Black Hole binaries as gravitational wave sources

The battle between GPU(Graphical Processing Unit) and CPU(Central processing unit) is even more fierce in the realm of science where many disciplines use simulations. One of this is astrophysics where simulations of physical processces in the Universe play an important role.

The main advantage of using the GPU is the price of hardware, you can have your own supercomputer at a small fraction of the CPU one and the disadvantage is that there are few programs that are using them and you would have to rewrite your code for using the power of the GPU.

In a recent article, S. Banerjee, H. Baumgardt and P. Kroupa (2009) from Argelander-Institut für Astronomie in Bonn used N-body simulations of star clusters using the NBODY6 code on GPU platforms trying to find if black holes binaries can be sources of gravitational waves.

So why this is important? Well, gravitational waves, the elusive space-time fluctuations predicted by Einstein’s theory of general relativity, if detected would open a new window of observation of the Universe in ways totally new. It is like we will receive a new pairs of eyes, ears, legs and arms to feel the reality around us.

In trying to detect them in present or future experiments like LIGO or LISA, one important thing is knowing where to look. And this study points to stellar mass black hole mergers that can happen inside star clusters within the first ∼ 4 Gyr of cluster evolution as one of the most important physical processes capable of generating observable gravitational waves.

In short the process follows like this, small black holes fall rapidly in the cluster core where they evolve bound by gravity, forming a dense sub-cluster. Here by close encounters pairs of black holes form, a process that can eject some of single black holes outside the system by 3-body interaction, further ejection of the pairs leave only a few inside the core where they can merge releasing large amounts of gravitational waves.

Considering that a galaxy like ours can have hundreds star clusters or that a neighbor galaxy like M87 even thousands, knowing that one of them can have one or two BH-BH mergers makes the job of future detection of GW much easier.