Given that the flux is in direct contact with the solder powder, this allows for the flux activators to interact with the solder powder even while the solder paste sits on the shelf. Those activators can begin to "react" with the powder, and, given enough time, can "clean" the powder surface to the point where the solder particles will actually "weld" together. So, now instead of the paste containing free-flowing powder, it contains clumps of welded together solder particles. Those clumps often increase the viscosity and can clog stencil apertures and dispensing needles. For these reasons, the paste manufacturer will require refrigerated storage of the paste in order to realize the optimum shelf life.
As a rule water-washable solder pastes often include activators that are more aggressive than the activators found in no-clean and RMA type solder pastes. This is because water-washable flux residues are designed to be washed off. So, there is no concern about the flux causing corrosion over the life of the product. On the other hand, a no-clean flux generally has milder activators, because the flux residue may remain on the device indefinitely; where corrosion would be detrimental to the performance and life of the device. As a result, no-clean type solder pastes typically have a longer shelf life and are more tolerant to higher storage temperatures than water-soluble/washable solder pastes.
A solder paste typically has a shelf life of 6 months when refrigerated. One may ask what happens if the paste has been refrigerated for 2 months, then thawed to room temperature, remains at room temperature for 12 hours and is then re-refrigerated....Will it still have a 6 month shelf life? That is a very difficult question to answer. The same situation could arise with a perishable food item that requires refrigeration, such as milk. Lets say that one buys a container of milk at the store and it has an expiration date that is 5 days away. After having it home, properly refrigerated, for 2 days, one of the kids leaves the milk on the counter for 3 hours before anybody notices it and puts it back in the refrigerator. Can one expect the milk to stay good for the remaining 3 days? What about if it is left out of the refrigerator for 1 hour? or 5 hours? You can see how difficult the questions become to answer. What is the impact if a solder paste is exposed to elevated temperatures when it is 3 days old or 3 weeks old or 3 months old or with 3 days left to expiration????? The answer is not fully known. It is impossible for the solder paste manufacturer to study every possible scenario for its impact on the shelf life of the paste.
The best and only sure approach is to refrigerate solder paste immediately upon receipt and only thaw when needed, in amounts that will be completely consumed. Avoid thawing and re-refrigerating pastes as much as possible, in order to take advantage of the full shelf life.
The particle (mesh) size of the solder powder can also impact shelf life. As the powder size decreases, the surface area per volume or mass of powder drastically increases. More powder surface area means more area for the flux to react with, and more surface area for welding to occur. Therefore, a type 3 solder paste that has a shelf life of 6 months may not provide a full 6 months of shelf life with a type 6 solder powder, all other things being equal.
For the most part, solder paste manufacturers are conservative in assigning shelf life. It is highly unlikely that a properly stored solder paste's performance is going to collapse 1 day after the expiration date. In fact, depending on the paste, it may still be good for months beyond the expiration date.
How does one know if their solder paste is still usable? This can be determined rather easily. As mentioned earlier, one artifact of a degrading paste is a rise in viscosity. So one can perform a simple printing or dispensing test to see if it still performs adequately in that regard. Another aspect that often suffers is coalescence. As the flux degrades it loses its ability to adequately remove oxides on the solder powder. In order to gauge the degradation, it is best to put a small amount of paste on a non-wettable substrate, like a piece of ceramic. Reflow the paste and see how well it coalesces. If coalescence is good, the solder paste will reflow into a ball, surrounded by a flux pool that is relatively free of uncoalesced solder particles. If the paste has significantly degraded, the paste will not coalesce well and there will be a significant amount of uncoalesced solder particles in the flux pool.
Please see this IPC test method for determining the coalescent properties of a solder paste.