Simplified: Freeze-in of Dark Matter in Five Steps

With my newly acquired free-time, I wondered whether it would be possible to elaborate on my actual research in a manner accessible to general audiences. This is the first attempt at that. I take a shot at explaining the main ideas behind the paper which resulted from my Master’s thesis. And I do this without using any equations. You can read the paper here: arxiv: 1809.04849. We’ve been told it’s pretty good. :p

Questions welcome!


If you’re a regular reader of this blog or a general science enthusiast, you’ve probably heard that 25% of our Universe is composed of Dark Matter. In fact, you might have seen a plot that looks something like this:

Screenshot 2018-09-18 11.15.04
Contents of the Universe

In the simplest of terms, dark matter is something which we cannot see directly (as one would see planets or stars) because it doesn’t interact with light or normal matter (cups, cars, 18th century French Guillotine etc.). Despite not having seen dark matter directly, we are reasonably sure that it exists and there’s a whole bunch of it out there. For a quick refresher on why, you can read A Beginner’s Guide to Dark Matter over at the TTK blog.

Once we establish the existence of dark matter, the next question to answer is where did it come from? More specifically, how did we end up in a Universe where there is five times as much dark matter as normal matter?

Conventionally, the explanation goes like this: In the very early Universe, there was a hot particle soup which included, along with “normal” particles like electrons and photons (aka SM particles), “dark” particles. (We don’t have to specify the nature of these particles yet). These dark particles collided and annihilated and otherwise transformed into SM particles and their number decreased.


Screenshot 2018-09-18 14.18.50.png
Dark particles annihilate into standard model particles.


But as the Universe expanded and distances between the particles grew, these collisions became less and less likely, and eventually stopped. Thus, starting from some large dark matter abundance, we ended up with the 25% dark matter abundance we see today. This mechanism, known as thermal freeze-out, has been at the forefront of dark matter research for the past few decades. However, this is not the only way to produce dark matter.

A recent production mechanism that has gained traction is freeze-in. (We can be a tad bit uncreative with names). What freeze-in says is that the early hot particle soup did not include dark particles. Instead, all of dark matter was produced by collisions, annihilations or decays of SM particles.


Screenshot 2018-09-18 14.18.40.png
Standard model particles annihilate into dark particles.


Essentially, visible or normal matter produced dark matter until once again the expansion of the universe put a stop to such processes giving us the abundance of dark matter we observe today.

Now I know what you’re thinking. Who cares? Why should I care about freeze-in? And the answer has to do with the nature (properties) of particles that we think constitute dark matter. In the standard freeze-out scenario the particle candidates are called WIMPs (Weakly Interacting Massive Particles). And although the WIMP interaction with visible matter is tiny, it is not undetectable. So, for the past couple of decades we’ve been working like crazy to detect WIMPs. We’ve set up elaborate direct detection experiments, we’ve tried to produce them at colliders like the LHC and we’ve also considered the possibility that WIMPs might decay into SM particles which we can then “see”. But we haven’t found anything yet and so the WIMP model is becoming problematic*. In science-speak:

Paper.Sketches.24 2

The bottom line is, it has become kind of important to think about alternative dark matter production mechanisms and freeze-in fits the bill very well. In the paper, we consider a specific dark matter model in which the production is via freeze-in.

Next, let’s dive into the particle physics part of the paper. We’ve been talking about producing dark particles from SM particles. But when coming up with a model, one needs to be slightly more specific than that. The Standard Model of physics has 17 particles (shown below).


As you can see, it has no dark matter particle so studying dark matter invariably leads to adding particles to this model. In our paper, we extend this model by adding a dark matter particle represented by ss is like the Standard Model Higgs boson except it is very light. The Higgs has a mass of 125 GeV. can have a mass anywhere between a few keV to MeV, orders of magnitude smaller than the Higgs mass (1 GeV = 103 MeV). Next, we assume that s only interacts with the Higgs boson so that it can only be produced via decaying Higgs bosons. Now we have all the ingredients needed to study the freeze-in production of dark matter (or s). And using a bit of math and a bit of code, we can compute the number of dark matter particles present at the end of freeze-in.

The story doesn’t end here, however. Sure, after freeze-in has ended, dark matter can no longer be produced by decaying Higgs bosons. But the amount of dark matter can still change. How?


By design, s also has the fun property where it can breed more s-particles or cannibalize to reduce the number of s-particles. So you collide two s particles and end up with three of them. Or, two particles can “feast” on a third so that at the end of a 3-particle interaction, we’re left with only two particles (no party like a cannibal party). The point being, we have to keep track of how much dark matter we have throughout the history of the Universe. And this is basically what we do in the paper.

The paper is titled, Freeze-in production of decaying dark matter in five steps, and each “step” is essentially one way in which the dark matter abundance changes. At the end of it all we have a robust dark matter model, the predictions of which we can compare to the experimental results. We can use the absence of a clear dark matter signal to limit parameters of the model. Conversely, we can use the parameters of the model to predict potential signals which could be seen at future experiments.

And now you know what I’ve been working on for the past year.


Like I said, I’m not completely sure if this makes sense. But it was fun to write!







5 thoughts on “Simplified: Freeze-in of Dark Matter in Five Steps

  1. Hey, you did awesome at dissecting the topic into a fairly understandable model of thought. I was able to follow all of it. Tell me, do you believe freeze-in is the real deal or you feel a better approach will come about soon? Kind of stupid, considering all ideas grt revamped later on pretty much, but I hope you get the point


    1. Thanks! Of course, I can’t say whether or not freeze-in is the real deal. As physicists, our main task is to see how well the model explains the available data. In case of dark matter, we haven’t been able to detect a clear signal so it is significantly harder to make a blanket statement such as “freeze-in is correct (or incorrect)”. What we can say is that freeze-in models easily evade the strong constraints set by direct detection experiments and as these constraints get stronger with better experiments, we have to consider the possibility that the conventional assumption of WIMPs and thermal freeze-out requires some serious re-evaluation.

      Liked by 1 person

      1. Hm, it reminds me of the Quantum theory vs the Classical EM theory. And I guess I didn’t really have a question, haha. I did enjoy the way you talked about it, though. It fascinated me. And dark matter has always been something I wanted to know more about. Thank you for the informative post!


      2. I wouldn’t draw an analogy between classical/quantum EM. Classical EM is just the large-scale manifestation of Quantum electrodynamics, whereas here we really have many competing models to explain one phenomenon. Glad to know you found this helpful!

        Liked by 1 person

      3. I meant that for photoelectric effect. Studied that during school, the whole intensity doesn’t effect the energy but rather just the number and hence the photoelectric current only. Whereas for a large time, the former was believed true. Anyway, sorry for the ramble.

        Liked by 1 person

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s