Gauge invariant incorporation of particle widths?

[EL]

Introducing particle width via the Breit-Wigner propagator can break gauge invariance.
Anyone know of some "nice" way to incorporate widths while still retaining gauge invariance?

 

[EL]

Hi EL!
For example, have a look at section 2 in this paper.
http://www.arxiv.org/abs/hep-ph?papernum=9602220

 

[Meir Achuz]

You could try replacing m^2 in denominators by (m-i\gamma/2)^2.
If that breaks GI, just stay in that particular gauge.

 

[EL]

I'm not really able to see how that method gives hope to retain gauge invariance if the ordinary Breit-Wigner fails? I mean, all it does just to add one more term, quadratic in "gamma", in the denominators?

(And for two simple cases I've checked, Compton scattering and electron-positron annihilation, it doesn't help to restore GI.)

 

[Meir Achuz]

If an approximation breaks GI. You can do everything in that gauge.
You don't need GI to complete an approximate calclation.

 

[EL]

I would like to have a gauge invariant amplitude since I suspect strong cancelations are important in the certain process I'm interested in.

If I do as you suggest and just stick to a certain gauge, which one is the "correct" to be in when I insert the width in the propagators?