[g-bones]

I am looking for a relatively simple method of forming the molecule shown (a phenyl acyloin).  I was going to use -OH to Sn2 a corresponding alpha bromo ketone (which we have), but fear elimination; especially because it would form an enone (conjugation).  any suggestions/references?

Thank you

[sjb]

Can you add Me^- to phenylglyoxal? I note you mention the S_N2 displacement of the bromoketone - does this mean you have chirality there that you need to preserve?

I think you may be able to form the keto-protected variant of phenylglyoxal semi-directly via benzaldehyde -> dithiane -> formyl addition with DMF and then add in the methyl group as above if you're worried about regiochemistry - there are a number of chiral adducts like ephedrines or TADDOLs / alkylzincs you can use to enhance the stereoselectivity of the carbonyl addition.

S

[kiwi]

Can you add Me^- to phenylglyoxal? I note you mention the S_N2 displacement of the bromoketone - does this mean you have chirality there that you need to preserve?

I think you may be able to form the keto-protected variant of phenylglyoxal semi-directly via benzaldehyde -> dithiane -> formyl addition with DMF and then add in the methyl group as above if you're worried about regiochemistry - there are a number of chiral adducts like ephedrines or TADDOLs / alkylzincs you can use to enhance the stereoselectivity of the carbonyl addition.

S

a couple of other ways to go about this.
(i) hit the available bromoketone with DBU (bromide -> a,b-unsaturated ketone) then Markovnikov hydration of the alkene to the alcohol. this reaction might turn out to be rubbish due to equilibrating elimination of the product back to the alkene, but its quick and easy.
(ii) ketone (or bromoketone?) -> silyl enol ether with base and TMSCl, then epoxidation (mCPBA) and in-situ rearrangement to the TMS-protected form of your product, which will deprotect on workup. this is called the Rubottom oxidation i think, and is a net alpha-hydroxylation of a carbonyl.
(iii) a number of other alpha-hydroxylation approaches are hanging out in the shadows, like radical acetyloxylation followed by hydrolysis. these seem a little dodgy in concept, but might work.
(iv) if you go via sjb's lithated dithiane, you might be able to get it to attack acetaldehyde and save a step.

good luck

[movies]

Rubottom oxidation of propiophenone is probably easiest.  Hydration of the enone won't work because it would go the wrong way.  The dithiane would probably work too, but it's kinda roundabout.