Thinking about Friedel-Crafts acylation using carboxylic acid anhydride got me wondering about how exactly aluminium chloride ends up in that reaction.
Generally, despite the fact that Friedel-Crafts alkylation can be carried out using catalytic amounts of aluminium chloride - when chloride of a carboxylic acid is used as an acylating agent in Friedel-Crafts acylation at least 1 equivalent of aluminium chloride has to be used. This is due to the fact that aluminium chloride is complexed by resulting ketone (please see figure below).
Formation of such complex takes aluminium chloride out of action: 1 mole of aluminium chloride gets deactivated for every 1 mole of ketone produced, so it can not activate another molecule of acid chloride. And this is the reason why Friedel-Crafts acylation needs stochiometric amounts of aluminium chloride. (This is often omitted when Friedel-Crafts acylation is being introduced in organic chemistry textbooks, but it is sometimes mentioned later on - especially when it comes to actual lab work and real life conducting Friedel-Crafts acylation.) In the end formation of a ketone-aluminium chloride complex is not a big problem because when chemist who was running the reaction decides to isolate the product, the ketone can be liberated by hydrolysis of that complex using aqueous acid (note that water is unwanted in Friedel-Crafts reaction).
In instance of using carboxylic acid anhydride at least 2 equivalents of aluminium chloride are needed. One surely ends up bounded to the ketone - similarly as described above, but I am not sure what exactly happens with the second equivalent. My guess is that it reacts somehow with carboxylic acid that was produced from anhydride while the other part of anhydride got bounded to aromatic unit.
Hence my questions:
(1) What happens with that second equivalent of aluminium chloride?
(2) How does aluminium chloride react with acetic acid in nonaqueous medium? (My guess would be that this is something like in the figure below. However, I don't know if reaction goes all the way to chelate or not.)