[wildfyr]

Hey everyone,
I'm confronting a difficult problem and I'm reaching out for help. I am trying to get to sulfonamide (1). Eventually I'd like to try other functional groups on the sulfur side, but tosyl or benzene seems a good place to start



I am starting with 2-Hydroxy-4-nitrobenzaldehyde. Ive done several approaches trying to mildly reduce the nitro without touching the aldehyde, but I think it has been either reduced, not reacted, or oligomerized (unsurprisingly).

The scheme I am attempting now involving a simultaneous diemethyl acetal protection of the aldehyde along with the reduction of the nitro in one step as seen using phoshomolybdic acid in dry methanol, followed by NaBH4 in the same pot in this paper http://www.sciencedirect.com/science/article/pii/0021951789902674 (Joshi et al. Journal of Catalysis 1989). I decided to try to basically "trap" the amine with the tosyl chloride, knowing that the workup would be tough for the intermediate, especially since if I deprotected the aldehyde with acid in the presence of the amine I could get Schiff base oligomers, and that I would be working with a pretty water soluble intermediate in my aminophenol.



I ended up with the acetal protected sulfonate ester with the nitro still intact (rather pure). I only did the reduction for 30 minutes, rotovapped the methanol, added THF and tosyl chloride and letting that stir overnight. Even if I push the nitro to better conversion, I don't feel confident in trying to separate sulfonamide and sulfonate ester without a column (not feasible industrially).

I've tried again, letting the nitro reduction step go for two days and pulled some aliquots, and I can't get much into my organic phase. I'm trying to figure how I can react the tosyl chloride at the amine while leaving the acetal intact so I don't get Schiff base side reaction. Just to be clear, in the end this pretty much has to be a one-pot-ish reaction from start to finish for it to be commercially viable. If anyone sees another, better pathway for this, I'm all ears. Sulfonyl chlorides react with anilines extremely readily in an almost "click" like fashion, so they can handle almost anything else in the pot aside from another very good nucleophile like phenolate. I also would be OK with exploring the sulfonyl being on the benzaldehyde side instead, but I couldn't find a starting material that contributed to that pathway.

[Dan]

OK, so in my opinion, when trying to develop a one pot procedure for multiple reactions, you are better off starting by testing each step separately - isolate the product of each step (using chromatography if necessary), check for and identify any side products because this is vital information. The first priority is making sure that every step works in isolation and that you are not producing side products that interfere with subsequent steps. Eliminating workup and chromatography can come afterwards.

I would actually make the nitro-acetal first, purify it, and then work on the reduction. I'm surprised you can't isolate the aniline, I would have thought a not-too-acidic workup, e.g. ammonium chloride wash or bicarb wash, would not harm the acetal. That said, the amino group para to the acetal will make it much more susceptible to hydrolysis (iminoquinone methide-type mechanism) than in the nitro system. So it is not unreasonable to hypothesise that the product of nitro reduction might undergo polymerisation reactions and give an intractable mixture, this would explain why all the material you recovered in the first run (yield?) still had the nitro intact (because the amine decomposed). If the dimethyl acetal really is too unstable, you could switch to a cyclic acetal. If I was doing this, priority #1 is proving that the amino-acetal is present and stable before trying to trap it. If you can't get a stable amino-acetal, you need to explore other options for masking the aldehyde (e.g. alcohol, acid derivative or vinyl/alkene).

[phth]

I don't think the immine formation is such a problem if you're willing to optimize things. Anilines as catalysts for oxime formation, for example, has been optimized over a range of pH.  I would try a differnt reducing agent like something that can be done under neutral conditions.  There are an enormous amount of reducing conditions for nitro groups because they have been studied for hundreds of years.... e.g. trinitrotoluene

[wildfyr]

I tried several reducing conditions. the most neutral one I found was using Iron NPs in water (sigh, if only that one worked), the rest generally involve some degree of acidic or basic conditions, like the Bechamp reduction, and got back either gunk or starting material.

I'm going to try the bicarb wash on my current reaction and see what I get.

[phth]

You just need a source of electrons!  The overall process is 4 electron redox.  adding 1 electron to the molecule creats the radical anion, which is in resonance with the nitro group; when 1 electron adds to the system to create a radical anion, the structure is equivalent to an enolate to the ring blocking the aldehdye electrophilicity.  The next electron turns the radical anion into an anion which gives its electrons to nitrogen .  If you have access to a Rayonet photochemical reactor, then I can tell you what to do because I've done it before.  Light 365nm 128W, sodium thiosulfate 9-hydrate, and ice temperature...  There is an enormous amount of literature precedent that aldehdyes withstand the reaction: e.g. photochemistry with o-nitrobenzyl groups to form nitrosobenzaldehydes and anilines.  There is no problem at all with schiff base formation at low concentration(!)~10-50 mM. Try widening your search parameters to search for reductions to nitroso/nitrosyl complexes and nitrosyl reductions to amines and hydroxylamine reduction to form amines.  Samarium is another good choice.  You can do the reduction electrochemically, also.

[phth]

Protecting the aldehyde is extremely bad decision in my opinion.

[wildfyr]

This reaction has to be scalable to the hundreds of kilos, photochemical methods are just not going to be feasible. This has to be a cheap approach, so I don't think samarium is gonna fly. And I didn't want to protect it, but I got enough brown goo that I didn't feel I had a choice. Its an interesting idea to retry some of my other conditions at much lower concentrations, but that will increase the price of the reaction a lot due to requiring orders of magnitude more solvent.

I will look for some nitroso/nitrosyl reductions, since generally they can be pushed the rest of the way to aniline with simple methods.

[phth]

It makes sense now that you wanna use a heterogeneous catalyst on a scale like that.  This is a tough problem to get to go nicely on so large of a scale because of the azoxy dye byproducts...   Electron transfer is extremely fast, so reactions can be run at very low concentrations which may even be a problem to do a 100kg reaction on a sub millimolar scale....  some people describe the reduction as a "complex mixture" as the reason for avoiding the use of the o-nitrobenzyl group in the literature but the system is not as complicated because there is no antranil formation etc.  I found that the byproducts can be suppressed by running the reaction in an ice water jacket and using a strong reducing agent like thiosulfave+365 nm hv--> e^-(aq)+thiosulfate dimer work extremely well because they dimerize meaning the rate equation has a squared term.  The paper that made me decide to use these contions is : http://www.sciencedirect.com/science/article/pii/001346869380238U  which yielded a green nitrosyl.

[wildfyr]

Hey guys,
I've made and isolated the dimethylacetal protected 2-hydroxy-5-nitrobenzaldehyde. I've tried reducing it with Na₂S₂O₄ in ethanol/water at 55°C. Within minutes the solution turns bright red. At the end, I get red solid only soluble in water (perhaps sparingly soluble in highly polar organics), and nothing in the organic layer of an EtOAc extraction. NMR shows a little aryl junk, and IR isn't very helpful either. I'm suspicious that I'm basically making a diazo dye of some kind, but I wouldn't expect it to be only water soluble. Any ideas what I'm making, and perhaps another neutral or basic way to reduce nitrobenzene derivs to anilines?

[rolnor]

It really feels like your strategy is very troublesome, I understand that you cannot alter it but it is perhaps the only way, change startingmaterial. I have reduced aromatic nitro to amin with NaBH4/Pd/C in methanol, also iron in 80% acetic acid.

[wildfyr]

I can definitely try the Pd/C route. That actually might work just fine since my first step is MeOH, catalytic phosphomolybdic acid and NaBH4 already. I would have low hopes of the aldehyde not doing Schiff base with the amine under the acetic acid conditions.


I'm happy to change the starting material, do you have a suggestion? I started with a salicaldehyde because making an aldehyde from an alcohol or ester is usually finicky or multi-step production

[rolnor]

Perhaps you can start with 3-aminophenol, put on the sulfon and then make a Duff reaction (Wikipedia)?

[wildfyr]

I poked around a little at this reaction when I was researching a few months ago. It seems like a good idea, but I wonder how well it will work, and how specific it would be. It seems to need pretty activated benzene rings, this substrate isn't one.

While poking around in that wiki article though, I came across the Riemer-Teimann reaction, which may suit my needs perfectly as long as the sulfonamide doesn't hydrolyze with KOH.

[rolnor]

But you have a phenol and also the sulfonamid activates a little? The position you want should be the most activated?

[wildfyr]

I was wondering whether the sulfonamide activates it. The sulfonyl side is definitely electron withdrawing, I'm not sure how I would characterize the nitrogen side