity is the amount of electromagnetic radiation a body radiates over time. The luminosity of astrophysical objects like stars is often given in terms of solar luminosities, where 1 L⊙=3.846×1026W.
The total luminosity of η Car is found to be 106.7 L⊙, for a distance of 2,300 pc (Davidson and Humphreys, 1997). As I mentioned before, there exists a theoretical limit for the mass of a star at a given luminosity, the Eddington Limit. This limit has to do with maintaining hydrostatic equilibrium, i.e., balancing outward radiation pressure with inwards gravitational force. This limit determines that the present-day stellar mass of Eta must be ≥ 90 M⊙. Its luminosity is appropriate for an evolved star with initial mass ~160M⊙, and given that these stars lose much of their mass during evolution, now a good estimate for Eta’s mass would be ~120M⊙ (Davidson and Humphreys, 1997).
During the Great Eruption, Eta’s luminosity rose to ~107.3L⊙, exceeding the Eddington Limit by a factor on the order of 4 (Davidson and Humphreys, 1997). At this time, Eta’s photosphere must have been as big as the orbit of Saturn.
In addition to the fact that we see it in the sky, we receive a lot of information about η from non-visible wavelengths. η has an unusually large present-day IR luminosity, and hot thermal X-rays.
Most of Eta’s luminosity today, in fact, emerges at IR wavelengths, in a comparable amount to what was seen in visual wavelengths during the Great Eruption. This suggests that most of the energy from Eta is absorbed by the dust in the nebula, and is re-radiated as thermal IR emission (Davidson and Humphreys).
NASA’s X-ray observatory has been key to providing data that has made astronomers certain that Eta is a “colliding wind binary”, as opposed to a single star. In the image, the X-rays reflected by the optical nebula (in yellow, inside) come from very close to the star itself, and are generated by the high-speed collision of wind flowing from ηA’s surface with the wind from the companion star. The yellow debris is a gas flow that exists outside the Humunculus Nebula, whose material includes nitrogen that has formed in the nucleus of the star and was dragged up to the surface.
We have observations from η Car dating back to the 17th century. In 1677 English astronomer Edmon Halley noted it as a fourth magnitude star (Encyclopaedia Britannica). From that time until the 1830s, it was reported as a star between the second and fourth magnitudes. In 1838, John Hershell observed it as a first magnitude star (Encyclopaedia Britannica). For the next 20 years, it fluctuated between first and zero magnitude, reaching its peak brightness mV≈ -1 in 1843 (which made it the second brightest star in the sky, after Sirius) in an event that we call the Great Eruption. After this, η has appeared to be more stable than in the centuries preceding the Great Eruption (in a reminiscent way of the irregular activity preceding a big eruption in a volcano or a geyser), settling in the 7th or 8th magnitudes, with the exception of the Lesser Eruption between 1887 and 1895. Since 1940, Eta has gradually brightened, and is currently a star of the fourth magnitude (4.47 in February 2011, Fernández 2011).
Given Eta’s distance, of the order of kiloparsecs, the Great Eruption was an extremely luminous event. In a few years, Eta produced almost as much visible light as a supernova explosion, but it survived. These types of stellar explosions that do not destroy their progenitor stars are referred to as supernova impostors. The Great Eruption is the greatest well-documented non-terminal stellar explosion. It ejected the expanding bipolar lobular cloud, which amounts to around 1 M⊙.
What is extremely peculiar about this event is that it is not understood at a surprisingly basic level (Davidson, 2002). In fact, we would not know about it had it not been so accessible for observation. According to Davidson, “theorists have repeatedly failed to predict crucial effects for very massive stars in general, and for η Car in particular” (Davidson, 2002), which can suggest that something critical is missing from the models that seek to explain these phenomena.
Eta is surrounded by a cloud of dust called the Humunculus Nebula. This cloud is an emission Nebula that was ejected during Eta’s outburst, whose light reached Earth in 1841. Eta Carinae is usually classified as a Luminous Blue Variable (LBV) with a companion star. LBV are extremely massive stars, so large that they approach the theoretical upper limit for stellar mass, known as the Eddington Limit. The Eddington Limit is the maximum luminosity where the radiation force outward balances the gravitational force inward, in hydrostatic equilibrium. Because these stars are so large, gravity is almost insufficient to balance the radiation pressure from within the star, causing a stellar wind that constantly ejects matter, decreasing the mass of the star. This mass forms a nebula around the star, as a cloud of dust, hydrogen, helium and other ionized gases.
Eta’s ejecta is characterised by two polar lobes and an equatorial disk. Polarimetry and spectroscopy suggest that at visual wavelengths the Humunculus is a reflection nebula. Most of the light that we see from the Humunculus comes from the star. The apparent magnitude of the system is mV≈5.7, where the light coming from the central system is mV≈8.4. The circumstellar dust extinguishes part of the star’s light; given Eta’s size and distance, we would expect the central star to be roughly mV≈4 if the Nebula was not there.
Most of the radial velocity data that we have from the nebula is consistent with a model in which the lobes are nearly spherical, hollow, and tangent to each other at the star’s location. The edges of the polar lobes were found to have extrapolated origins in the time of the Great Eruption. (Davidson and Humphreys, 1997).
The motion and radial velocities of some condensations in the equatorial disk suggest that the disk was ejected in 1890 (Smith and Gehrz, 1998), coinciding with what is known as Eta’s Lesser Eruption, a second eruption of a smaller scale and less dramatic mass loss than the Great Eruption of the mid-nineteenth century. The disk has a radius of the order of 1011m, and is surrounded by an outflow of a much larger disk of a scale ~1015m, with a morphology of radial “fans” (Davidson and Humphreys). This information, along with other electromagnetic evidence, was what prompted astronomers to believe that Eta might not be a single star, but a binary system.
Eta Carinae is around 8,000 lightyears, or 2500 parsecs away from us. Estimating the numerical distance between astrophysical objects that are far is not a trivial task. For objects that are in our immediate neighbourhood, the Solar System, we can very accurately know distances through sending radio waves to a particular object and measuring the time the waves take be reflected and come back. This method is not useful for objects that are further away, not only because it is not realistic to wait for the signal to come back to Earth, but also because it would be too faint for us to detect.
Another method to detect astronomical distances is that of stellar parallax. It consists of detecting the apparent motion of nearby stars as the Earth orbits around the Sun. However, Eta is also too far for it to noticeably change its position as the Earth moves around the Sun.
There are several methods to determine the distances of stars in the Milky Way, such as Eta Carinae. One way is looking at standard candles that are near the star that we are interested in. Standard candles are objects of known luminosity (such as stars with characteristics that determine a particular spectral type, like Cepheid variables). There is a relationship between apparent magnitude (how bright the object appears to us as we observe it), absolute magnitude (how bright the object intrinsically is), and its distance to us.
Another way in which the distance to Eta has been determined is by looking at the angular size of its surrounding halo (the Nebula), combined with spectroscopically observed velocities of expansion and the fact that we know the time elapsed since the Nebula was ejected (the time of maximum light during the Great Eruption). In addition to these two methods, it is possible to correlate absolute magnitudes of ordinary novae with properties such as the rate of decline of light from the maximum since the Great Eruption. These methods are discussed in Thackeray, 1956, and all give an approximate distance of 2.5 kpc.