Friedel-Crafts formylation?

[bottleneck]

Hello.

Why doesn't Friedel-Crafts acylation work with formic acid? There are plenty of patents describing making acetophenones by acetic acid and propiophenones by propionic acid, but none that I can find using formic acid to make benzaldehydes.

Can anyone explain why formic acid doesn't work?

[Rhodium]

I don't know why it doesn't work, but it can be worked around by using "synthetic equivalents" to formic acid, such as dichloromethyl methyl ether, which together with a lewis acis such as SnCl4 or TiCl4 will formylate aromatic compounds.

[Aurelius]

formic acid decomposes too easily

[Rhodium]

That is well known, but how and why would it do that with a weak lewis acid?

[lugh]

According to Calloway's article on the Friedel-Crafts Synthesis in Organic Reactions, formyl groups are introduced by formic acid, hydrogen cyanide and hydrogen chloride (the Gatterman reaction), and carbon monoxide and hydrogen chloride (the Gatterman-Koch reaction). See Ind Eng Chem 26 1317 (1934) for an article by Groggins about using formic acid directly  

[bottleneck]

Rhodium:

>I don't know why it doesn't work, but it can be worked around by using "synthetic equivalents" to formic acid, such as dichloromethyl methyl ether, which together with a lewis acis such as SnCl4 or TiCl4 will formylate aromatic compounds.

Okay. But does that mean it still works like a Friedel-Crafts reaction? Can all FC catalysts be used, including such as FeCl3?


Aurelius:

>formic acid decomposes too easily

What does it decompose into? If it decomposes into carbon monoxide, wouldn't it work as a type of Gattermann-Koch reaction which would just as well result in the benzaldehyde?


Lugh:

>According to Calloway's article on the Friedel-Crafts Synthesis in Organic Reactions, formyl groups are introduced by formic acid, hydrogen cyanide and hydrogen chloride (the Gatterman reaction),

So formic acid works fine in that reaction? What is the role of the HCN in the reaction? Do you think some other substance could be used in its place?

>and carbon monoxide and hydrogen chloride (the Gatterman-Koch reaction).

Yes, I have been reading a little in some patents using this reaction. The materials are so available that it is appealing at first glance. CO can be made by heating CaCO3 and zinc to 700-750 deg C or by dehydrating formic acid with sulfuric acid.

The addition of cupric chloride apparently allows the reaction to take place at atmospheric pressure (because the CuCl helps the CO dissolve in water??), but in the industry, the reaction is apparently usually carried out under increased pressure, which I don't know how one deals with in a "kitchen lab" type of arrangement available to most people.

>See Ind Eng Chem 26 1317 (1934) for an article by Groggins about using formic acid directly

I will try to get the library to get it, but can you tell a little more about the conditions under which the reaction takes place? Do they use a catalyst?

Thank you.

[lugh]

I will try to get the library to get it, but can you tell a little more about the conditions under which the reaction takes place? Do they use a catalyst?

The name of the author of that article was Goggins, there may be other articles/patents of interest to you. No further information is available presently  

[Rhodium]

It should behave like an ordinary FC reaction, but I have never seen it being used with other lewis acids than TiCl4 and SnCl4, perhaps it is because those milder agents gives beter yields than FeCl3/AlCl3 etc.

[bottleneck]

Lugh:

>The name of the author of that article was Goggins, there may be other articles/patents of interest to you. No further information is available presently

Okay, thanks. I couldn't find references to his name in the patent database, but I will definately try and get the article you mentioned.


Rhodium:

>It should behave like an ordinary FC reaction, but I have never seen it being used with other lewis acids than TiCl4 and SnCl4, perhaps it is because those milder agents gives beter yields than FeCl3/AlCl3 etc.

Okay. What's your opinion on the feasibility of successfully using zeolite-like catalysts for Friedel-Crafts acylation in a very basic lab? I was impressed by a reported 90% conversion of anisole to p-methoxyacetophenone with acetyl chloride in 50 seconds by some minerals known commercially as "Envirocats".

With the number of potential Friedel-Crafts catalysts apparently available, I am surprised there haven't been reports of anyone repeating Assholiums synthesis of 2CH using nitrosation of 2,5-DMacetophenone. That seems to have the potential of being a "golden bullet"-synthesis of 2CH requiring only OTC chemials or reagents easily synthesizable from OTC chemicals.


Thanks.

[Rhodium]

It is very easy to make 2,5-dimethoxyacetophenone and nitrosating it (I have very many references for this), but the only reduction I have seen for this is catalytic hydrogenation under pressure.

Catalytic transfer hydrogenation at atmospheric pressure using ammonium formate as a hydrogen donor works both for aryl alkyl ketones, and for oximes, but I have never seen an example where both such a ketone is reduced together with an oxime on the same molecule.

If Assholium did not give any more details on the reduction, we are either stuck with hydrogenation under pressure, or someone has to try the unknown reaction of catalytically hydrogenate alpha-oximino-2,5-dimethoxyacetophenone at ambient pressure (which probably would work, but noone has tried that one yet).

[bottleneck]

>Catalytic transfer hydrogenation at atmospheric pressure using ammonium formate as a hydrogen donor works both for aryl alkyl ketones, and for oximes, but I have never seen an example where both such a ketone is reduced together with an oxime on the same molecule.

Now I am far from being a real chemist, but can't you just reduce the groups with the normal reducing agents, maybe in a larger amount? Does the presence of the two groups close together "deactivate" them to reduction?

>If Assholium did not give any more details on the reduction, we are either stuck with hydrogenation under pressure, or someone has to try the unknown reaction of catalytically hydrogenate alpha-oximino-2,5-dimethoxyacetophenone at ambient pressure (which probably would work, but noone has tried that one yet).

Since 1,4-dimethoxybenzene almost makes itself of benzene and methanol, it seems desirable that someone would try to perfect the synthesis at some point.


Thanks.

[Rhodium]

Of the "normal" reducing agents which are capable of reducing aryl alkyl ketones to hydrocarbons, I don't know any that also reduces oximes smoothly.

And those which reduce oximes smoothly will only reduce the ketone to the alcohol, and we know from the ephedrine experience that benzylic alcohols are not very easy to reduce to hydrocarbons (both the RP/I2 and the birch reduction would destroy the molecule).

[bottleneck]

>Of the "normal" reducing agents which are capable of reducing aryl alkyl ketones to hydrocarbons, I don't know any that also reduces oximes smoothly.

>And those which reduce oximes smoothly will only reduce the ketone to the alcohol, and we know from the ephedrine experience that benzylic alcohols are not very easy to reduce to hydrocarbons (both the RP/I2 and the birch reduction would destroy the molecule).

Ah, okay. I didn't know it was so difficult to reduce the groups. What about LAH? I thought it removed everything.

If an alcohol is the result, what about the suggestion on your site about dehydrating with KHSO4? If a doublebond formed where the -OH had been, would it be difficult to reduce?


Thanks.

[Rhodium]

LAH will reduce some benzylic ketones to hydrocarbons, but I have no idea about when it does and when it doesn't.

If we dehydrated the Hydroxy 2C-H, a so-called enamine would be formed, and it would rearrange to 2,5-dimethoxyphenylacetaldimine, which would need to be reduced once again to form the 2C-H. It would not be too hard, but I believe that that imine would be unstable under the conditions required for the alcohol dehydration. If it was that easy, then people would use that method to make meth from (pseudo)ephedrine.

[bottleneck]

>LAH will reduce some benzylic ketones to hydrocarbons, but I have no idea about when it does and when it doesn't.

Okay, thanks. I guess it's difficult to predict in advance?

I am interested in electrolytic reduction, as the apparatus required seems potentially available to everyone. There are of course a number of journal articles describing the reduction of nitrostyrenes to phenethylamines. Three abstracts from Russ. J. Electrochem. Vol. 32, No. 1, 1996  (which I can't seem to get at the library) suggest that it maybe could also work in reducing oximes to amines.

Do you know of any case of ketones having been reduced to hydrocarbons electrolytically?

>If we dehydrated the Hydroxy 2C-H, a so-called enamine would be formed, and it would rearrange to 2,5-dimethoxyphenylacetaldimine, which would need to be reduced once again to form the 2C-H. It would not be too hard, but I believe that that imine would be unstable under the conditions required for the alcohol dehydration.

Thanks for the explanation.

>If it was that easy, then people would use that method to make meth from (pseudo)ephedrine.

I guess that's probably true!


I have another question, if you know. Could you oxidize styrene in a similar way to isosafrole, to give the phenyl-2-ethanone? Could the ethanone then be reductively aminated to phenethylamine?


Thanks.

[Rhodium]

Could you oxidize styrene in a similar way to isosafrole, to give the phenyl-2-ethanone? Could the ethanone then be reductively aminated to phenethylamine?

Phenyl-2-ethanone would be phenylacetaldehyde, and it would aminate to phenethylamine. I guess the reaction conditions has to be pretty mild for the oxidation (like the buffered performic, or even better one of the epoxidation methods) as phenylacetaldehyde is prone to polymerize.

Another interesting thing to do with phenylacetaldehyde imines instead of reducing them is to react them with CH3MgI to form amphetamine or methamphetamine:

https://www.thevespiary.org/rhodium/Rhodium/chemistry/meth.phenylacetaldehyde.html

[bottleneck]

>Phenyl-2-ethanone would be phenylacetaldehyde, and it would aminate to phenethylamine.

Thank you, great! There must be many efficient routes to styrene by now, so maybe something along those lines has a future?

> I guess the reaction conditions has to be pretty mild for the oxidation (like the buffered performic, or even better one of the epoxidation methods) as phenylacetaldehyde is prone to polymerize.

Okay. Otherwise, could one maybe try the direct oxidative amination proposed by Drone #342 in "Photoamination of Methylstyrenes" for an even simpler route?

>Another interesting thing to do with phenylacetaldehyde imines instead of reducing them is to react them with CH3MgI to form amphetamine or methamphetamine:

Isn't that a Grignard reaction? I thought it was very hard to get it to work in practise, because the need for dryness is more than modest equipment can provide?

Thanks for the link to the article by the way!

[Rhodium]

Yes, it is a grignard reaction. Some grignards can be very hard to start, but there are ways of making it work even in damp ether, just use an ultrasonic bath (there is a ref on my page for that). Grignards can be made in the kitchen, if the ether is dry.

[bottleneck]

>Yes, it is a grignard reaction. Some grignards can be very hard to start, but there are ways of making it work even in damp ether, just use an ultrasonic bath (there is a ref on my page for that). Grignards can be made in the kitchen, if the ether is dry.

That's fabulous!

Do you think one could use a Grignard reaction to make phenethylamine from 2-chloroethylamine and a halobenzene?


Thanks.

[psyloxy]

Nope. The amine is too acidic for the R-MgX. But there are some less reactive
organometallics that could do it. Zn I think ... I'll look it up.

--psyloxy--

[Rhodium]

React a halobenzene grignard with ethylene oxide to get the phenethyl alcohol. This can either be oxidized to the phenylacetaldehyde and be reductively aminated to a PEA, or you could make the bromide (with PBr3) or the tosylate (with tosyl chloride) and react that with sodium azide followed by NaBH4 (the latter two PTC catalyzed) to get the PEA.

The general procedure for bromide/tosylate swap to the azide to the amine can be found at:

https://www.thevespiary.org/rhodium/Rhodium/chemistry/mda.azide.html

[bottleneck]

Psyloxy:

>Nope. The amine is too acidic for the R-MgX. But there are some less reactive organometallics that could do it. Zn I think ... I'll look it up.

Thank you.

Would it be best to make the reagent of 2-chloroethylamine or of the halobenzene?

Would it be better to use a chlorobenzene or a bromobenzene?

Rhodium:

>React a halobenzene grignard with ethylene oxide to get the phenethyl alcohol. This can either be oxidized to the phenylacetaldehyde and be reductively aminated to a PEA,

Great! I guess you can do a lot of interesting things with these types of reagents. I also like the ultrasonic bath thing very much.


Thank you.

[Rhodium]

Bromobenzenes are way better than chlorobenzenes when it comes to formation of grignard reagents.

[bottleneck]

Okay, thank you.

Are there other ways to substitute an aromatic halogen than these organometallic reactions?


Thank you.

[Rhodium]

What do you want to substitute the halogen with?

[bottleneck]

Anything that could eventually be extended to aminoethane.

I thought the halogen on the benzene would be a nice "target" for other reagents, but I can't find a lot of reactions where the halogen is substituted, except Grignard reactions, making phenols, making sulfonic acids, etc.

I am looking for a way in which one could use 1-chloro-2,5-dimethoxybenzene as a precursor.


Thanks.

[psyloxy]

Had a look in that book on organometallics again and, well, there's a lot left to
explore and no really hard facts.

It might work like this:

Br-Me-CN + Zn --> BrZn-Me-CN

BrZn-Me-CN + CuCN*2LiCl --> BrZn[Cu(CN)-Me-CN]

BrZn[Cu(CN)-Me-CN] + (MeO)₂Ph-Br --> (MeO)₂Ph-Me-CN

but you better forget about this and try to find some useful info on
higher order cuprates and organo-zinc reagents on the net yourself.
Sorry for the disappointment.

--psyloxy--

[bottleneck]

It's okay.

Thank you very much.