Sciencemadness Discussion Board

Skraup Reaction- Preferred Workup?

UnintentionalChaos - 10-12-2010 at 08:26

So, yesterday I ran the widely disliked Skraup reaction, which is used to prepare quinolines from anilines. The conditions I used are covered in the attached paper under General Procedure for quinolines, method B, except that I ran at 1/10th that scale. It went very smoothly with no hiccups whatsoever. In a way, it was kind of boring- but I appreciate that when 140C sulfuric acid is involved.

What I'm not sure about, however, is the workup. I see numerous different techniques out there. Some go out of their way to destroy excess aniline while others seem to not have any present.

The paper I followed simply makes the reaction mixture neutral and extracts with toluene, strips the solvent, and distills the product.

The preparation referenced here takes a slightly different route: http://orgprepdaily.wordpress.com/2009/11/10/8-bromoquinolin...

They make it strongly basic and extract with ether.

Orgsyn has a prep here: http://www.orgsyn.org/orgsyn/orgsyn/prepcontent.asp?print=1&...

They steam distill the product, diazotize, re-steam distill, and fractionate under vacuum.

Vogel (pages 828-829) advocates a similar procedure to orgsyn, but it utilizes the formation of a quinoline-zinc complex instead of the diazotization.

So, does anyone have experience with which workup (or combination thereof) is the least massive pain in the ass and is most amenable to small reaction size?

Additionally, I have never run a steam distillation where steam is injected into the reaction mix, nor do I have glassware to carry it out (though I could modify one piece if it's necessary). Would simply boiling the crude product in water (with addition of more water as needed work- even if it is slower?

[Edited on 12-10-10 by UnintentionalChaos]

Attachment: Ozonolysis of Quinolines - A Versatile Synthesis of Polyfunctional Pyridines.pdf (179kB)
This file has been downloaded 602 times

Magpie - 10-12-2010 at 09:04

I trust you've seen this?

http://www.sciencemadness.org/talk/viewthread.php?tid=8442#p...

DJF90 - 10-12-2010 at 12:34

Your procedure is very similar to one I have done, the main differences being that I used iodine as the catalyst and 98% H2SO4. In my preparation, p-toluidine was used as the amine. Standard procedure in the lab book I was following was to add acetic anhydride and stir for a couple of hours, then do an acid/base extraction. This supposedly works quite well. However I decided to follow another route, influenced by vogel. Here's what I did...

The apparatus was set up for downwards distillation with a claisen adapter. In the straight neck of the adaptor was an addition funnel containing water. The water was added dropwise to the cooled reaction mixture with magnetic stirring. Once the reaction mixture/tar had been reasonably diluted, the mixture was heated in an oil bath to ~180*C. A hot air gun was used to pre-heat the claisen head to speed up the steam distillation. Water was added as necessary (dropwise) and the steam distillation continued until no more milky distillate was obtained. The steps in vogel were then followed IIRC, and thankfully the zinc chloride-quinoline complex formed with my specific substrate (I dont know if its a general reaction?!). The precipitate was a kind of rosy-peach colour. This was dissolved in excess 5M NaOH, the intermediate precipitate od Zznc hydroxide redisolving. This was then subjected to extraction with ether, separation of the phases, a quick wash with water/brine, and then removal of solvent on the rotavap. The quinoline was obtained as a pale yellow oil, with a pyridine-like odour, and the 2D NMR came out very clean! Yield was a bit on the low side however.

UnintentionalChaos - 10-12-2010 at 12:58

Quote: Originally posted by DJF90  
Your procedure is very similar to one I have done, the main differences being that I used iodine as the catalyst and 98% H2SO4. In my preparation, p-toluidine was used as the amine. Standard procedure in the lab book I was following was to add acetic anhydride and stir for a couple of hours, then do an acid/base extraction. This supposedly works quite well. However I decided to follow another route, influenced by vogel. Here's what I did...

The apparatus was set up for downwards distillation with a claisen adapter. In the straight neck of the adaptor was an addition funnel containing water. The water was added dropwise to the cooled reaction mixture with magnetic stirring. Once the reaction mixture/tar had been reasonably diluted, the mixture was heated in an oil bath to ~180*C. A hot air gun was used to pre-heat the claisen head to speed up the steam distillation. Water was added as necessary (dropwise) and the steam distillation continued until no more milky distillate was obtained. The steps in vogel were then followed IIRC, and thankfully the zinc chloride-quinoline complex formed with my specific substrate (I dont know if its a general reaction?!). The precipitate was a kind of rosy-peach colour. This was dissolved in excess 5M NaOH, the intermediate precipitate od Zznc hydroxide redisolving. This was then subjected to extraction with ether, separation of the phases, a quick wash with water/brine, and then removal of solvent on the rotavap. The quinoline was obtained as a pale yellow oil, with a pyridine-like odour, and the 2D NMR came out very clean! Yield was a bit on the low side however.


Thank you. The iodide in the procedure I followed is rapidly oxidized by the hot H2SO4. It crystallized on the roof of the reaction vessel before a good amount of condensate on the walls started washing it down.

I'm guessing that you added NaOH to the reaction mix before the steam distillation, lest you be trying to steam distill quinolinium bisulfate.

I actually used p-toluidene as well for this preparation. :D The formation of an analagous complex with zinc is not surprising. A methyl on the 6-position carbon should not change much over at the nitrogen. If it were a 2 or 8 position methyl (or larger), sterics might start mucking things up though.

I guess I'll steam distill by the method you mention, use a diazotization to kill off any aniline (I'm not confident that the zinc complex doesn't lose significant product), make alkaline (to keep the phenolate in the aqueous layer), extract, and strip solvent. That should be pure enough for my purposes.

How bad was the yield? (to prepare me for disappointment!)

[Edited on 12-10-10 by UnintentionalChaos]

DJF90 - 10-12-2010 at 14:39

Starting with 750mg p-toluidine, the yield was about 10%. I recall alot of tarry material in the flask, which you may have circumvented by using less concentrated sulfuric acid. The reaction mixture was of course basified before steam distillation, thanks for pointing this out. There seems to be a large loss over the course of the workup - I'd suggest using the minimal amount of manipulations possible, especially if you're working on a small scale in "standard" size glassware. The advised workup in the book is as follows:

Allow the reaction mix to cool and then add ice water (10ml). Transfer the viscous liquid to a 100ml conical flask and cool it in an ice bath. Wash the reaction vessel with ice water (5ml), add the washings to the cold conical flask and make the reaction mixture basic with 5M sodium hydroxide. Add ether (25ml) to the flask and vacuum filter the biphasic mixture with the aid of a celite pad. Wash the conical flask with ether (2x20ml) and pour the organic solution through the filter pad. Separate the biphasic mixture, wash the aqueous layer with ether (20ml) and combine all the organic fractions.
Dry the ethereal solution over magnesium sulfate, filter under suction and concentrate on the rotavap to obtain an oil. Add acetic anhydride (1ml) and leave the remaining solution to react for 30min. Add 1M HCl (20ml) dropwise and stir for a further 10mins. Transfer to a sep funnel and wash the reaction flask with ethyl acetate (20ml)., before adding the solvent to the funnel. Separate the layers, wash the aqueous layer with more ethyl acetate (20ml) and neutralise the aqueous layer with 2M NaOH. Finnaly, extract the aqueous solution with ether (2x20ml) and retain the ethereal solution.

Fill a small sintered funnel with an ethereal silica gel slurry (30ml). Carefully pour the ethereal solution of the product onto the silica and apply suction until the level of liquid is just above the top of the silica. Add more ether (40ml) to the top of the silica and pass it through the filter using suction. Remove the solvent from the filtrate to obtain the product as an oil.

garage chemist - 10-12-2010 at 14:54

I have done this reaction together with a colleague on a substantial scale, yielding well over 250ml of quinoline (unsubstituted).
Making the zinc complex would have wasted a huge amount of ZnCl2, so we did it differently:
After the reaction, the mixture (still highly acidic) was diluted with water and steamdistilled until no more nitrobenzene came over (important!).
Then it was basified and the quinoline steamdistilled, together with small amounts of aniline. This took two full days of work and over 10 L of water had to be distilled.
No steam was blown into the flask, just 1L water added, distilled off, another L added, and so on.
The aqueous phases were again subjected to steamdistillation, the depleted water in the flask discarded, more quinoline-bearing water distilled, and so on, until there was no more water. This way, we elegantly circumvented extraction, which would have required more than 1L of ether.
The crude quinoline was dried and vacuum fractionated over a good column, which got rid of the aniline in the forerun.

DJF90 - 10-12-2010 at 15:27

Very elegant GC! Distillation is definately a good method if the scale is large enough. Steam distillation seems to give a clean quinoline-aniline crude, at least in my experience. Removal of the nitrobenzene in your case, before basifying, is a neat trick.

One thing I did mean to ask UC. Did you notice any H2S or SO2 evolution from the top of your condenser? Both gases are formed in the oxidation of iodide to iodine. How did you deal with it? I didn't bother to check when I did the experiment as I was working in a fume hood at the time...

not_important - 10-12-2010 at 15:52

The couple of times I've done a Skraup I worked up similar to GC's method. Returned the condensed water to the still pot to reduce losses through the slight water solubility of the product. Think we did a distillation just to drive off most of the aromatic amine, then treated the remainder with succinic anhydride followed by extraction with base to remove the amide formed (Still has a free CO2H), finally fractionated the product.

UnintentionalChaos - 11-12-2010 at 18:12

Quote: Originally posted by DJF90  
Very elegant GC! Distillation is definately a good method if the scale is large enough. Steam distillation seems to give a clean quinoline-aniline crude, at least in my experience. Removal of the nitrobenzene in your case, before basifying, is a neat trick.

One thing I did mean to ask UC. Did you notice any H2S or SO2 evolution from the top of your condenser? Both gases are formed in the oxidation of iodide to iodine. How did you deal with it? I didn't bother to check when I did the experiment as I was working in a fume hood at the time...


SO2 was evolved slowly, throughout the 4 hour reaction period. There was no detectable H2S, which 1) probably would not form with the excess of sulfuric acid around and 2) would have been rapidly converted to sulfur in the presence of SO2 and water vapor (A pale film of solid did form on the walls, but I think that was polyacrolein).

As I was working on a 100mmol scale, the amount of SO2 was rather small and it dispersed enough to not bother me. I did not use a condenser. Water was continuously distilled off during the reaction through a 3-way adapter connected directly to a vacuum adapter and a small receiving flask. The 3C ambient temperature did all the condensing I needed with that short bit of glass. The output from the vac adapter was connected to a hose which was placed in a large erlenmeyer in case of catastrophic runaway.