i've never seen codeine as any salt except the phosphate. it causes some problems, because phosphate is triprotic, so there's actually three codeines bound to each phosphate.
hmm.. if you don't understand the salt thing... drugs themselves aren't normally soluble in water, because molecules need to be polar to dissolve in water. so in order to dissolve the molecule (which is all bristling with non-polar hydrogen-carbon bonds) you have to create a charge on it. this is done simply by reacting it with an acid; the acid donates one of its protons to the drug (usually at a nitrogen atom, for example an amino group) and the drug acquires a positive charge that makes it perfect for bonding with water... the acid (well, actually, the conjugate base of the acid =)), of course, will dissolve in water because it's negatively charged (and will donate its protons to water if it needs to...)
so codeine phosphate is simply codeine reacted with phosphoric acid.
now, this acid-base extraction.. not going into too much detail on the procedure because a) it's really basic and easy b) whoever made your Panadeine doesn't want you to do that.
but this is the theory...(using a nice inert molecule... aniline, which is simply a benzene ring with an amino group)...
first, if you react aniline (which is not soluble in water) with hydrochloric acid, you get anilinium chloride (-ium means positive ion[cation]). (if this was a drug, you'd probably call it "aniline hydrochloride", same thing; and i'll use that term to avoid confusion)
now suppose you were to react this water-soluble salt with a strong base like sodium hydroxide (or, in some cases, a weak base like sodium bicarbonate), that extra proton on the aniline will transfer to the strong base, because it has a much greater affinity for them than aniline does. (and meanwhile the chloride ions from the hydrochloric acid have been turned into a salt too, sodium chloride, because Cl- has a much greater affinity for sodium ions than it does for those positively-charged aniline ions) so now you have sodium chloride and aniline rather than aniline HCl. aniline is not soluble in water, and it will precipitate out of solution (technically it's an oil, but let's pretend it's crystals)
you could filter it, but it's much easier to then add something that aniline *will* dissolve in. say its diethyl ether. you'll get a layer of ether on top of the water, and the aniline will dissolve in that when you shake it. you could then evaporate the ether to yield pure aniline.
if you need the salt again (which you would if you planned on eating it), then it's not so simple. at least in theory, you can add acid to the ether and shake, and you will get the salt precipitating out.. only, since that acid has water in it, the salt will dissolve in the water.. so you have to be very careful, and neutralize any excess acid with sodium bicarbonate in the hopes that you don't add too much again, and neutralize the salt. you could then evaporate that, or better yet, recrystallize the product from it.
sorry for not doing the example with codeine but i think you know why i didn't =P if you want to try that, you need to look into whether codeine base (not phosphate) can be absorbed easily.. it's possible that it will be absorbed intranasally but i can't say for sure.
