In SUSY models, you can have the case that sparticles and their decay products have near-degenerate masses. For example

$$ m(\tilde \chi^\pm_1) - m(\tilde \chi^0_1) < 1\,\mathrm{GeV}$$

Then in the leptonic decay mode

$$ \tilde \chi^\pm_1 \rightarrow \tilde \chi^0_1 + \bar\nu + e^-$$

you'd have the electrons basically produced at rest, or their momentum will be at least well below what you can comfortably detect at a hadron collider. How would you detect SUSY in such a scenario?

(Assume that you can't just search for other SUSY particles or decay chains, since you are either doing a model-independent search and don't know how heavy gluinos etc. are, or your other particles are out of range, or also degenerate in masses. Nature can be nasty.)

  • You certainly can't just look for a drop in total cross section (disappearing events), since the SUSY cross sections are way too low.
  • I was wondering if you could use associated production to find these kind of decays, for example if you produce a Z along with your SUSY particles, you'd see a boosted Z decay balanced by nothing to the other side. Would that work?
  • I heard these scenarios would be easier to study with the ILC. Is that just because of the cleaner events, and that you'd be able to see those displaced decays easier?
  • $\begingroup$ You are underestimating the vertex detectors and trackers . The could detect an electron of 100 MeV particularly if it came with a decay vertex twiki.cern.ch/twiki/bin/view/CMSPublic/PhysicsResultsFSQ12014 $\endgroup$
    – anna v
    Commented Jan 15, 2014 at 15:01
  • $\begingroup$ @anna v: True, that's why I wrote comfortably detect. The problem is for one the trigger, you have a minimal $p_T$ threshold to trigger on single electrons or muons, and if it's too low, you have to run prescaled (I think at the moment at ATLAS the lowest is at 10 GeV, and then 18 GeV, but I'm not sure). The other thing is that (assuming your event triggered on something) you don't know the trigger efficiencies very well below 15 or 10 GeV (because they were determined e.g. from $Z\rightarrow\ell\ell$. So very often, you can't use those soft leptons as you'd like. $\endgroup$
    – jdm
    Commented Jan 15, 2014 at 15:14
  • $\begingroup$ Are you maye confusing the "at rest" in the cms of the decaying particle with at rest in the lab? If the energy of LHC is high enough there will be enough momentum for detection in some mass ranges of the particles. One has to model this in the monte carlo. $\endgroup$
    – anna v
    Commented Jan 15, 2014 at 16:16
  • $\begingroup$ These are events with a low cross-section and you can't rely on the high-momentum tail to boost the leptons above trigger thresholds. Although the actually difficult case, higgsinos, can give you >~10 GeV leptons some fraction of the time, but the rates are low and the backgrounds are large. $\endgroup$
    – Matt Reece
    Commented Jan 15, 2014 at 16:30
  • $\begingroup$ @annav: Thanks, I forgot to take the random boost of the initial state into account! That could help, even without additional emitted particles. Still, just from experience, there are parameter ranges areas where its not enough and a e.g. multilepton search is just not sensitive. $\endgroup$
    – jdm
    Commented Jan 15, 2014 at 17:31

1 Answer 1


It depends on the degree of degeneracy. Winos are split primarily by loop effects (because the leading tree-level splitting arises only from a dimension 7 operator) and have a mass splitting $m_{{\tilde W}^+} - m_{{\tilde W}^0} \approx 166$ MeV. Due to this small splitting, the charged wino propagates distances of order centimeters before decaying (usually to a soft pion and the neutral wino). So the experiments can look for a disappearing track. ATLAS has done this and excluded wino LSPs below about 270 GeV.

Higgsino LSPs have a larger tree-level splitting: it arises from dimension 5 operators and is of order $\frac{m_Z^2}{M_{1,2} - \mu}$. This is typically of order 10 GeV. This makes higgsinos very difficult to find. Associated production is the way to look, either in the monojet channel (missing energy recoiling against a hard jet) or in the vector boson fusion channel (tagging the event with two forward jets). You can see some recent theoretical work on such signatures here, here, or here, for instance. (This is not a complete list, but it should get you started.) The conclusion is that the LHC is only sensitive in such channels if the masses are very small. This is probably why we haven't seen experimental results along these lines yet: the reach just isn't good enough to say anything interesting yet.

The ILC (or TLEP, or any other future $e^+ e^-$ collider that may be built) would be much cleaner and doesn't have huge QCD cross sections to contend with, so yes, it would do better. LEP, for instance, has ruled out higgsinos up to about 100 GeV (approximately its kinematic limit).


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.