Higgs production with large width

Hi Olivier,

Many thanks for your reply. Actually I have also tried to decay Higgs in MG5. I have tried the following:
1) pp>h>zz [real=QCD]
then Z bosons have exactly the mass of 91 GeV. I think this is an expected feature for Madgraph_aMC@NLO.
You can see the plots at: https://lxu.web.cern.ch/lxu/public/MG_v2.3.2/MG_v2.3.2_Heft_ZZ_nodecay.html
The Z boson decays are handled by Pythia 8.
I know that Madspin can be used for h->ZZ decay with proper Z boson width effect. I'll try that later.

2) pp>h>eemumu [read=QCD]
then I found that there are huge contribution from h->aa, as you can see the plots at:
https://lxu.web.cern.ch/lxu/public/MG_v2.3.2/MG_v2.3.2_Heft_ZZ_decay.html
So my question is, why there should be Hgammagamma vertex in MG5.
You can see the Feynman diagrams generated by MG at:
https://lxu.web.cern.ch/lxu/public/MG_v2.3.2/PROCNLO_heft_truncated-no_b_mass_0_P0_gg_h_epemmupmum.pdf

3) to remove the contribution from photon, I then did
pp>h>eemumu/a [read=QCD]
by doing this, the results look more reasonable to me now:
https://lxu.web.cern.ch/lxu/public/MG_v2.3.2/MG_v2.3.2_Heft_ZZ_decay_noGamma.html

Hi,

2) pp>h>eemumu [read=QCD]
then I found that there are huge contribution from h->aa, as you can see the plots at:
https://lxu.web.cern.ch/lxu/public/MG_v2.3.2/MG_v2.3.2_Heft_ZZ_decay.html
So my question is, why there should be Hgammagamma vertex in MG5.
You can see the Feynman diagrams generated by MG at:
https://lxu.web.cern.ch/lxu/public/MG_v2.3.2/PROCNLO_heft_truncated-no_b_mass_0_P0_gg_h_epemmupmum.pdf

Well “ggh” and “aah” are both loop-induced vertices which are added by the heft model.
This is the content of the model. If you use the “sm” model then you will not have any of those two vertices.
This is just a model choice. I do find natural to have either both or none since both are loop induced vertex with top which can (partly)
be approximate to an effective vertex. Note that since arXiv:1507.00020, we are able to handle loop-induced processes as well in MG5.

Cheers,

Olivier

PS: Note that one think that I do not understand is why you use the [real=QCD] syntax?
Is their any particular reason to compute only a part of the NLO computation?

Hi Olivier,

Many thanks for your explanation. Do you think it makes sense to use the process
pp>h>eemumu/a
by excluding "aah" loop-induced vertices? Or should I include it?
Or it makes more sense to use
pp>h>zz
with MadSpin?

For the syntax [real=QCD], that's also one of my questions. Actually I got this syntax from someone else. I'll ask if there is any
reason for that. Thanks a lot for point this out.

Thank you!
Lailin

Hi,

The main question is which peaks are you interested in.
In the final state to lepton, you have two area.
One where the Higgs is on shell and one of the Z off-shell
and one where the Higgs is off-shell and both Z on shell.

You will only have have the second if you do pp>h>zz.
So this is fine if you are interested in that contribution only.

> pp>h>eemumu/a

Includes both contribution.
Obviously for the the off shell Z contribution, the interference with the photon might be important (I do not know) if this is the case
the above syntax is obviously not relevant.

> For the syntax [real=QCD], that's also one of my questions. Actually I got this syntax from someone else. I'll ask if there is any
> reason for that. Thanks a lot for point this out.

Except very specific need (like if you build a custom NLO computation with various tools) , you should not use that option.

Cheers,

Olivier

On 31 Aug 2015, at 08:46, Lailin Xu <email address hidden> wrote:

> Question #270891 on MadGraph5_aMC@NLO changed:
> https://answers.launchpad.net/mg5amcnlo/+question/270891
>
> Status: Answered => Open
>
> Lailin Xu is still having a problem:
> Hi Olivier,
>
> Many thanks for your explanation. Do you think it makes sense to use the process
> pp>h>eemumu/a
> by excluding "aah" loop-induced vertices? Or should I include it?
> Or it makes more sense to use
> pp>h>zz
> with MadSpin?
>
> For the syntax [real=QCD], that's also one of my questions. Actually I got this syntax from someone else. I'll ask if there is any
> reason for that. Thanks a lot for point this out.
>
> Thank you!
> Lailin
>
> --
> You received this question notification because you are an answer
> contact for MadGraph5_aMC@NLO.

Hi Olivier,

Thanks a lot for your detailed clarifications!

Lailin

Thanks Olivier Mattelaer, that solved my question.