It is indeed a difficult question. Let me answer in a different way. What is your min level of lab analysis of Hg ? If it is 1 micrograms/Nm3 then this would be your lowest detectable limit. This kind of analysis is difficult not to mention the importance of proper sampling of the gas. With atomic absorption spectroscopy you can measure in levels of nanograms.
In general a level below 0.1 micrograms/Nm3 may be considered safe. You've indicated Al-air exchangers - is it something to do with regassification of LNG ? The Aluminium alloy is important if it has more of Mg it will be more susceptible to Hg attack. But remember even if the gas has Hg if the temp is below its melting point i.e. about -34 C Hg attack will not progress.
Air or moisture will do the damage. But first liquid mercury has to form the amalgam with Al which will then extract H2 from moisture and Al starts corroding. The key thing is 1) avoid Hg to go through 2) keep it cool below the melting point of Hg.
This is an interesting subject but unfortunately limited resources are available.
You should also be aware that mercury will form an amalgam with the porous gold electrode present in some moisture measurement sensors.
The resulting amalgam isolates the sensor's metal-oxide layer from the moisture in the natural gas resulting in a non-responsive sensor. This is particularly dangerous as the sensor will still output a signal thereby appearing as if it is still operating properly.
If your natural gas contains mercury, I would suggest using a technology other than a metal-oxide sensor.
