Feynman gauge vs. unitary gauge

[Martin]

Dear Florian,

We were using SARAH to generate some CalcHep Output. We have used MakeChep[FeynmanGauge -> True]. The generated model files seem to be fine, but in the created *.cpp files there is the line

Code: Select all

ForceUG=1;

This forces micromegas to use unitary gauge and the results are likely to be wrong. It is easy to fix by hand, but I was wondering, if this could be treated directly by SARAH in the future.

Thanks,
Martin

[FStaub]

Hi Martin,

why are you so sure that the results are wrong? Could you please give an example.

My experience is that for a consistent input both gauges give the same results (as they should). Unitary gauge is usually faster because of less diagrams and therefore the default choice.

Cheers,
Florian

[Martin]

Hi Florian,

I am not totally sure about this. In the model we studied I could not yet find a point which differs.

But one of my collaborators, who has much more experience with CalcHep and micromegas, had problems with this kind of conflict before. He said this is more obvious in Higgs observables and Higgs portal dark matter models. In the past, he even encountered negative cross sections when he tried to use a model file in Feynman gauge while micromegas assumed unitary gauge. It sounds reasonable that if you force micromegas to use unitary gauge it might ignore the diagrams involving the Goldstone bosons leading to wrong results.

In any case, this is not really a problem of SARAH and an experienced CalcHep user should know this anyway. It is more of a suggestion to set this flag consistently automatically.

Thanks,
Martin

[FStaub]

Hi,

sorry, I must say that I'm not really convinced. I tried many models so far with CalcHep/micromegas and never encountered this problems once the model and the input was consistent. It sounds more like a buggy model file or a not well defined model where gauge invariance is broken from the very beginning (like simplified dark matter models with massive vector mediators). So, if there is a significant difference between Feynman and unitary gauge, then I wouldn't trust neither of the both!

Cheers,
Florian

[Martin]

Dear Florian,

Indeed you are correct. The problems my colleague mentioned occurred in the context of simplified models.

Sorry for bothering and thanks a lot,
Martin