There was a discussion on how to transfer beer and minimize staling in the beer here.
https://www.homebrewersassociation.org/forum/index.php?topic=23575.msg301175#msg301175Sean makes the statement that if you have a DO meter you become shocked as to how much purging is necessary to achieve low levels. How much is that?
There are procedures to purge vessels of toxic or explosive gases. If you search for "pressure cycle purge equation" you will get hits to some links with information. Without going into all the details and the derivation, you can estimate how many purge cycles you need to reach a level of gas concentration.
N=ln(C
_{low}/C
_{original})/ln(P
_{low}/P
_{high})
N=Number of purge cycles
C
_{low}=Lower concentration target
C
_{original}=Original concentration, so for air that is 0.21
P
_{low}=Low pressure, 14.7 PSI absolute for our application
P
_{high}=High pressure for the purging, I use 30 PSI gauge, so 44.7 PSI absolute
For our case.
N=ln(C
_{low}/.21)/ln(14.7/44.7)
selecting a sequence of 0.1, 0.01, 0.001... we can calculate the number of cycles. Round up for your application.
Concentration #purge cycles.
0.1 0.6671342451 0.01 2.7375696785 0.001 4.8080051119 0.0001 6.8784405452 0.00001 8.9488759786 0.000001 11.019311412 0.0000001 13.0897468454 0.00000001 15.1601822787 0.000000001 17.2306177121
So to get close to 1 ppm (0.000001) you need 11 purge cycles. State of the art canning and bottling lines have about 50 ppb TPO or less.
That will use a boatload of CO2 for an empty keg, so the technique of pushing sanitizer out saves on CO2. If the head space is purged 5 times at 30 PSI, then the headspace will have about 1000 ppm O2. The headspace is small, so there is not a lot of waste for that process.
I hope this helps.
