Some research and published guidelines into this is long overdue.
Mitchell and Gant's presentation did also address the subject of scrubber storage. A previous reference they cited was a Canada Defence document that concluded "
Storage sealed in a rebreather for one week cause no change in duration".
They performed an experiment to test "overnight open" and "1 month, open and sealed" storage of Inspiration (Sofnolime 797) canisters with 90 minutes of prior use. Their ambient atmospheric condition was 20 degC, 60 %RH, 0.03% CO2. After 28 days storage the "open" stored canister showed approximately 40 minutes less duration than the "sealed" stored canister. The overnight "open" canister was very similar performance to the 28 day "sealed" canister. Their summary:
Outcomes
Storing a scrubber between dives can influence CO2 absorbing capacity
Prolonged storage (28 days)
Sealing scrubber from the atmosphere increases scrubber endurance
Brief storage (overnight)
Unprotected scrubber performance is equivalent to a sealed scrubber after 28 days storage (implying that overnight unprotected storage is OK)
It is probably fair to summarise as follows .. if you are intent on storing scrubber canisters between dives and are faced with a moderate lay over then storing sealed is a very good idea but you don't have to be worried about overnight storage.
IIRC an interesting observation that was discussed but not part of the presentation was that the baseline test - fresh pack and consume in a single operation - gave approximately 200 minutes duration (under their test criteria). The "28 day sealed" and "overnight" canisters gave slightly over 220 minutes duration. I believe it was Neil Pollock who suggested a hypothesis that not all of the reactions happening in the Sofnolime are occurring at the same rate and therefore introducing a 'dwell' might allow some element of 'recovery'. It is important to note that there was
no repacking of any scrubber canisters. This observation has no practical use to CCR divers which is why it was not directly referenced in the presentation but the graphs do clearly show the different end point duration's.
Simon and others have presented this at a number of venues around the world and I would encourage everyone to attend a session. Not only to hear first hand from the experts but also to support their work. Waiting for snippets to leak out on forums opens the door to recollection errors and opinion bias.