negative cross sections when using the hGG/AA coupling model

I should add I get negative cross sections regardless of whether im using 1d integration or the monte carlo simulation

For SM( with hGG/AA) implemented in CalcHEP for Pcm=1TeV I have got

Cross section 20.65 pb.ggs

I see the same result  removing  h->AA vertex.  It should not play a
role here because of small contribution from virtual Higgs.

Do you have negative cross section without h->AA ?

Which model  you use?

Best

    Alexander Pukhov

On 03/15/2018 07:17 PM, chris wrote:
> New question #665678 on CalcHEP:
> https://answers.launchpad.net/calchep/+question/665678
>
> Hello,
>
> I am currently using calchep to calculate the cross sections of different interactions, and I am running into a problem when using the hGG/AA coupling where some interactions that can only happen through that coupling have negative cross sections. For example, with both particles momentum set to 1 TeV, I get negative cross sections for w+ w- > g g, z z > g g, and z z > a a. Is this the hGG/AA model behaving as expected, or is something going wrong here?
>
> Thanks,
>
> Chris
>

I'm using the SM(CKM=1 with hGG/AA ) model. The specific values for the cross section I 'm getting at 1 TeV are

ww>gg: -3.22e-3 pb
zz>gg: -4.14e-3 pb
zz>aa: -1.23e-4 pb

Where you testing w+ w- > a a? I do get a similar positive result for that one at 1 TeV, though I'd note that at Pcm = 1e6 GeV the cross section becomes negative (speciffically calchep gives -2.86e-5 pb). In this case I can resolve the problem by excluding the diagram with the hAA vertex, but I was hoping to resolve this problem for interactions that can only occur through hGG/AA coupling, such as the ones above.

Thanks,

Chris

While not strictly related, another problem I am having is getting different result for the cross section of zz > aa (as I previously stated I get -1.23e-4 pb) vs aa > zz (for this calchep gives -2.74e-4). Again these results use Pcm=1 TeV and the SM(CKM=1 with hGG/AA ) model. Do you have any idea why the two interactions give different cross sections?

Dear Chris,

please try this Standard Model implementation
http://hepmdb.soton.ac.uk/hepmdb:0318.0304

In this version we have implemented the effective HGG and HAA vertices
via gauge invariant H^2 F^2 interactions.

Previous implementation of HGG and HAA was valid for the the Higgs being close to on-shell state
since we were assuming that off-shell case which gives low cross sections is not so interesting.

The latest implementation above is general and should solve the problems you have faced with.
We confirm that the problem you have faced is related to not very generic implementation of the effective
 HGG and HAA vertices.
[The model above will replace the old one in the online CalcHEP version tomorrow.]

Regards,
Alexander on behalf of the CalcHEP team

Hello,

Using this model did resolve the negative cross section problem, but I am still getting different cross sections depending on the direction I run the Interaction in. For instance, zz>aa now gives a cross section of .0148 pb at Pcm=1 TeV, whereas aa>zz gives .0328 pb. Do you have any idea what might be causing this?

Thank you for your help,

Chris

In cross section formula we have a non-symmetric  factor     1/(
n_polarizarion  states for incoming particles)

Arithmetic:  0.0148*9/4=0.333

Another such factor is   Pcm_out/Pcm_in.  It should be close close to
one in  both cases  for  1 TeV.  May be you'll get 0.0328 taking this
factor into account as well.

Best

    Alexander Pukhov

On 03/16/2018 05:47 PM, chris wrote:
> Question #665678 on CalcHEP changed:
> https://answers.launchpad.net/calchep/+question/665678
>
> Status: Answered => Open
>
> chris is still having a problem:
> Hello,
>
> Using this model did resolve the negative cross section problem, but I
> am still getting different cross sections depending on the direction I
> run the Interaction in. For instance, zz>aa now gives a cross section of
> .0148 pb at Pcm=1 TeV, whereas aa>zz gives .0328 pb. Do you have any
> idea what might be causing this?
>
> Thank you for your help,
>
> Chris
>

Dear Chris,

cross sections for
zz->gg vs gg->zz
MUST be different because of the different averaging
for the initial state,
see explanation of Alexander Pukhov

Alexander Belyav

On 16/03/18 14:47, chris wrote:
> Question #665678 on CalcHEP changed:
> https://answers.launchpad.net/calchep/+question/665678
>
> Status: Answered => Open
>
> chris is still having a problem:
> Hello,
>
> Using this model did resolve the negative cross section problem, but I
> am still getting different cross sections depending on the direction I
> run the Interaction in. For instance, zz>aa now gives a cross section of
> .0148 pb at Pcm=1 TeV, whereas aa>zz gives .0328 pb. Do you have any
> idea what might be causing this?
>
> Thank you for your help,
>
> Chris
>

--
Prof. Alexander S Belyaev (<email address hidden>)
https://www.hep.phys.soton.ac.uk/content/alexander-belyaev

School of Physics & Astronomy, University of Southampton, Office: 5047
SO17 1BJ, TEL.: +44 (0)23 8059 8509; FAX.: +44 (0)23 8059 3910
.....................................................................
Particle Physics Department, Rutherford Appleton Laboratory, Didcot,
OX11 0QX, TEL.: +44 (0)1235 445562; FAX.: +44 (0)1235 446733
.....................................................................
CERN: Office: 40 1-B20 Mailbox: E27910; ccid: 532076
TEL: +41 22 76 71642
______________________________________________________________________

Thank you for all your help! As one final note after further testing I noticed that , even with the updated model, for the process ww>ha I'm still getting a cross section of -5.2e-5 pb at Pcm=1e6 GeV. Aside from that interaction the new model seems to have resolved the problem.

Thanks again,

Chris

Dear Chris,

ww>ha is DIVERGENT process if you do not apply any cuts. Did you?

Before claiming the problem, please make sure that the accuracy for the result you are getting is reasonable , say ~ 1%

Is this the case?

If it is not there could be several problems -- e.g. you did not apply cut on th ephoton PT which make the tree-level cross section finite. If this is not the case (i.e. your process in divergent) the numbers you are getting from the integration does not make sense.

Regards,
Alexander