Finkelstein reaction

I went ahead and tried it with KI anyway, since I didn't have any NaI around, and it worked just fine. I did however, use a 3x molar excess of KI. The fact that the end product was EtI and not EtBr could be seen both by its boiling point (when I distilled off the solvent) and because I did a quick AgNO₃ test to check that I truly had an iodinated species, not a brominated one.
And you're right, it didn't take nearly three days. Overnight was enough for a total transformation (unless the EtBr boils off with the acetone as an azeotrope or something...)

The link below shows a picture of the silver nitrate test (left: Iodoethane. Right: Bromoethane)
https://ibb.co/MNmxNL3

Cheers

Probably because you can cook 2-butanone (usually called MEK or methyl ethyl ketone) a little hotter.

I've done it, very common reaction, and definitely use NaI, what rolnor says about potassium salts is important. You want to run it a long time to get very high conversion because theres a good chance you can't separate the different haloalkanes unless they are distillable.

You should use NaI, KBr is more soluble in acetone than NaBr. Usually roomtemp over night is OK, you dont need reflux.

In my hands the reaction is sometimes sluggish with chloroalkanes but somewhat more facile with bromoalkanes.  @OP, How does the insolubility of NaBr play into the overall completeness of the reaction?  The concentration of reagents is usually high, in my experience.  Occasionally this reaction is run in 2-butanone, but I could only speculate as to why.