Eta Carinae: Nature’s own Large Hadron Collider
To say that Eta Carinae is one of the most remarkable and marvellous stars in the sky is probably an understatement of hyperbolic proportions. It is one of the most fascinating objects in the universe. Not only is it one of the most massive stars in the Universe (weighing approximately 100 solar masses), it is also amongst the most luminous known (with a luminosity some five million times that of Sol), and is extremely unstable and dynamic, undergoing periodic nova-events (including the famous outburst of 1843 where it became briefly the second-brightest star in the sky after Sirius).
For a comparatively long time now we’ve known that Eta Carinae sheds vast quantities of material via its extremely strong stellar winds -the 1843 event is calculated to have involved the loss of 10 solar masses of material (Smith et al. 2003, ADS), and its ongoing mass loss is perhaps as much as 10-3 solar masses annually (van Boekel et al. 2003, ADS/arXiv) – by contrast, Sol (thankfully!) loses only ~1/1014 of its mass annually via its own stellar winds. The mass loss from the 1843 event created the nebula around Eta Carinae now known as the Homunculus:
As this material streams away from Eta Carine it (de)accelerates, and hence produces synchrotron radiation (De Becker 2007, ADS/arXiv), which has been detected at radio wavelengths. But as we’ve already seen, in other exotic systems such as Pulsar Wind Binaries, even higher-energy emission is possible where particles in the stellar wind collide with ambient particles in the local Interstellar Medium producing energies up to gamma-ray range (> ~100 GeV), and such emissions have indeed been detected recently from Eta Carinae (Leyder et al. 2010, ADS/arXiv), which suggests that Eta Carinae could also be binary system. The binarity of Eta Carinae is very important, as we will see later in this post.
But this marvellous star is the site of something even more stranger & energetic, according to a new report (A&A 526, A57 (2011), ADS/arXiv) to be published in the February issue of Astronomy & Astrophysics and authored by a small group of Swiss & Belgian Astronomers led by Christian Farnier of the Data Centre for Astrophysics at the University of Geneva.
They report, using long-term observations of Eta Carinae with our old friend Fermi/LAT, evidence of two separate mechanisms for the production of the observed gamma-rays.
As well as the ‘standard’ mechanism for production of the gamma-rays by inverse Compton scattering of stellar photons by extremely relativistic electrons, the authors found evidence for an additional and different mechanism for production of gamma-rays; a process called neutral pion decay. Pions are subatomic particles with one quark and one antiquark and are the lightest type of mesons. They come in two forms: charged (π+, π–) and neutral (π0). Neutral pions are typically produced by interactions between accelerated protons (hadrons). They then decay extremely quickly (they have a lifetime typically of only 9 X 10-17 seconds) into gamma-rays, producing a small amount of electrons and positrons, and possibly neutrinos (although unfortunately our current instrumentation is not yet sensitive enough to detect these neutrinos at the Earth’s surface).
So what in the Eta Carinae system could be responsible for the production and acceleration of such protons? We know that the stellar winds of massive stars such as Eta Carinae contain atomic nuclei ranging from helium up to oxygen. These hadrons are then photodisintegrated into neutrons and protons (Bednarek 2005, ADS/arXiv). Farnier et al. suggest that a process called diffusive shock acceleration, or Fermi acceleration could be responsible for the acceleration of these protons to speeds where collisions could produce neutral pions. This process would occur where the stellar wind collided with a previous disturbance (or “shock”).
Is this a realistic scenario? There are a number of relevant factors that the authors evaluate which could impinge upon the probability of gamma-rays being produced in the above way:
- Proton Density: The observed gamma-ray flux requires a certain density of protons as a prerequisite . Unfortunately, the most obvious candidate region for this, the interface between the Homunculus nebula and the stellar wind from the primary star in the system does not meet this threshold. However, the authors calculate that a region in the centre of the system where the stellar winds from both stars collide with each other producing a hydro-dynamical shock would contain enough protons for diffusive shock acceleration to theoretically occur.
- Magnetic Field Strength: A known constraint upon Fermi acceleration is the necessity of a strong enough magnetic field. As with the proton density criteria, the interface region between Homunculus and the primary stellar wind does not appear to contain a field of sufficient strength. However, regions of space closer to the primary star, such as the interface region could be sufficiently magnetised for Fermi acceleration to be a feasible source of proton acceleration.
- Available Energy: To drive the acceleration of protons, there needs to be sufficient energy available that can be transferred to the protons. The authors calculate the amount of mechanical energy in Eta Carinae’s stellar wind and find that it is many times the amount necessary – they suggest provisionally that maybe only ~1% of the total stellar wind energy is transferred in this way.
So this model for the gamma-ray production looks pretty solid. But it rests or founders on the binarity of Eta Carinae. Fortunately, the evidence for this is now reasonably conclusive (see for example Abraham & Falceta-Goncalves (2010), ADS/arXiv).
So the next time you look up into the night sky (although Eta Carinae is only visible from the tropics and the Southern Hemisphere), or watch a documentary on the Large Hadron Collider, remember the universe’s own particle acceleration laboratory, Eta Carinae.
Farnier, C., Walter, R., & Leyder, J. (2010). Eta Carinae: a very large hadron collider
Astronomy and Astrophysics, 526 DOI: 10.1051/0004-6361/201015590
Abraham, Z., & Falceta-Gonsalves, D. (2010). Precession and nutation in the Eta Carinae binary system: evidence from the X-ray light curve Monthly Notices of the Royal Astronomical Society, 401 (1), 687-694 DOI: 10.1111/j.1365-2966.2009.15692.x
Bednarek, W. (2005). GeV gamma-rays and TeV neutrinos from very massive compact binary systems: the case of WR 20a Monthly Notices of the Royal Astronomical Society: Letters, 363 (1) DOI: 10.1111/j.1745-3933.2005.00081.x
Becker, M. (2007). Non-thermal emission processes in massive binaries The Astronomy and Astrophysics Review, 14 (3-4), 171-216 DOI: 10.1007/s00159-007-0005-2
Leyder, J., Walter, R., & Rauw, G. (2010). Hard X-ray identification of Carinae and steadiness close to periastron Astronomy and Astrophysics, 524 DOI: 10.1051/0004-6361/201014316
Smith, N., Gehrz, R., Hinz, P., Hoffmann, W., Hora, J., Mamajek, E., & Meyer, M. (2003). Mass and Kinetic Energy of the Homunculus Nebula around η Carinae The Astronomical Journal, 125 (3), 1458-1466 DOI: 10.1086/346278
van Boekel, R., Kervella, P., Scholler, M., Herbst, T., Brandner, W., de Koter, A., Waters, L., Hillier, D., Paresce, F., Lenzen, R., & Lagrange, A. (2003). Direct measurement of the size and shape of the present-day stellar wind of Eta Carinae Astronomy and Astrophysics, 410 (3) DOI: 10.1051/0004-6361:20031500