How We Learn What Happened in a Proton-Proton Collision

It’s not enough to make spectacular collisions of ultra-energetic protons.  We could pat ourselves on the back for the achievement, but we can’t do any science unless we can figure out what’s actually happening in those collisions.  That’s no mean feat.  Remember the colliding protons are 100,000 times smaller (in radius) than an atom.  My eyes aren’t that good, and yours aren’t either: in fact, it is literally impossible to see anything that small using visible light… so we don’t just take a photograph! 

[Why is it impossible to see something the size of a proton with visible light?  And what is the difference between visible light and other types of light?  Check back for answers at a later time.]

Instead, the strategy for figuring out what’s going on in a collision is to stand back and observe all the junk — and the few pearls — that come flying out of the collision location.  Then we work backwards — given what we see coming out, what must have gone on during the collision?   It’s vaguely similar to a detective, looking at distant video of the expanding debris from an explosion, trying to figure out where and why it took place, what the explosive material was, and whodunnit.

That flying junk from the collision is all in the form of particles … tens or hundreds of them, none bigger (and some smaller) than a proton, and moving at or just below the speed of light.  We can’t see these particles with our eyes (or cameras) either!  So how do we observe them?

We use the fact that fast moving particles traveling through ordinary materials — silicon chips, for example, or lead blocks, or more unfamiliar substances such as liquid argon or lead tungstate — can create electrical signals that we can pick up using metal wires and measure using little electrical devices not so different from (but often more sophisticated than) what we have in our homes and cars and computers.

What we do, in a nutshell, is we surround each collision point with a `detector’, a big machine made from various devices, called `sub-detectors’, organized concentrically in a structure like an onion.  Each sub-detector is made from a particular material, can detect electrical signals, and can send them out to computers that sit far away in another room.  When a particle comes from the collision point in the center of the detector and passes through some of the sub-detectors, the electrical signals it leaves are recorded on those computers.  Detailed analysis of where the electrical signals originate and how strong they are in the different sub-detectors allow us to answer these questions:

1. what type of particle was it?
2. in what direction was it heading?
3. how much energy was it carrying?

In order to observe all the particles coming out of the collision with the requisite precision and reliability, it turns out that the LHC detectors have to be the size of moderately big office buildings!  Different technologies are used in the two big detectors, ATLAS and CMS, and that is why CMS is the size of a six story building and ATLAS reaches nine stories.  But in general they are very similar in their strategies, and comparable in their capabilities.

Using the detectors to figure out which particles were produced, where they went and how much energy they carried, we have the ingredients to make some educated guesses about what may have occurred in a given proton-proton collision.  How we go from these guesses to making discoveries of something new is a story of its own [soon to come].

Here are a few more details, if you are interested:

The location where the electrical signals originate tell us where a particle traveled, and thus the answer to (2) in our list of questions above.  The nature of the electrical signal in the various devices tells us most of the answer to (1), because each type of particle behaves differently in different types of matter.  (For instance, a muon will pass through lead with almost no effect, but a proton will create a large electrical signal.  [More details to follow soon.] )  And then there are two more pieces of information we can use.  The first is the overall strength of the electrical signal left by a particle in certain sub-detectors.  The second is the degree of bending of the particle’s path in the presence of a magnet, which we install in some of the sub-detectors.  One or both of these bits of information can be used to figure out the answer to (3), and part of the answer to (1).  And that’s it!   The rest is just technical details.     [More details to follow soon.]

Both ATLAS and CMS are very good at detecting and distinguishing the following classes of particles — the only ones we know about that exist for long enough to be detected (with one exception, see below):

a) electrons [and their anti-particles, `positrons’ or `anti-electrons’]
b) muons [and their anti-particles, `anti-muons’]
c) photons
d) hadrons [particles made from quarks, antiquarks and gluons, including protons] that carry positive electric charge [and their anti-particles, which carry negative electric charge]
e) hadrons that carry no electric charge [such as neutrons]

and one more thing that’s not so obvious:

f) the presence of particles, such as neutrinos, that can travel undisturbed through matter leaving no trace.

This last step is done by inference, by tallying up the “momentum” of all the particles that are detected.  If it doesn’t add up to zero, as it should, there must have been one or more particles produced that the detector failed to observe.  [Small white lie here, to be explained elsewhere.]

Note the detectors are not designed to distinguish among hadrons with the same electric charge, such as positively-charged-pions and protons. For technical reasons it turns out we generally don’t need to know this, so it didn’t make sense to spend the money to put in equipment that could provide this information.