Suggestion for a Synthesis of Paracetamol — plausible?

Question

I have seen a number of different plausible syntheses for paracetamol but they suffer from problems such poor regioselectivity of aromatic substitution (usually when nitrating phenol getting a mixture of undesired ortho and desired para). However, is the following not a plausible synthesis?

1. Take aniline (I believe it's commercially available but can be synthesised from mononitration then reduction).
2. React with acetic anhydride to make the amide — still ortho-, para-directing and mildly activating and the directing effects are not acid sensitive as in aniline.
3. Do a Friedel–Crafts acylation with the an acyl chloride.
4. React with hydrogen peroxide and sodium hydroxide to get paracetamol via a Dakin reaction.

Of course the acylation will suffer a little with regioselectivity issues but I feel like it will suffer less as the amide group is larger than a hydroxyl so favours para on steric grounds more strongly and is thus a better solution than nitration of phenol and subsequent reduction.

Nitration of phenol then reduction and reaction with acetic anhydride is the most common hit on google so I assume it's the most common method. But what's wrong with what I have suggested?

Answer 1

Your route sounds possible, yes. (Aniline is commercially available.) There would very likely be side reactions of the amide under Baeyer-Villiger conditions but nothing exceptionally out of control—except you lose a little bit of yield.

There seems to have been a little race concerning the shortest and most environmentally favourable route to paracetamol. To the best of my knowledge, and assuming that their NMR spectra are unaltered and directly from the reaction published, Joncour et al. have the shortest route in 1 (one) step starting from (also commercially available) hydroquinone and utilising ammonium acetate in acetic acid.^[1]

However, for the industry to decide to use a non-established method, that new method must offer a significant advantage over what is presently used. ‘Significant advantage’ typically means much safer or much cheaper. The industrial paracetamol synthesis as employed by Hoechst–Celanese starts off from phenol (extremely inexpesive), does a Friedel–Crafts type acylation with acetic anhydride and $\ce{HF}$ (cheap materials), transforms the ketone into a ketoxime (cheap $\ce{NH2OH}$) and finally uses catalytic acids ($\ce{CF3COOH, SOCl2}$; the latter being rather cheap) to induce a Beckmann rearrangement.

Viewed zealously, this is technically already a more optimised synthesis than the one you propose—it is a step shorter and uses even cheaper reagents. There is probably not much room left for optimisation along these traditional lines.

Referene:

[1]: R. Joncour, N. Duguet, E. Métay, A. Ferreira, M. Lemaire, Green Chem. 2014, 16, 2997. DOI: 10.1039/C4GC00166D. The openly accessable supplementary information with experimental details and the NMR spectrum is available on the same page.

Answer 2

The synthesis proposed would not seem to offer much advantage as the question around selectivity is a matter of conjecture. There are two major industrial syntheses of paracetamol, mainly occurring in India and China. One is the nitration of phenol because it is so cheap. The other is cheaper but has technology barriers to technically inferior producers. This is the Bamberger rearrangement. Starting with nitrobenzene, partial hydrogenation is carried out to phenylhydroxylamine. Being able to do this without full hydrogenation to aniline is the technical differentiator. Acid catalysed rearrangement of phenylhydroxylamine yields para-aminophenol. The hydrogenation and the rearrangement can be done in one step. The amino group then reacts preferentially to form the acetaminophen. Mallinckrodt use this process. This is extremely atom efficient as the most amount of waste is in the final step. On paper, the first two steps have no reagent byproducts.