MadGraph5_aMC@NLO

How to simulate events in MG5 if any process has multiple off-shell particles?

Asked by Abhaya Kumar Swain

Hi MG5 experts,

I want to generate events for a process where multiple particles are off-shell. Based on my understanding regarding off-shell particles, I generate events by mentioning its decay products. Since I have multiple off-shell particles, if I mention the decay products in total I am generating a 2->8 process. For this MG5 can list the Feynman diagrams but freezes when saving them in an output file before generating events.

Any help to resolve this issue would be very helpful. Thank you in advance.

Thanks,

Abhaya

Question information

Language:
English Edit question
Status:
Solved
For:
MadGraph5_aMC@NLO Edit question
Assignee:
No assignee Edit question
Solved by:
Olivier Mattelaer
Solved:
Last query:
Last reply:
Revision history for this message
Olivier Mattelaer (olivier-mattelaer) said : 		#1

Hi,

Sorry, but with that little information, I can not help you.
Please give us the model, the process that you are interested in (and the benchmark) without that it is impossible to comment and see if a strategy (other than the 2>8) is possible or not.

Cheers,

Olivier

Revision history for this message
Abhaya Kumar Swain (abhaya.swain) said : 		#2

Hi,

Thank you very much for your quick reply.

I am using a simplified model where the SM is extended with two scalar singlets and a a heavy neutrino. The process I am interested in is,

p p > H > S N all all all, (S > N all all all).

Where H and S are the scalar singlets with mass 270 and 150 GeV. So one S from the decay of H is off-shell. And N is the heavy neutrino with mass 100 GeV. With this mass spectrum, S can not decay to both N on-shell, that is why I generate S> NN*. Since N decays through W boson, the off-shell N will decay through off-shell W boson that makes three body decay (N* > all all all). This make it a 2 > 8 process.

I am sending the model file through email.

It would be very helpful if you kindly suggest anything to generate this process.

Thanks a lot,
Abhaya

Revision history for this message
Best Olivier Mattelaer (olivier-mattelaer) said : 		#3

The issue here is that you have a lot of diagram and a lot of color ordered amplitude.
A lot of those are independently gauge invariant of what seems to interest you and therefore you might want to remove all those diagrams in order to get something manageable.

The only solution here is likely to write some diagram filtering function to achieve what you want.
Additionally, you can split your contribution into two part:
1) generate p p > H, H > .... #H onshell
2) generate p p > H > S S $ H , S > ... #H offshell decaying into two S onshell
You can even continue the splitting with the N

Cheers,

Olivier

PS: Note that if you use
generate p p > H > S N all all all
You might have some symmetry factor issue for the diagram which is p p > H > S S* > S N all all all

Revision history for this message
Abhaya Kumar Swain (abhaya.swain) said : 		#4

Thanks Olivier Mattelaer, that solved my question.