What's the recipe for dark matter?
Well, there are many hypothetical models, but the most popular candidate for dark matter is a particle that is:
a) Stable (ie. doesn't decay)
b) Weakly interacting (interacts with other particles through the weak force and gravity)
c) Massive (ie. has a mass)
The acronym used to refer to a weakly interacting massive particle is, not surprisingly, "WIMP."
Dark matter is also often specified as being "cold" (ie. moving at non-relativistic speeds). The reason why "hot dark matter" isn't so favourable by itself is because in the early universe, dark matter composed only of these fast-moving particles would have effectively destroyed (smoothed out) the small density structures that later gave rise to big density clumps (galaxies) as the universe expanded. The only way galaxies could actually form in that scenario is by enormously huge clumps of matter eventually fragmenting to make smaller galaxies - but that doesn't really fit what we see. Cold dark matter in the early universe, on the other hand, is much more consistent with the finer density fluctuations in the early universe (we study this by looking at the cosmic microwave background radiation -
an extremely low-energy "thermal glow" permeating the universe, which is actually a relic from when it was only 378,000 years old).
A WIMP can arise in many theoretical extensions to the standard model (the model of physics as we currently know it). For example, there is a branch of physics called "supersymmetry," which
posits that there is a larger symmetry connecting the force particles (bosons) and the matter particles (fermions). A major result of this is that the standard model particles that we know of have "partner" particles called "superpartners". The superpartners of quarks are known as "squarks," the superpartners of leptons are known as "sleptons," the superpartner of the gluon is known as a "gluino", and so on. In these models, dark matter often assumes the form of a light supersymmetric particle.
Ultimately, any model of dark matter that scientists come up with has to fit the observational evidence - and there's a lot of it. For example, galaxies don't rotate in the way that we expect them to based on the matter that we can see (optical and radio observations). Stars and gas on the edge of galaxies orbit around the centre of the galaxy at high speeds that indicate that the amount of actual mass contained in their orbit is much, much higher than the amount of visible matter. This is the galaxy rotation curve problem.
Also, gravitational lensing refers to the technique of estimating the amount of matter in a cluster by how much light bends around it (light is affected by gravity too!). Based on how much light bends, there's a substantial amount of matter that isn't accounted for by the galaxies (and intergalactic gas) alone.
One really,
really cool image often featured in explanations of dark matter is
an image of the Bullet Cluster (which is actually two whopping clusters of galaxies colliding with each other):
What you're seeing there is a digitally enhanced composite of several images. The pink parts of the image represent the hot extra-galactic gas, mapped from X-ray observations by the Chandra observatory. The blue parts of the image represent the dark matter mapped out by weak gravitational lensing (from various optical telescopes, like the Hubble Space Telescope). What this shows is the gas lagging behind the dark matter, indicating that it is much more affected by the drag force of the on-going collision (the dark matter by contrast appearing to have phased through it).
