MadGraph5_aMC@NLO

Generating semi-leptonic WWgamma with 0,1,2 jets

Asked by Alessandra Baas

Hi

We want to generate a WWgamma sample with one W->jj and the other W->l vl with up to two additional hard jets. Since we also would like to study b-tagging requirements later, we include b quarks in the p and j definition. So we generate:

define j = g u c d s b u~ c~ d~ s~ b~
define p = g u c d s b u~ c~ d~ s~ b~
generate p p > e- ve~ w+ a, w+ > j j
add process p p > e- ve~ w+ a j, w+ > j j
add process p p > e- ve~ w+ a j j, w+ > j j

In order to account for matching we set ickkw=1, xqcut=20 (also ptj=ptb=20) and maxjetflavor=5. in the run_card.py.

The first question which came up is if it is necessary to explicitly include QCD radiation from the decaying legs:
add process p p > e- ve~ w+ a , w+ > j j j
add process p p > e- ve~ w+ a j, w+ > j j j
add process p p > e- ve~ w+ a, w+ > j j j j
I found a bug report from 2012 (https://bugs.launchpad.net/mg5amcnlo/+bug/1017279 comment 3) where you explicitly say that this should not be done, since the matching with pythia will go wrong and one should restrict the QCD radiation to the core process. Is this still valid and to you know if pythia also has a build-in matrix element correction for the parton showers of W decays?

Another question which came up is if it is ok to use the default 'sm' model, include the b-quarks in the p and j definitons and require then maxjetflavor=5 in the run_card.py. Or should we set maxjetflavor=4, since we want the b quark to be massive (in the run_card it says 'maximal pdg code for quark to be considered as a light jet' )?
Or do we also need to switch to a different model, such as 'sm-no_b_mass'? Could you please clarify when such a model (with no b quark masses) is needed?

Thanks for your help,
Cheers,
Alessandra

Question information

Language:
English Edit question
Status:
Solved
For:
MadGraph5_aMC@NLO Edit question
Assignee:
No assignee Edit question
Solved by:
Valentin Hirschi
Solved:
Last query:
Last reply:
Revision history for this message
Valentin Hirschi (valentin-hirschi) said : 		#1

Dear Alessandra,

Yes, the 2012 question and answer is still valid and the syntax to be used is

add process p p > w- w+ a , w+ > j j, w- > e- ve~ @1
add process p p > w- w+ a j, w+ > j j, w- > e- ve~ @2
add process p p > w- w+ a j j, w+ > j j, w- > e- ve~ @3
...

(It is not clear to my why you didn't seem to also use the decay chain syntax for the w- decay).
The variables 'maxjetflavor' only decides whether the jet cuts of the run card should also apply to external state b-quarks.
When the b-quark is massless, you certainly want maxjetflavor set to 5. When the b-quark is set massive, you can chose maxjetflavor to be either 4 or 5 depending on what you need.

I know that Pythia can decay W's while retaining some of the spin correlations in the decay product, but you will have to ask the authors for details.

If you want to keep the b-massive, you of course cannot use the -no_b_mass restriction. This restriction is used to simplify the model and is preferable (and mandatory at NLO) whenever setting the b massless.
Given that including b-quarks in the initial state, with their associated PDF is inconsistent when considering the b-quark massive, I would advise you to use the b-quark massless altogether, and within the sm-no_b_mass model..

Revision history for this message
Alessandra Baas (alessandra-baas) said : 		#2

Hi Valentin,

Thanks for the helpful answer, that already made a lot clearer.

> (It is not clear to my why you didn't seem to also use the decay chain syntax for the w- decay).

We chose the other notation for two reasons. First we recognized, that when using the decay chain syntax for both w's, there is no photon radiation from the decaying lepton included. Compared to the notation with only one w giving in the decay chain, we see that this diagram is included. And second, we might not be able to set a cut on the transverse mass on the leptonically decaying w boson and hence we thought it would be better to not restrict the l vl system to arise from an onshell w. Altogether we get a difference in the cross sections of ~30% between the two notations (when only considering the 0 jet case). Or are there good reasons for using the decay chain syntax for both w's?

About the b-quark question, if we understand correctly our options are: if we want to consider b quarks in the initial state, we should use the sm-no_b_mass model and set maxjetflavor=5. For massive b quarks, we should not include them in the initial state and decide to set maxjetflavor to 4 or 5.

Thanks again,
Cheers,
Alessandra

Revision history for this message
Best Valentin Hirschi (valentin-hirschi) said : 		#3

Your understanding of the decay chain is correct and I agree that in your situation it is best not to force the w- for the leptonic decay.

And regarding the b-quarks, yes again, the two possible so-called flavor schemes are:

> The 5-flavor scheme with massless b-quark and included in the initial state (i.e. non-zero PDF's for them).
> The 4-flavor scheme with massive b-quarks not included in the initial state (make sure to use the 4-flavor scheme PDF fit when using it).

In your case, the 5-flavor scheme should be preferred.

Revision history for this message
Alessandra Baas (alessandra-baas) said : 		#4

Thanks Valentin Hirschi, that solved my question.