Formation of dark matter structures

[andrew s 1905]

TL;DR

I can't understand how dark matter can form structures from the initial near uniform distribution.
So my question is what physics allows this to happen?

I can't understand how dark matter can form structures from the initial near uniform distribution as shown in simulation like those of the Eagle project http://icc.dur.ac.uk/Eagle/index.php. I understand how the initial density fluctuation are amplified and dark (and ordinary) matter start to collapse However, unlike ordinary matter which can lose energy via electromagnetic interactions, dark matter will gain kinetic energy from the gravitational potential energy of the initial distribution but this cannot be dissipated and so should in my naive view oscillate about the local center of mass.

This must be wrong but what physics am I missing in understanding this. I have tried various google searches but not found a solution. Any pointers would be most welcome.

Regards Andrew

 

[Janus]

Dark matter can lose energy by gravitational interaction via the emission of gravitational radiation. Given that the gravitational interaction is much weaker than the electromagnetic, This is a much slower process. Also, gravitational interactions between DM particles can transfer momentum from one to another( much like a gravitational slingshot does), which can also aid in forming structures.

 

[Keith_McClary]

How can dark matter collapse without collisions or radiation?

 

[andrew s 1905]

Thanks, for the replies. I think I have worked it out. My assumption about the potential energy was wrong.
Regards Andrew

 

[kimbyd]

One interesting thing about this problem is that on very large scales, structures don't collapse. This derives from using linearized gravity to estimate how structures evolve over time. This fact probably matches more with your original intuition: that conservation of energy would prevent collapse.

So how can there be these huge structures at all if they don't collapse? The answer is pretty simple: at some point, the rate of expansion was slow enough and the systems were dense enough that they settled into stable orbits instead of continuing to expand. Meanwhile, the overdense regions continue to get further apart and the rest of the universe gets less dense. So the density of these large regions is sort of "baked in" by how long ago they stopped expanding, with some settling into stable configurations younger than others.

That said, gravity isn't a linear theory, so this approximation breaks down as you look at smaller structures (such as galaxy clusters and individual galaxies). This leads to more complicated evolution. And dark energy also changes things so that the gravitational potentials of these large regions aren't actually constant.

Finally, note that my description is compatible with the one that Keith posted above. It's just described from a different perspective.