What is Dark Matter?

CERN's celebratory picture for dark matter day

Dark Matter is just matter that doesn't interact via the Electromagnetic (EM) or Strong Nuclear forces. No EM interactions means that it can't give off light, or absorb light, or reflect, refract, or scatter light in any way. This, naturally, makes it rather difficult to see (thus "dark" matter, although I suppose it's more "transparent" than "dark"). Our current best measurements indicate that something like 85% of the matter in our observable Universe (about a quarter of the total mass-energy content) is Dark Matter.

Isn't Dark Matter weird/spooky?

Not at all. Neutrinos, for example, satisfy the definition of Dark Matter, they just represent such a tiny fraction of the total DM in the Universe that people tend to neglect them when they ask, "what is Dark Matter made of?"

There is nothing at all strange or unusual about certain particles not interacting in certain ways. Neutrons have no electric charge (although they do have EM properties, but that's neither here nor there), and electrons don't interact via the Strong Force, so why shouldn't there be particles that interact with neither, like the neutrinos? Saying that interacting with light is "normal" is purely human bias, because we rely so much on sight. Having lots of DM in the Universe is in no way "weird"; it just means that the Universe doesn't revolve around what humanity finds convenient!

Why are we confident that DM exists?

This is by no means a complete list, but it should give a sense of the kinds of evidence we have. Each of these would take at least a chapter of a book to explain properly, but hopefully this will give the general idea.

• Galactic rotation curves.
  
  When one object orbits another, the orbiting object has to be constantly accelerating towards the central object (or, more precisely, they both accelerate towards their combined center of mass). Without that acceleration, the orbiting body would just fly off. 
  
  The faster the orbiting body is moving, the more acceleration is needed to keep it in its orbit. Since in this case the acceleration is due to gravity, this means that the central mass has to be bigger. For a circular orbit of a small object m at distance r and velocity v around a large (and assumed stationary) object M, the acceleration requirement gives
  
  v2r=GMr2
  which in turn gives us the relation
  
  v=GMr−−−√(I'm doing this with Newtonian gravity for simplicity; to do it with full rigor would require General Relativity. In this situation, the Newtonian approximation is actually generally pretty good.)
  
  For a more complicated object than just two point particles, as long as there's enough symmetry, the gravitational version of Gauss's lawsays that the relevant M is the total mass of everything in the galaxy that's at a distance less than r  from the center. 
  
  This allows us to "weigh" different parts of the galaxy, by measuring the relationship between r and v. (We can measure the rotational velocities by comparing redshifts on the approaching and receding sides of the galaxy.) 
  
  This image from Wiki shows the result of this measurement:

The "expected from visible disk" line is determined by adding up the masses of all the parts of the galaxy that we can see. (How we measure that mass is a whole different discussion.)

• Gravitational Lensing
  
  In General Relativity, whenever light passes through a gravitational field, that field bends its path slightly. This acts like a Gravitational lens, and can produce, for example, "Einstein Rings", like this image from Wiki:

The "ring" is a distorted image of a single blue galaxy located behind the red galaxy at the center. Light from the blue galaxy goes out in all directions, but is bent by the red galaxy's gravity. This means that the light that starting out on a "direct path" to us never reaches us, but light that was originally missing us by a specific amount (in any direction) gets bent back towards us, which makes it look like it's coming from a bunch of different directions, resulting in the ring image seen here.

This is a highly dramatic example of gravitational lensing, but there are much more subtle effects that can still be useful. In Weak gravitational lensing, statistical analysis of distortions in the light we receive allows us to "map out" the gravitational field between us and distant galaxies. Often, this just shows more mass than we know how to account for, but that could be explained away by just assuming that our understanding of gravity is off.

There's something else, however, that's a lot harder to explain away in that manner: the Bullet Cluster.

Image from A Matter of Fact on nasa.gov

                       

What's going on here? Well, two galaxy clusters collided with each other, and this is the aftermath. The red coloring represents where the visible matter is, while the blue coloring represents where the dark matter is, as inferred by gravitational lensing. Why are they separated so much? Well, most of the luminous matter in a galaxy cluster is in the Intracluster medium, a hot, dense, plasma. When these plasmas collide with each other, a significant amount of the matter slows down. However, since Dark Matter interacts only very weakly, the DM components of the two clusters were free to pass through each other unimpeded, resulting in a separation (as seen here). Not only is there "not enough" luminous matter... it's in the wrong place! A small number of scientists remain committed to finding ways to explain this without DM, and they have had partial success, but only by including as-of-yet-unmeasured things that are far stranger than DM (for example, a rank-3 tensor field, which, while possible, would be the first tensor field of such a high rank ever found).

• Effect on the Cosmic Microwave Background.
  
  For the first few hundred thousand years after the Big Bang, The Universe was hot enough that it was highly ionized, which made it more or less opaque to light; photons were pinballing around just like any other particle. However, once things cooled down enough, significant amounts of the protons and electrons combined into neutral Hydrogen, which is (more or less) transparent to most of the light that was around at the time. This happened fairly quickly (in terms of cosmological time), and so it was as if all of the light pinballing around all over the Universe were suddenly released all at once, effectively capturing a snapshot of the Universe at that moment in its evolution.
  
  Since this light was released everywhere in the Universe, we can point our radio telescopes in any direction we like, and there it is: the Cosmic microwave background (CMB). It's almost the same temperature in every direction, but there are small differences (generally around one part in 105${10}^5$), and we have measured these tiny variations with extraordinary accuracy: first via the COBE satellite, which was then replaced by the more advanced WMAP, which was then replaced by the more advanced Planck (spacecraft), which is currently in operation.
  
  These tiny variations can tell us a lot about the early universe. For example, statistical analyses of these variations show the distinct signature of pressure waves propagating through that early plasma, and the nature of these Baryon acoustic oscillations can tell us a lot about what kinds of things were around. Specifically, the protons and electrons would be dragged along by the (dominant) photons, becoming part of the wave, but Dark Matter wouldn't, and would only be indirectly affected by the resulting small changes in gravity. The presence and abundance of Dark Matter therefore affects how these waves impact the temperature variations in the CMB.
• The formation of large-scale structure.
  
  The standard story given in popular science explanations goes like this: the Universe started out hot and dense and more or less uniform, then it expanded and cooled and clumped into stars and galaxies. However, this story is incomplete, in a way that means that galaxies wouldn't exist without dark matter.
  
  At a surface level, the story makes sense; heck, I got almost half-way through a Ph.D. in Physics without noticing any problem with it! It sounds so plausible because of how gravity works: if matter is distributed more or less evenly, but some places are a tiny bit more dense than others, gravity will tend to make those overdensities bigger and bigger. Why? Well, even if a region is just a little denser than its neighbors, it's still going to win the gravitational tug-of-war and gradually accumulate more and more mass. Of course, once it has more mass, it wins the tug-of-war by even more, and so it's a run-away process that ends in big gravitationally-bound clumps.
  
  So, what's the problem? Well, consider the air in the room with you right now. Are there tiny density variations? Of course, since perfect uniformity is impossible. But, is it forming into exponentially denser clumps? Certainly not! The reason for this is that, under these kinds of conditions, when the density of a gas goes up, so does its pressure. That pressure makes the over-dense region expand outwards again, returning the density back to average!
  
  Now, of course, the scales and temperatures involved are totally different. A huge region of gas will have more gravitational attraction than a tiny pocket of denser air in your room, and gas in space has no need to be at room temperature, either. So, if the gas can cool down enough, that can reduce the pressure enough for gravity to win. But, the more it compresses, the more it heats up, because it's converting gravitational energy into thermal energy, and so the pressure goes up again. This means that forming a galaxy is a very gradual process, during which it has to constantly be getting rid of tremendous amounts of energy. If it were just a cloud of gas, without any outside interference, this process wouldn't be nearly fast enough, and we wouldn't have galaxies today.
  
  But, as you know, hot things give off heat much faster than cold things. So, if we want galaxies to form by the present day (or, indeed, before the expansion of space makes matter too dilute to form galaxies at all), something has to be forcing the gas to compress and become denser and hotter than it would be able to under its own gravity. Enter: Dark Matter. Because it interacts only weakly, Dark Matter doesn't have pressure like gas does. So, the argument about the run-away gravitational process actually works for Dark Matter. DM can't get rid of energy very easily, and so conservation of energy and angular momentum mean that it can only collapse to about 200 times the background density, but the resulting Dark matter halo provides enough of a gravity well to "seed" the formation of visible galaxies. So, it's not a coincidence that the galactic rotational curves showed large amounts of dark matter... the galaxies wouldn't have formed there without it!
  
  As a result of this DM "seeding" process, theoretical models and computer simulations of the formation of DM structures have been fairly successful in describing the statistical properties of how galaxies are distributed now, as well as how they were distributed earlier in the history of the Universe (which we can measure by looking at very distant galaxies).

Okay, so that's why we think Dark Matter exists. But, the next obvious question is... well... what is it? What's it actually made of? What are its properties? Here, we have only very partial knowledge, and multiple different theories, any combination of which could be correct (or, perhaps, none of them). This leads us to:

What do we know about the properties of Dark Matter?

Again, this is in no way exhaustive, but it should give you a decent idea.

1. It's "cold".
   
   This is why the current dominant model of cosmology is called the Lambda-CDM model: "Lambda" (Λ   ) stands for the Cosmological constant, and "CDM" stands for "Cold Dark Matter".
   
   When an astrophysicist describes something as "cold", they generally mean that the associated thermal velocity is much less than the speed of light. By this standard, the air in Death Valley is "cold". But, then again, to cosmologists, galaxies are basically point-particles, so everything's a matter of scale and perspective!
   
   So, why does DM need to be "cold"? Well, remember that DM clumping together was integral to the formation of structures like galaxies. However, if DM were very hot (and therefore the particles were moving very fast), this would prevent it from clumping properly. I have explained it in very vague terms, but the effects are actually well-understood mathematically. In fact, it's similar to something that doeshappen, with photons: Diffusion damping (or "Silk damping"). However, in the case of Dark Matter, the result would be a significant delay (or even outright prevention) of the formation of galaxies, to the extent that "warm" Dark Matter can be observationally ruled out.
   
   Incidentally, this is how we know that it's not all just neutrinos: given what we know about the early Universe, they would be far too hot!
2. It interacts only very weakly.
   
   This is somewhat part of the definition of Dark Matter, but it's good to see observational confirmation. I know fewer of the details on this, but I do know that there are observational bounds in the "interaction cross-section" of Dark Matter, both in terms of its interactions with luminous matter and for its theorized self-annihilation processes (in which two DM particles could interact and annihilate each other). Also, as discussed before, the Bullet Cluster shows giant dark matter halos more or less just passing through each other, which suggests very weak interactions.

So... given those properties,

What might Dark Matter be made of?

There are two leading theories (as far as I'm aware) that suggest the existence of specific types of new particles. Both are well-motivated theoretically (as in, we have good reasons for suspecting that particles with those particular properties might exist), but neither has been experimentally confirmed yet. Interestingly, the two predicted particles are totally different from one another, not slight variations on the same theme.

At the end of the day, either of these theories could be right, or both (given the existence of neutrinos, there's no need for all the rest of the DM to be a single type of particle), or, of course, neither.

So, what are the theories?

1. Axions.
   
   I feel morally obliged to put this one first, even though the other one is more popular at the moment, because my university is heavily involved in the Axion Dark Matter Experiment (ADMX), and so of course I'm rooting for my colleagues!
   
   The existence of Axions has been theorized since the 1970s, but we only recently have the technology to even properly start to try to measure them in the lab. Axions are extremely tiny particles (unlike WIMPs, the other leading option), and so they would have to exist in truly huge quantities. Still, because they interact so weakly, it's hard to detect them, even with billions of them passing through your detector in a tiny fraction of a second.
   
   The linked wiki articles do a better job of explaining the theoretical motivation and experimental search than I could. It's really quite elegant, and would solve a lot of outstanding mysteries in particle physics (like the Strong CP Problem), but I can't do it justice.
2. Weakly Interacting Massive Particles (WIMPs)
   
   (Wiki reference: Weakly interacting massive particles)
   
   To explain why WIMPs are theoretically attractive, I have to take a little detour into "relic abundance", i.e., how many particles of a given type were left over after the Universe cooled down and generally became a more stable place.
   
   In the early universe, everything was very dense, and many different kinds of particles were "tightly coupled" (i.e., they interacted with each other frequently). However, as the universe got larger and cooler, these interaction rates slowed down, more or less to a stop, a phenomenon known as "freezing out". The time at which something "freezes out" depends on a number of things, including its mass and how strongly it interacts with other things.
   
   This "freezing out" has a huge effect on the abundances of various particles in the Universe. For example, consider the process
   
   n⇔p+e−+νe¯¯¯¯¯,
   
   in which a neutron can turn into a proton, electron, and electron anti-neutrino (or vice versa). If you go back far enough, this reaction will be in thermodynamic equilibrium, just like with any chemical reaction that can go either direction. However, once the neutrinos "freeze out", equilibrium can't be maintained anymore (although other processes, like beta decay, can still happen). The relative abundance of protons and neutrons at the moment of freeze-out is determined by two factors: the mass difference $$Δm between the two particles, and the temperature of the Universe when freeze-out occurred. These factors combine to determine by how much the lighter particle is thermodynamically favored, with an exponential dependence:
   
   NnNp∝e−Δmc2/kT
   
   where k is Boltzmann's constant.
   
   This clearly impacts the "relic abundance" of the particles involved (here, protons and neutrons). So, in general, the abundance of a given particle in the Universe today is significantly influenced by the mass of that particle and how strongly it interacts (since that affects freeze-out time and thus freeze-out temperature).
   
   This brings us to the so-called "WIMP miracle": a particle that interacts predominantly via the Weak Nuclear Force, and that has a mass near the mass scale associated with Weak interactions (∼100 GeV/c2), would have a relic abundance that would match the measured abundance of Dark Matter in the Universe. Given that such a particle was already speculated to exist (in the context of Supersymmetry), this was very theoretically attractive to a lot of people, although our inability to find it so far has damped some of their spirits.

source - Erik Anson's answer to what is Dark Matter

This post will be followed by a post on Dark Energy and how these two are interlinked with each other. So stay tuned and dont forget to subscribe my blog and visit it regularly.

Ciao!