MadGraph5_aMC@NLO

VBS quark pt from 5f vs 4f

Asked by Jonathan Guiang

Hello,

We have been looking at generating VBS W events using a 5-flavor scheme. In particular, we were looking at the pT of the outgoing VBS quarks in the LHE files we produced and we noticed that the pT of outgoing VBS b quarks is noticeably larger than that of outgoing VBS udcs quarks. In addition, we see that the eta of the VBS b quarks is central, whereas the udcs quarks are more forward as we would expect. Is this correct?

You’ll note that in our proc card (below), we are specifically generating p p > l nu j j events, so we could also have WW and WZ events. We apply an Mjj cut of 150 GeV to remove these events.

Plot of the quark pT separated into events with VBS b quarks (yellow) vs.events without VBS b quarks:
http://uaf-10.t2.ucsd.edu/~phchang/dump/forJonathan/4f_v_5f/analysis/plots/Wgt__quarks_pt.pdf

Plot of the quark eta separated as above:
http://uaf-10.t2.ucsd.edu/~phchang/dump/forJonathan/4f_v_5f/analysis/plots/Wgt__quarks_eta.pdf

Proc card:
http://uaf-10.t2.ucsd.edu/~phchang/dump/forJonathan/4f_v_5f/ewkwplvjj_5f_LO/Cards/proc_card_mg5.dat

LHE file:
http://uaf-10.t2.ucsd.edu/~phchang/dump/forJonathan/4f_v_5f/analysis/ewkwplvjj_5f_LO_mmjj150.lhe

Thanks!

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

Those are plot at parton-level? (based on that lhef file?)

Cheers,

Olivier

Revision history for this message
Jonathan Guiang (jguiang) said : 		#2

Yes, these plots are made at parton-level, i.e. based on the LHE file I linked.

Revision history for this message
Jonathan Guiang (jguiang) said : 		#3

Hello,

After some deliberation, we believe we have found the issue. In the original proc card, we had the following:

generate p p > l+ vl j j / t t~ h QCD=0

However, it seems that this excludes a diagram where a top quark decays into the leptonically decaying W plus a b quark. Moreover, we believe that this excludes a significant interference term, and is thus causing events with an outgoing b quark to have high pT. To test this, we generated another set of events:

generate p p > l+ vl j j $ t t~ / h QCD=0

This gives a pT distribution[1] for outgoing b quarks that makes more sense to us. We believe this is the solution to the issue we were seeing before, but welcome any further input.

1. http://uaf-10.t2.ucsd.edu/~phchang/dump/forJonathan/4f_v_5f/dollarsign/Wgt__pdgid5_quarks_pt.pdf

Best,
Jonathan

Revision history for this message
Best Olivier Mattelaer (olivier-mattelaer) said : 		#4

Hi,

Including the offshell top diagram should change even more the distribution of the "b" compare to the other jet.
Since the top decays only into b (or preferentially depending of the approximation of the CKM).

Now one big difference of the "b" compare to other jet is that you do not the diagram with a S-channel W.
So for example you do have that type of diagram for
 u d~ > e+ ve q q~ QCD=0 / t t h @1
and for
c s~ > ...
but you do not have that contribution for "b" initial state.
and those can indeed impact the observed b spectrum and make it different from the other jet.

And I guess this explains why they do differ. Adding the off-shell top is important to have a more accurate description but only increase the change of behavior with the other flavor.

Cheers,

Olivier

Revision history for this message
Jonathan Guiang (jguiang) said : 		#5

Thanks Olivier Mattelaer, that solved my question.