Seeking New Particles with Long Lifetimes

For context, you may find it useful to read previous articles on supersymmetry first (this one, and this one, and this one) before reading this article, which assumes you have the knowledge they contain.

Let’s recall the classic three assumptions of the most popular form of supersymmetry:

1. in any process, the number of superpartners can only change by an even number;
2. the lightest superpartner [which is stable, by assumption 1]  is a superpartner of a particle we know (and therefore, to avoid conflict with other data, an undetectable neutralino or sneutrino);
3. the superpartners that are affected by the strong nuclear force are significantly heavier than the other superpartners of known particles.

If we violate either assumption 1 or assumption 2, then the lightest superpartner of any known particle — let’s call this particle the “S” for the purposes of this article — will decay.  It may violate assumption 1 by decaying into a set of particles that do not include a superpartner.  [This is called “R-parity violation.”]  Or it may violate assumption 2 by decaying to a light superpartner of some unknown particle, along with at least one additional particle. For instance, I have already covered the case where a neutralino decays to a gravitino (superpartner of the graviton) plus a photon, a Z particle, or a Higgs particle. [Since the graviton is a particle that has not been observed — though most of us are confident it exists — I don’t consider it “known” for current purposes.]

Because this S particle  is unstable, the reasons given in this article that require that it not be affected by electric and strong nuclear forces do not hold.  [A heavy stable particle with electric or strong nuclear forces would affect the abundance of various elements formed in the Big Bang, making them different from what we observe; but as long as the particle decays fast enough, this effect would not occur.]  Thus the S could be any of the superpartners: a neutralino, a chargino, a slepton, a squark, or a sneutrino. And that’s important, because if the S lives long enough to pass through part of the detector material before it decays, its signature in the detector can be extremely special and distinctive, allowing it to be discovered with relative ease.

As long as someone looks.

Which isn’t always as easy as it sounds.

The decay of the S is certain, but every type of decaying particle has an average lifetime, and nothing I’ve said tells us the lifetime of the S.  So if we want to go looking experimentally for the S, we have to consider all of the different possibilities.

• It might be so short-lived that it decays within a microscopic distance of the place where it was created, in which case the detectors can only observe the particles into which it decays. For reference, let’s call such an S “short-lived”.
• Alternatively, it might be so long-lived that it can pass straight through an LHC detector and ends up decaying halfway across Switzerland, or even on the moon. Let’s call the S “metastable” in this case — as far as the LHC detectors are concerned, it might as well be a stable particle.
• And finally, there’s the middle ground — it might decay while crossing through one of the LHC detectors. Let’s call this type of S “long-lived”; it lives rather a long while, but the LHC detectors will be able to observe it breaking apart.

Let’s start with the case of metastable particles — when the S travels all the way through an LHC detector, and decays long after it exits.

If the S is a neutralino or a sneutrino, it will pass through a detector leaving no trace, just like a neutrino.  For this reason, there’s no experimental difference between a metastable neutralino or sneutrino and a stable one — the same techniques have to be used to find them.  And the stable case is that of the most popular form of supersymmetry: the only sign of the S particle is an imbalance of momentum, or “missing energy”. Standard searches for supersymmetry constrain this case just as they do the most popular form of supersymmetry.

If the S is a chargino or slepton, then it appears in the detector as a particle very similar to a muon — it leaves a track in the tracker and travels straight through the detector to the muon detector — but on closer examination may exhibit odd features. For instance, whereas a muon with high-momentum should be moving at almost the speed of light, our heavy S could be moving well below the speed of light. There are several clever methods to try to measure this slower velocity. The first is obvious: timing.  A slow particle will simply take a few extra nanoseconds to cross the detector.  Any muon-like particle traveling well below the speed of light is clearly something exotic and new.   The other less obvious methods are based on how electrically charged particles interact with atoms as they pass through matter.  For instance, a particle traveling slower than the speed of light by 10% or more is much more efficient at stripping electrons off their atoms (“ionizing” the atoms) and this “high-ionization” can be measured in the tracker systems near the cores of the ATLAS and CMS detectors. A muon-like particle that is highly-ionizing is not a muon, or anything else known: it must be a new metastable charged particle.  Such a particle would be easily seen, if it were present.

If the S is a squark or gluino, the situation is similar but a bit more complicated. Since squarks and gluinos feel the strong nuclear interaction, just like quarks and gluons, they form hadrons (bound states of quarks, gluons, and in this case also a squark or a gluino). These exotic heavy hadrons, called (for historical reasons) “R-hadrons”, may or may not be electrically charged, and the expectation is that both types exist (for the same reason that there are electrically charged protons and neutral neutrons in nature.) If they are charged, they can be detected just the same way as a chargino and slepton. There are some subtleties in the measurement, which I won’t go into here, but suffice it to say that even with these subtleties accounted for, it is rather easy to detect these particles [with a small-print loophole if there all charged  R-hadrons decay instantly to neutral ones.]

Both ATLAS and most recently CMS have looked for highly-ionizing muon-like particles, and have supplemented this strategy with a couple of others to deal with the subtleties of R-hadrons. The results are spectacular, and devastating. The most recent CMS result requires that the mass-energy (E = m c-squared) of the S particle be larger than

• 800-900 GeV if it is like a gluino (the precise limit depending on some assumptions about how R-hadrons behave.)
• 500-600 GeV if it is like a squark (again precise limit depending on assumptions)
• 300 GeV if it is like a slepton

The second and third of these are very conservative limits; there are many contexts in which the limit would be raised. [For example, slepton S particles might be produced most often in the decays of heavier unstable gluinos; in this case both this measurement, and others, would put even stronger limits on the S mass.]

Nothing in these measurements makes any reference to the origin of the S particle — there’s no assumption about supersymmetry anywhere in these results.  So in complete generality, any metastable S particles affected by the strong nuclear  force are now quite constrained by the data.  Constraints on particles that are electrically charged but are not affected by the strong nuclear force are significant though somewhat weaker.

Searching for Particles that Stop Before They Decay

There’s one other way to detect metastable particles that is rather amazing. It works best for gluinos and squarks, which are easily produced in the collisions of quarks and gluons, and less well for charginos and sleptons, which are relatively rare. Like any heavy particle produced at a hadron collider, S particles are produced with a wide variety of velocities. Some of them will travel at half the speed of light, or even slower. These cannot be detected easily using the methods described in the previous section… because they take so long to cross the entire detector that a new collision has started before they get there! The LHC detectors were not designed to find these kinds of particles, and unfortunately the details of their hardware actually gets in the way.  Fortunately, such experimental design problems can often compensated by later creativity. It turns out that an S particle which travels slowly enough can actually bang into enough atoms to come to a stop! And it turns out that ATLAS and CMS are thick enough that sometimes this will happen. So imagine: right now, some of the S particles might just be sitting inside the detectors, waiting for their moment to decay.

Both ATLAS and CMS have looked for particles that decay, somewhere in the middle of the detector, at a random time when no collision is occurring. As you might guess, to do this requires specialized techniques — because the detector hardware and software is mainly designed to work when a collision is occurring, not when it isn’t. But the techniques work very well, and the limits on gluinos or squarks (or other new particles affected by the strong nuclear force) that stop in the detector are somewhere (from the most recent CMS paper on the subject) around 600 GeV, as long as the average lifetime of the new particle is somewhere between 10 millionths of a second and 1000 seconds.

Note the limits on the mass of the S from the R-hadron search described in the previous section are actually stronger.  But the assumptions in the two types of searches are somewhat independent, so it may be hoped that anything that could sneak through one of the two searches would eventually be picked up by the other one.

This page will be supplemented soon with a discussion of the fascinating case of “long-lived” S particles that decay while still flying across the detector.