Exclusion plots and cross-section?

[frazzle]

Hello everyone,

As I've mentioned in a thread in the cosmology forum, I'm currently reseraching dark matter.

I am often confronted with 'exclusion plots' for the results of WIMP direct detection (via nuclear recoil) experiments, where one axis depicts WIMP mass, and the other depicts cross section.

Am I correct in saying that these plots illustrate the experimental sensitivities for which the presence of a particle has been investigated and - in the event of a null result - ruled out?

In which case, I understand how an experiment could be sensitive to a range of particle masses/energies, but I am a bit confused about the cross-section part. Am I right in thinking that cross-section is the likelihood of a collision event?

here is an example!:

http://img55.imageshack.us/my.php?image=exclusionplot6sc.jpg

if anyone could go any way towards answering these questions I would really appreciate it!

 

[SpaceTiger]

In which case, I understand how an experiment could be sensitive to a range of particle masses/energies, but I am a bit confused about the cross-section part. Am I right in thinking that cross-section is the likelihood of a collision event?

It's proportional to it, yes. When you see constraints on the dark matter cross section, it's probably due to direct detection experiments where they look for the products of a dark matter particle interacting with something ordinary, like a nucleus.

As I understand it, mass constraints come also from collider experiments. If they were to create a dark matter particle in a reaction, then it would appear as missing energy; that is, the interaction would appear as if it weren't conserving energy. The same thing can be done to identify neutrinos. As they move to larger energy scales, it's more likely that they will produce the dark matter particle in a reaction, if it exists.

 

[Meir Achuz]

In order to put a mass limit, a size of the cross-section for its observation has to be assumed. The use of cross-section is:
N_seen=(N_incident per unit area)*(cross-section).

 

[frazzle]

Thankyou both for your explanations, they are a big help. I'm still a little confused though. When we talk about the cross-section, we are referring to a cross-sectional area? Is this the cross-section of the incident particle or the target?

It makes sense to me that a larger cross-section for either of these particles would increase the likelihood of a collision, and hence, as spacetiger mentioned, they are proportional.

My main source of confusion I suppose is this: within collision detectors particularly, I understand the exclusion of any given particle mass to be a result of no detected events in an apparatus which should be sensitive to the recoil energies caused the mass in question. However, I don't understand how the cross-sections (on the other axis of the exclusion plot) are ruled out.

Is it a case of, if a particle with mass (whatever)MeV and cross-section (whatever)cm should have been detected if present?

Or perhaps a case of varying the target nuclei (Ge, Xe, Si etc) and ruling out cross-sections that way?

Sorry if these questions are a bit silly! Thanks again.

 

[Meir Achuz]

I'm still a little confused though. When we talk about the cross-section, we are referring to a cross-sectional area? Is this the cross-section of the incident particle or the target?

"Cross-section" is not an actual area. It just has the dimensions of area.
Operationally it is used as in my first equation.
N_seen=(N_incident per unit area)*(cross-section).
You should read a QM or Nuclear Physics text to get more about it.

 

[Meir Achuz]

However, I don't understand how the cross-sections (on the other axis of the exclusion plot) are ruled out.

Some model is assumed to predict the number of particles entering the detector per unit area. Multiplying this by an assumed cross-section gives the number of partricles that would be observed if that were the cross-section.

 

[frazzle]

Some model is assumed to predict the number of particles entering the detector per unit area. Multiplying this by an assumed cross-section gives the number of partricles that would be observed if that were the cross-section.

thankyou :)