I've had good luck displacing bromides and even (primary)chlorides with azide. Dissolve everything in DMF, add a catalytic amt. of KI, and heat it up to 80° overnight. Works fairly well even with a small excess of azide, but goes nowhere without the KI. In the AM, dilute with plenty of water and extract the hell out of it. DMF tends to stay in the aqueous phase. Never tried it in the presence of a carboxyl though.
The only time I ever made an acyl azide was when I thought I was a hot shot who could do something in one pot. EDC-HOBt coupling of a halo-alcohol to an acid, followed by SN2 with NaN3. Turns out there was a lot of -OBt ester left over which converted to acyl azide. Upon heating it did a Curtius rearr. and I ended up with a bunch of urea decorated on both sides with my starting material, sans one carbon.
As for the acyl azide formation, what's your evidence? As long as you knocked the bromide off, it should be fairly easy to hydrolyze the acyl azide back to an acid. I recall reading once that HN3 and heat will make an acyl azide from an acid, but I'm not finding any evidence to back that up.
In general you want to keep the azide ion as far away from acid as possible, because it will form the volatile, poisonous, and explosive hydrazoic acid. In light of this, surya's advice about esterification is pretty good because if you want a decent extraction of the acidic compound you'd have to acidify your mix.
This goes for everyone: always remember to quench your azide with nitrite or triphenylphosphine before you throw it away. The hazardous waste removal people get pretty pissed when you blow up their truck because you threw out a bunch of azide.
ChemBuddy | 
Chem Reddit | 
Topic: azide preparation (Read 6594 times)