MadGraph5_aMC@NLO

Generating Higgs plus jet at NLO with light quark contributions

Asked by Nordin Breugelmans

Dear MadGraph experts,

We are trying to simulate p p > h j where j is any parton (in the 5 flavour scheme) at NLO. To this end, we generate the following process:

generate p p > h j [QCD]

In addition to the cases where the additional parton is a b/c quark or gluon, we are particularly interested in the contribution made by light quarks (uds) at NLO. However, if we use the default loop_sm model that is shipped with the MadGraph distributions, we are unable to properly generate the relevant processes. This is due to the loop_sm model only featuring bottom/charm - Higgs couplings thus meaning processes such as e.g. u g > h u are not produced by the above command due to being loop-induced and not existing at tree level.

We have so far considered 3 different solutions to this problem:

1) Use a SM UFO which includes light quark Yukawas and is capable of NLO simulation. However, we have not been able to find such a model and attempts at producing our own with FeynArts and NLOCT have been unsuccessful.

2) Attempt to stitch together a sample consisting of both the QCD and loop-induced contributions to the p p > h j [QCD] process by generating them individually and using appropriate flags for virtual, real etc. contributions. However, we are not sure if this is feasible and whether the results will be physical if e.g. destructive interferences play a role here.

3) Use a UFO model which implements the effective Hgg vertex. Ideally we would like to avoid this approximation.

We would be very grateful if you could weigh in on this and might be able to give us some advice on how to proceed.

Kind regards,
Nordin

Question information

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

Hi,

I'm not sure what you try to do.

In MG5aMC, you can not compute loop induced amplitude (so loop square) which are handle like LO computation
with any other type of computaton (either other tree-level LO computation or NLO computation which involves tree times loop computation).

So if you try to do loop-induced + NLO, then you are either need to deeply change the code or to do some stitching like you say in your option 2. The problem here is not at the UFO level but in the expansion since the loop-induced is part of the NNLO contribution.

Now for the loop-induced effect, we know that the u/d/s loop squared are fully negligible.
The inclusion of such loop actually leads to one percent effect but those are due to the interference with the top loop.
 Given that fact, I doubt that keeping the u u~ > h at tree level and doing NLO correction on top does that much sense since the interference with the top does not come to a rescue.

So my advice would be to add by hand the missing R2 counter-term within the UFO and do the pure loop-induced computation (potentially using diagram filtering). But as said above this might not be what you want to study.

Cheers,

Olivier

Can you help with this problem?

Provide an answer of your own, or ask Nordin Breugelmans for more information if necessary.

To post a message you must log in.