Interesting side note.
Way back when, the USAF research lab in Dayton at Wright Pat found that among the best "field" aluminum bonds resulted from a final scour/abrade using the actual adhesive during the process. Abrade, immediately wipe surface clean (no solvent!) Reapply adhesive, join,clamp, allow to cure.
Theory was that the surface was final cleaned/prepped with no oxygen exposure and therefore the oxides that can cause poor bonding never had the chance to form.
Someday I will probably try to verify he assumption if I live long enough