Hi Shane, 

I would change several moments in your process. First, take 1,05 eq NaNO2 and 5 eq HCl. The excess of NaNO2 produces NO+NO2 that nitrosates your dione. An excess of the acid you do not need either. For electron-rich anilines it is pretty enough. Then, it is always convenient to add the NaNO2 solution from a syringe with a long needle  under the surface of the stirred aniline solution, you avoid this way NO+NO2 losses. During the reaction of the diazonium with the dione pH should not be 1, rather 4-5 (otherwise the anion of the dione will not be present in a sufficient concentration), so I would take 3-4 eq of NaOAc instead.

Then, what is the product you are expecting? Is the benzoyl group the one to be lost? Why do you have a chlorine there, is that supposed to remain in the product? 

Electron-rich anilines are generally poor substrates for diazotation, you get anyway more  tarry products, and the solution is less stable. It is better to run the reaction ot -15 °C. For p-dimethylaminoaniline, naturally, you need an extra eq of HCl.

Thank you so much for your detailed response, Andrei! I'll give those things a shot today when I try this reaction again.

The product I'm hoping for is the hydrozone containing the benzoyl group and the chloride. This reagent will be used in a 1,3-dipolar cyclization with a rhodanine derivative.

I've noticed that NaOAc is not terribly soluble in EtOH, will this be a problem if I were to increase the number of equivalents from 1.6 to 4? Should I add a co-solvent? 

After reviewing the NMR spectra for some separated fractions, I *think* the reaction proceeds to the azo-ester, but just does not finish conversion to the hydrazone, so the pH issue that you brought up makes complete sense.

So you consider that the acetyl will be cleaved  - I am not sure, I would say benzoyl has a better chance to leave. Is it known in literature? If not, I would rather take chlorinated dibenzoylmethane. There are references for amides instead of a ketone in this reaction, here in a review, but I guess you have seen it also:

http://www.arkat-usa.org/get-file/35264/

As for NaOAc, you can add it as a solid, just finely grinded, it plays no role. I would rather avoid too much solvent, it makes the work-up tedious, while the product is not very stable.

I ran many of these reactions, with overall good yields, but I had a CH2 between the carbonyls, and was isolating the hydrazones without cleavage.  Excess HNO2 should definitely be avoided; it can be quenched with sulfamic acid but it's best to simply avoid it in the first place.  My diazotization procedure involved cooling the aniline solution until it began to freeze, then adding 1.1 equiv. solid NaNO2 with gentle stirring.  For the coupling, you might try methanol as a more nucleophilic cleaving agent. Note that you need enough acetate to neutralize all of the acid, so using conc. HCl might be overkill.

See:  H. Yao, P. Resnick, "Azo-hydrazone conversion. I. The Japp-Klingemann reaction."  J. Am. Chem. Soc., 48:3514-3517 (1962) 

http://dx.doi.org/10.1021/ja00877a018

Andrei, in regard to the p-toluidine reaction, I was able to increase my yield from ~9% to 40% using the conditions you gave me. I only allowed it to stir for about 30 minutes before I placed it in the fridge overnight so next time I will try longer stirring at low temp. I had read that partially neutralizing the diazonium solution (with 0.8eq NaOAc) before adding to the dione solution can help avoid drastic pH changes, but this caused the diazonium solution to become very viscous (almost solid), so I may not do this next time either. I tried the dimethylaminoaniline again last night and allowed that to stir overnight. By TLC, it appears to be at the same stage as the previous attempt which I believe to be the azo-ester. However, this time there are only 3 spots on the plate, rather than a mess of products. I had not previously seen the paper you referenced and also got some great information from the references in it. Thank you for 

James, thanks for the additional information! I will be attempting this reaction on a dimethoxyaniline next week and will try bringing the temperature down even lower, as well as methanol. I added a bit of methanol to the p-toluidine reaction because ethanol wasn't keeping things in solution, so I'll try this as the reaction solvent next time. Since you cooled your aniline solution to near-freezing, did you allow the temperature to climb after the addition was complete or did you add it immediately to the dione solution? Thank you for your response and I will definitely take a look at the paper you linked.

Since my yield was still a bit lower than I'd have liked, I saved all fractions after chromatography and I think the additional fractions contain the hydrazone. Can I just add this into methanol with some acetate to complete the conversion? If so, can this be done at room temperature? It seems like the temperature dependence is due to the stability of the diazonium, so if it isn't present, is temperature still an important consideration?

Again, thank you both for your information! I've been having these issues for months and not only has the yield increased, but the reaction is much cleaner and purification was not a nightmare as it usually is.

Shane, 

I'm glad those remarks were useful :)) I would also check whether you have among others a product with an acetyl group instead of benzoyl. 

I did not get about the azo-ester  - you have a 1,3-dione, do you mean that the acyl cleavage is too slow and you obtain the diketone with Ar-N=N- on it? If so, you need, in principle, a stronger acid as a leaving group, either 2-chloro-1,3-diphenylpropanedione-1,3 or even chlorinated benzoyltrifluoroacetone (the starting material would be Aldrich 217042). This means more work, but might improve the outcome dramatically.

As for the diazonium synthesis: with electron-rich anilines which you use the diazotation is very fast even at the freezing point. I think it is better to run the second step immediately as you get a clear diazonium solution. The longer you wait the more your diazonium decomposes.

I am not sure about p-dimethylaminoaniline - this substrate looks problematic, I would rather leave it aside until you establish the process well: it may undergo diazo-coupling itself, so the final mixture will be quite messy. p-Tolyl seems to be the best substrate for the optimisation. Dimethoxyaniline behaves not so nice either  - even the simplest reactions with this diazonium usually give yields about 60%. In these cases I would take an excess of diazonium, 1.2-1.3 eq to the dione.

Maybe you can make an improvement if you isolate the diazonium salt as a tetrafluoroborate - just add to the diazonium solution 1,5 eq NH4BF4, filter the solid off,  wash it with a bit of water, acetone until it's white, and dry in vacuum at rt. This compound - in case of tolyl - is sufficiently stable and can be stored at 0 °C for a long time (for dimethoxy- it is worse, but the hexafluorophosphate must be better). Some info is here: http://www.orgsyn.org/demo.aspx?prep=CV2P0299

So for the second step you can take 1.1 eq of NaOAc (or KOAc - better soluble) and work much more cleanly.

Sorry for the very delayed response, there was a bit of a time-crunch at work so I had to put this project on the side to finish up some stuff.

I had already tried the reaction with p-dimethylaminoaniline before your last response and got a very dark mixture that I set aside for a few days (I assumed it hadn't worked). After purifying it yesterday, it appears that I have isolated the intermediate azo-compound (the acyl group has not been cleaved). It is a very bright red, quite different from the others I've made in this series. I will take a small amount of this and see if I can push the hydrazone formation with acidic or basic conditions. I believe this is the same thing I was seeing before (I incorrectly called it the azo-ester when it is the diketone, as you said). I read in a paper (I believe one of you shared it) that there was some success in converting the azo-diketone to the hydrazone by boiling in alcoholic phosphoric acid (10%). 

I tried this reaction with p-isopropyl, as well, and this gave none of the desired product; instead, the benzoyl-group was cleaved, leaving the undesired acyl. I've ordered 1,3-diphenyl-1,3-propanedione to try this reaction again. The only concern I can think of is that the additional aryl group may slow this reaction due to a bulkier nucleophile. I'll try it with p-toluidine first, to see what kind of yield I get.

If these other things don't work, I'll give the tetrafluoroborate a shot.

Sounds good. A benzoyl is slightly larger than acetyl, but the diketone is more acidic, so you have more of the anion. I think the reaction should be even faster. p-Dimethylamino has a push-pull system around N=N bond, like in Methyl Red, so it is certainly a good dye. Acidic conditions to get the hydrazone are probably better suitable, as a base may attack the product (it resembles acyl chloride, although is surely more stable toward bases).

Just wanted to share an update. I was able to get this reaction to work with o-anisidine and m-anisidine (it was a bit more difficult to isolate the o-anisidine hydrazone). I was also able to isolate the 3,5-dimethoxyaniline, but this reaction seemed to favor formation of the acyl hydrazone rather than the desired phenyl hydrazone. The desired product that DID form was difficult to separate from the undesired acyl product and I ended up doing it by flash chromatography with toluene:EtOAc followed by recrystallization from toluene by addition of hexanes.

Thanks again for all your help!