off-shell W boson and Z boson decay in madspin

Asked by Di Zhang

Dear MadGraph Team,
          I generate p p > x1- n2 and p p > x1+ n2, and in proc_card.dat, I do:
          import model MSSM_SLHA2 --modelname
          generate p p > n2 x1-
          add process p p > n2 x1+

          The masses and decay modes of particles in param_card.dat are:
           23 9.118760e+01 # MZ
          24 7.982901e+01 # MW
          1000022 235.0 # Mneu1
          1000023 250.0 # Mneu2
          1000024 250.0 # Mch1
          1000025 2.5e+03 # Mneu3
          1000035 2.5e+03 # Mneu4
          1000037 2.5e+03 # Mch2
           .......

          DECAY 23 2.411433e+00 # WZ
          DECAY 24 2.002822e+00 # WW
          # BR NDA ID1 ID2 ...
         1.000000e+00 2 23 1000022 # 0.020777
         DECAY 1000024 1.704145e-02 # Wch1
         # BR NDA ID1 ID2 ...
        1.000000e+00 2 24 1000022 # 0.01704145
         ......

         In run_card.dat, I do:
         15 = bwcutoff
          .......

          The following switches determine which programs are run:
          1. Choose the shower/hadronization program shower = Pythia8
          2. Choose the detector simulation program detector = OFF
          3. Choose an analysis package (plot/convert) analysis = OFF
          4. Decay onshell particles madspin = ON

           In madspin_card.dat,I do:
           set spinmode none
           set max_weight_ps_point 400
           decay x1- > w- n1, w- > l- vl~
           decay x1+ > w+ n1, w+ > l+ vl
           decay n2 > z n1, z > l- l+

           I got:
           Command "generate_events " interrupted with error:
           ZeroResult : No Phase Space. Please check particle masses.

           I think there must some thing wrong, but I do not know where it is. Exactly, I want to know how to let the off-shell W boson and Z boson decay to leptons in madspin?

            Thank you in advance! And I am looking forward to your reply soon!

Yours sincerly,
Di Zhang.

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,

the decay syntax (the one with the comma) assume that the particle is (quasi) onshell.
and therefore if I read correctly your syntax this is not the case and therefore you can not use the syntax
decay x1- > w- n1, w- > l- vl~
in this context (since you are in spinmode=none, i.e. no offshell effects and no spin correlation) you should use
decay x1- > n1 l- vl~

Cheers,

Olivier

Revision history for this message
Di Zhang (ddou) said :
#2

Hi,
      Tank you for your reply!

       Firstly, if I am in spinmode = none, the offshell effects will not be considered. Am I right? If it is indeed, witch spinmode should I choose to only consider the offshell effects?

       Secondly, How many choices of spinmode dou I have? I set spinmode = full, however, it told me that it is not valid.

       Thirdly, I set spinmode = none and use
       decay x1- > n1 l- vl~
       decay x1+ > n1 l+ vl
       decay n2 > l- l+ n1,
       but I got:
       Command "generate_events " interrupted with error:
InvalidCmd : The bridge mode of MadSpin does not support event files where events do not *all* share the same set of final state particles to be decayed. One workaround is to force the final cross-section manually.
       So, what is the correct decay syntax?

      Thank you in advance! And I am looking forward to your reply soon!

Yours sincerly,
Di Zhang.

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

We do have three mode

"full" described in 1212.3460. limited to a series of two body decays.

"none" which is just gluing together two sample file (one for the production and one for the decay). This is in general quite unaccurate method for decaying particle (no spin correlation between production and decay and no offshell effect) . And was released only because of loop-induced process for scalar particle.

"onshell" which works as none but implement a re-weighting on top to restore spin-correlation (but no off-shell effects implemented )

So in your case, we do not offer any solution with offshell effects.

For your third question, the easiest workaround is to split your generation in two
The issue is that your sample does not have the same particle decaying in all event.
          generate p p > n2 x1-
          add process p p > n2 x1+
Some event will decay x1- and some other x1+. This case is not handle.
the easiest solution is to do two separate run and merge the hepmc or root file (or any format you want) at the end of the run.
The other option is to tell madspin to not compute the new cross-section and just provide it as input but this is quickly quite complex.

Cheers,

Olivier

Revision history for this message
Di Zhang (ddou) said :
#4

Hi Olivier!
     Thank you for the detailed explanation!

Yours sincerly,
Di Zhang.

Revision history for this message
Di Zhang (ddou) said :
#5

Thanks Olivier Mattelaer, that solved my question.