identical particles affect cross section

Asked by Meng Lu on 2021-04-02

Dear experts,

I'm trying to understand the cross section of the following processes, (other kinematic cuts are the defaults, using MG273)

scheme1, p p > l+ l- l+ l- : XS = 0.02456 pb ( l- = e-, mu-)
scheme2, p p > z l+ l-, z > l+ l- : XS = 0.03904 pb (l- = e-, mu-, and set 'True = cut_decays')
scheme3, p p > z z, z > l+ l- : XS = 0.01766 pb (l- = e-, mu-, and set 'True = cut_decays')
scheme4, p p > l+ l- l+ l- $z: XS = 0.00075 pb ( l- = e-, mu-)

scheme1 is the inclusive one, and the motivation of scheme2 is to avoid two off-shell Z.

it seems to me that the two on-shell Z in scheme2 has been calculated twice, as the difference of scheme2 and scheme3 is 0.02138 pb, which is smaller than the inclusive one. do i understand correctly?

Cheers, Meng

Question information

Language:
English Edit question
Status:
Solved
For:
MadGraph5_aMC@NLO Edit question
Assignee:
No assignee Edit question
Solved by:
Olivier Mattelaer
Solved:
2021-04-02
Last query:
2021-04-02
Last reply:
2021-04-02

Hi,

Syntax 2 does not make sense, if the process ZZ does exists and is not killed by phase-space cut or similar.
The issue is not that the computation is done twice but that the symmetry factor can not be set in a consistent way.
In this case you seem to be dominated by the ZZ contribution which indeed needs an additional factor of 2 as symmetry factor but if you process was 100% dominated by ZH then you would not need such symmetry factor. This syntax is just ill defined.

Cheers,

Olivier

Meng Lu (meng-lu) said : #2

Hi Olivier,

Thanks for your prompt reply. i have several further questions, what you said will affect all the similar syntax? i.e., p p > l+ l- z X, z > l+ l-, for example X is a photon.
and what if i use madspin to decay the z in the syntax (leptonical decay), will the issue still there?
The process i really interested is 4lep + 1photon, so the recommended syntax for the process would be p p > l+ l- l+ l- a, is that right? (in this case, i checked that the zza is not the dominant one)

Cheers,
Meng

Dear Meng,

> what you said will affect all the similar syntax? i.e., p p > l+ l- z X, z > l+ l-,

Yes I would expect the same effect/issue for that process.
I have never study such issue in details, my current feeling (but it need to be validated) is that if you label the lepton like this
p p > l1 l2 z X, z > l3 l4
and the you have cuts on l1 and l2 to prevent M(l1 l2) to be close to the z mass, you should be able to have physical cross-section.

One way to enforce that is to use the syntax:
p p > l+ l- z X $ z, z > l+ l-

> The process i really interested is 4lep + 1photon, so the recommended syntax for the process would be p p > l+ l- l+ l- a, is that right?

The syntax "p p > l+ l- l+ l- a" is the safest syntax and should be the one against to compare to see if you can use any other syntax for your analysis. It might be slower to generate events obviously since it includes more Feynman-Diagram/physics.

Cheers,

Olivier

Meng Lu (meng-lu) said : #4

Thanks Olivier Mattelaer, that solved my question.

Meng Lu (meng-lu) said : #5

Hi Olivier,

Thanks very much, it's clear to me now.

Cheers, Meng