Detecting dark matter at the LHC

One way physicists can better understand particles, and look for new species like the dark matter that Frank Close talked about, is to produce them in a lab. And one way to do that is by smashing together existing particles - like protons - at extremely high speeds. This can break them apart to reveal their constituents, and the behaviours of those constituents. Scientists doing this work famously say “you break it, make it, or shake it!” and this is what’s done at the CERN Large Hadron Collider - LHC, buried under Geneva. With the energies this 27 km long machine can harness, it can even recreate the conditions of the very early Universe, giving us insights into how key particles might have behaved at that time. Chris Smith spoke with Sarah Williams from the University of Cambridge, who works on one of the experiments at the LHC about their searches for dark matter...

Chris - Frank Close was just saying, Sarah, that we're now on the plains and that we need to strike out in search of that next mountain range, which is dark matter. You're part of the expeditionary force. Which way are you heading?

Sarah - That's actually one of the biggest challenges associated with this mission, in that while we know some of the characteristics we expect from dark matter, based on how it behaves in the Universe, there's actually a wealth of possible candidates. And that means that in order to maximise our chances of discovery, we have to systematically investigate all these possibilities. So I've been using the collisions in the ATLAS experiment at the LHC to search for new particles predicted in an extension of our current recipe book, the Standard Model, which is called Supersymmetry. This predicts that for every particle we currently know about, there's a heavier superpartner and introducing these new particles can actually solve several of the outstanding mysteries in our universe, including providing us a candidate for dark matter.

Chris - We can infer that dark matter is there, Sarah, when we look at distant galaxies and we can see the stars moving around in ways that can only be accounted for if there's something a lot heavier there, than what we can see. And that's how we worked out partly that it must be there. Given that, how are you trying to recreate something that's happening on the scale of galaxies in a particle accelerator?

Sarah - You're correct that what you're seeing in the Universe is the gravitational effects associated with large amounts of dark matter. However, we also expect that dark matter particles should have some weak interaction with particles in the Standard Model, like the protons that we're colliding in the LHC. And the nice thing about these LHC collisions is they happen so frequently - we're talking about colliding bunches of protons 40 million times a second - and that means that even if we'd only have very rare or weak interactions between the Standard Model particles and dark matter, in our very enormous data sets we might be able to see interesting hints that we've produced dark matter particles in these collisions. One of the problems, of course, with this is that actually when producing these very rare processes, it could also be mimicked by other processes that we already know about. So we have to be very careful with the design of our searches and also understand the known Standard Model processes very well.

Chris - So it's not that you can see the dark matter you're making. You are inferring that you're making it by onward domino effects on other particles, the way it interacts with other products of those collisions. And you can say, well, we would see that if dark matter is being made there and it fits our current theory for what it is.

Sarah - As you say, one of the big challenges with our experiments is, unfortunately, that a lot of these heavy particles we're trying to produce, and either discover with dark matter or now measure in the case of the Higgs particle, is that we can't just see them directly. So, most heavy particles would decay almost instantaneously, and then it's the particles they decay into that we're measuring. One of the challenges with dark matter is, as we said, because it would interact so weakly, it would actually not leave any visible signature in our detector if we were to produce it. So what we have to do is look at all the other particles that we produced in the collision and try to infer the possible existence of a dark matter particle being produced. It's a very challenging job, but one that with these very large data sets, we hope to be able to do.

Chris - Have you got anywhere yet? Are you seeing things that tantalisingly make you think you're on the right path across Frank Close's planes to that next mountain range? Or is it very much a wilderness at the moment?

Sarah - Unfortunately, one of the problems when embarking upon a journey where you don't know which way to go is that you have to be prepared to gather information about which paths are going to lead to fruition and which aren't. So, unfortunately, I've been working on the ATLAS experiment now for 10 years, where we've performed many searches for candidates of dark matter, and we haven't seen any signals yet. However, this information allows us to refine our future searches and gives us a better idea of where to look. And, actually, one of the exciting things at the moment is that when we start taking data again next year, we've got a few more years of data taking ahead of us, and we can use the lessons learnt so far to really design our searches to target some of the areas of possible dark matter space that could lead us to a discovery.

Chris - This is a horrible question - to say you have to answer this in 30 seconds, Sarah - but there are rival theories for what dark matter is. Are you considering all these theories or are you favouring just one when you're designing your experiments?

Sarah - Unfortunately, there is absolutely no way we could design a catch-all dark matter experiment. So we rely within the community of having complementary activities going on, some of which might be more sensitive to particular dark matter candidates and others that might be designed to target others. At the LHC, we're very good at targeting certain types of weakly-interacting dark matter particle. However, there are other well-known candidates that would require dedicated experiments, and also within high energy physics, there are others working on those, and I guess you could make up a whole hour or so discussing the different dark matter experiments if you wanted to.