[billnotgatez]

This is an experiment I might attempt.
- Background -
A container of helium (like a balloon) attempts to achieve equilibrium with the outside air. So the container becomes polluted with air and loses some helium to the outside air. Neglecting humidity and given enough time one would expect that both the outside and inside of the container would approximate the below consistency.

 
Clean Dry Air Components Percent
N2  Nitrogen             78.0836
O2  Oxygen               20.9466
Ar  Argon                 0.9345
CO2 Carbon Dioxide        0.0325
Ne  Neon                  0.0018
He  Helium                0.0005
All the rest              0.0005
 
This is an approximation
Clean = no particulate matter
Dry = no humidity
All the rest includes
Kr Krypton
SO2 Sulfur Dioxide
CH4 Methane
H2  Hydrogen
N20 Nitrous Oxide
Xe  Xenon
O3  Ozone
NO2 Nitrogen Dioxide
I2  Iodine
CO  Carbon Monoxide
NH4 Ammonia
 

- The Experiment -

Maintaining safety precautions
Fill a closed container of helium to about 90%
Fill the rest with air
Using a separate smaller container that is heat resistant and non easily subject to corrosion
This container should also have electrolysis electrodes
Put in sodium hydroxide
Heat to melting and perform electrolysis
Remove heat and electricity and place in the larger polluted helium container
Let stand for several hours to allow for full reaction of the sodium
 with the helium / air in the container
Measure the purity of the helium
If the helium is fairly pure remove the sodium container
Repeat using the same sodium container to see if it can be reused
 even with the new compounds formed

-The expected results is that all of the reactive gases will be removed from the helium container
Only helium and argon should be left
The hope is that the sodium container can be reused over and over
Even if it has sodium compounds containing nitrogen oxygen and carbon

[Mitch]

Sodium might not be reactive enough for your purposes. You might have to upgrade to potassium. You'll be loosing Helium through the ballon at a sufficient rate and air will be coming in through the pores of the ballon. Perhaps if you tell us why the helium has to be so pure we can all come up with something more efficient.

[billnotgatez]

I was using a balloon as an example of something that out-gasses helium and in-gasses air. The application will most likely applied to small airship situations where 95% helium purity may be sufficient. Of course 1% of that would be argon. The old method of doing this was a cooling compression system which does not lend itself to small operations. I am using sodium here as a proof of concept. There might be better substances. I think potassium may be too reactive. It may be that we experiment with several different compounds or elements. I just chose sodium hydroxide as an inexpensive plentiful substance.
Regards,
Bill

[billnotgatez]

Some additional notes -
This is an experiment using containers that will not out-gas or in-gas readily. We will by extension apply it to the situation where the container has the out-gassing and in-gassing problem (like a balloon) at a later time. But for our initial experiment we want to control the concentration of gasses.

The selection of sodium hydroxide as a cleansing substance was made for the following reasons -
Cost
Disposability
Relative safety
The discussion about alkali metals already on this web site.

Some other points -
The determination of helium purity is not an easy matter. My initial thought is to use a Wheatstone capacitor bridge with one of the capacitors being 2 plates with the helium and / or air in between.

The concept behind all this is to mimic the idea of a pool filter system, in that you filter out the unwanted stuff and back flush the filter to do it again.

The final procedure may have to rely on solar energy in the form of a solar oven and electric energy coming from photovoltaic cells.

But let us not jump ahead - the goal hear is to test a method to increase helium purity.