MadGraph5_aMC@NLO

Gluon as mother particle for photon in MG5_aMC_v2_9_15

Asked by Andrej Saibel

Hello,

I run MG5_aMC_v2_9_15 with python2.7

loop_sm-no_b_mass
generate p p > t t~ a [QCD]

I want to quickly check the LHE files and I get events like these:

<event>
  5 0 0.63305562E+01 0.43975668E+02 0.75467711E-02 0.10526398E+00
       21 -1 0 0 503 502 0.00000000000000000E+00 0.00000000000000000E+00 0.59906249675831475E+02 0.59906249675831475E+02 0.00000000000000000E+00 0.0000E+00 0.9000E+01
       21 -1 0 0 501 503 0.00000000000000000E+00 0.00000000000000000E+00 -0.84148361406974732E+03 0.84148361406974732E+03 0.00000000000000000E+00 0.0000E+00 0.9000E+01
       22 1 1 2 0 0 0.96421558175218962E+01 -0.12987351029689830E+02 -0.52292951036017797E+02 0.54737511668261938E+02 0.00000000000000000E+00 0.0000E+00 0.9000E+01
        6 1 1 2 501 0 -0.48061513370337032E+02 0.30947582381212840E+02 -0.59864651862625294E+03 0.62575899209361876E+03 0.17300000000000000E+03 0.0000E+00 0.9000E+01
       -6 1 1 2 0 502 0.38419357552815136E+02 -0.17960231351523007E+02 -0.13063789473164519E+03 0.22089335998369816E+03 0.17300000000000000E+03 0.0000E+00 0.9000E+01
#aMCatNLO 1 0 0 2 4 0.32500759E+02 0.00000000E+00 9 6 0 0.00000000E+00 0.00000000E+00 0.00000000E+00 0.00000000E+00 0.00000000E+00

To my understanding, the event indicates that the mothers of the photon are partons 1 and 2 which are both gluons. Is this expected or am I missing something?

Kind regards,
Andrej

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,

It just means that we do not force the parton-shower to use some specific parton-shower history associated to that photon.
For the events that you quote this is indeed the correct things to do.

Note that 3 particles have 1 and 2 has mothers. This does not mean that you have a g g fusion producing a photon.

Cheers,

olivier

Revision history for this message
Andrej Saibel (asaibel) said : 		#2

Hi Olivier,

Thank you very much for the answer! It's very concise. Just one additional question (I promise it's the last one in this thread):

I'm trying to look at the top mass dependence of the photon radiation. I see a suppression at low delta R to the closest top quark, so I assume the effect is implemented. However, if I change the top quark mass to about +-2.5 GeV, I don't see an impact on the normalized differential distributions of delta R or the closest top quark energy times delta R.

Is the mass variation just too small to see effects in the LO calculations or is the mass effect on the radiation not taken into account in MG5?

Thanks again for the help!

Kind regards,
Andrej

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

Hi,

Not sure which effect you are looking for/speaking about. In our code, we use the full matrix-element square. The top-quark will be massive and the propagator will have a width (which is technically higher order). Likely the effect you are looking for is part of amplitude computation and in that case is indeed simulated.
For example, since we include all the Feynman Diagram, the cancellation between the soft photon emission from the top and the anti-top will automatically be present.

Concerning how much should be the impact on small shift of the top mass. I just never study such impact and do not have intuition here. I would just make one point is that your shift is at the same order as the width. And the presence of such width might shadow some of those effects. It maybe interesting to look at this process (at LO) when setting the top width to zero.

Cheers,

Olivier

PS: for the syntax that you quote above (a NLO syntax), the top width might be set automatically to zero. If you run at LO with the above syntax, I'm not sure if it is done or not, the best is that you check the stdout of the program to see if you see an information line telling that the top width was set to zero or not.

Revision history for this message
Andrej Saibel (asaibel) said : 		#4

Thanks Olivier Mattelaer, that solved my question.