---

"The Preparation and Properties of Codeinone" J. Am. Chem. Soc. v72, p3247, 1955

The present investigation was undertaken because of the reputed difficulty of preparing codeninone by means of the chronic acid oxidation of codeine.  This difficulty seems to both the sensitiveness of codeinone and to the conditions necessary for its formation and to the throublesome separation from degradation products.  There is recorded, to be sure, a German patent according to which this ketone is obtained ... by the addition of aqueous chromium trioxide to a solution of codeine in dilute acetic acid, but it seems that no ordinary ability can produce the results claimed for this method.  Although codeinone has also been prepared by other means, it was decided that some variation of a chromic acid oxidation would afford convenient access to this ketone.
An observation made during the course of unsuccessful efforts to duplicate the claims of the patent procedure suggested that codeinone resulted from the oxidation of a crystalline polychromate of codeine. *snip*

Experimental

Preliminary Oxidation Experiments -- A mixture of codeine (10g) redistilled cyclohexanone (100ml) dry toulene (300ml) and Raney nickel (20g) was refluxed for 24h.  The catalyst was removed and the base shaken into N hydrochloric acid.  After making ammoniacal the aqueous phase was extracted several times with ether and the base removed.  Three crystalizations from ethyl acetate furnished 3g. (30%) of dihydrocodeinone, mp 193.5-195.5; no depression of melting point after mixture with an authentic sample of dihydrocodeinone.  Its oxime had mp 259-264 dec., (reported 264).
On long standing in sunlight a mixture of codeine, benzophenone and tertiary butanol deposited benzopinacol, but no codeinone could be isolated.
Observations on the Patent Procedure -- According to this method by adding chromium trioxide to a solution of codeine in aqueous acetic acid and rubbing, a chromate salt of codeinone is formed which on appropriate treatment gives 40% of pure codeinone.  Several experiments, in each of which 20 g. of codeine was used, demonstrated that, on this scale of operations at least, this process is unreliable.  Even in the most successful experiments the crude product aggregated less than 40% of the theoretical yeild and contained phenolic impurities which were difficult to remove.
*snip*
Hydroxycodeinone -- The filtrate from the chromate salt contains phenolic by-products and some codeinone which mixture can be recovered by adding alkali and extracting with ether.  In a large scale operation this material was found to contain a small amount of insoluble, high melting base which was separated from the accompanying impurities by its insolubility in ethyl acetate.  From ethyl acetate it furnished a methiodide, mp 249.5 and crystalized from benzene in brown rosettes, mp 269.

---

I had tried to find the reference to the direct chromic oxidation listed here http://hive.dopers.org/14-hydroxycodeinone_and_oxycodone.txt
as "The direct allylic oxidation of codeine to 14-hydroxycodeinone using chromic acid", H.L.Humes et al, J Am Chem Soc, vol 69, 1947, p 1966 but that is not the article at that page.

I guess I'm worried that when this paper "Studies Into the Direct Oxidation of Codeinone to 14-Hydroxycodeinone" on rhodium's site (../rhodium/pdf /codeinone2hydroxycodeinone.pdf) suggests "The majority of reagents gave little or no desired product." for going from codienone to 14-hydroxycodeinone, which seems to be in line with the other ref's for similar reactions, then it may be too good to be true that the direct oxidation copied from the german patent results in a high yeild from one step with chromic acid.  I remember something about german science around the nazi time wasn't so honest too.

I want to believe that when Dope Amine says he has "taken an H-NMR of the compounds at each step of the way and found them to be highly pure with no observable starting material in their spectrums" he was right, but I wonder if a mixture of 14-hydroxy and codienone wouldn't be hard to differentiate using NMR (I know nothing of NMR..)  Proceeding with the hydrogenation would give a product that wouldn't be that obviously not pure oxycodone, if it contained a mixture.  The difference in activity would be more pronouced after demethylation.

Can somebody reassure me?  The dichromate procedure does work and for sure produces 14-hydroxy??  Are there any other references to this?  Or the correct reference for Humes 1947?  I guess the only way to really find out is to do it, but then how would I get an idea of the content of hydroxyc. vs. codienone?