Hydroxymethylation of aromatic ethers

[bottleneck]

Patent US5315018

describes the para-hydroxymethylation of 3,4-methylenedioxy-1-propylbenzene in 59% yield.

In fact, because the reaction is carried out in a carboxylic acid the benzyl alkanoate is formed rather than the free alcohol. However, according to the article the benzyl acetate can be oxidized to benzaldehyde just like the benzyl alcohol, even without deprotection, although no specific working methods of oxidation are listed.

This appears at first glace to be worth looking into as an alternative to chloromethylation, as no gassing is involved and carcinogenic chlorinated methyl ethers are avoided.

The hydroxymethylation of phenols with formaldehde is known, but the ortho-directing effect is very strong, with 2-pentyl-1,4-dihydroxybenzene giving 6-hydroxymethylated compounds, at least when not catalyzed. (see Casiraghi et al. "Selective Reactions Between Phenols and Formaldehyde. A superior Synthesis of Salicyl Alcohols," Synthesis, 124-125 (1980)) ( online and free at

http://www.thieme-connect.com/

)

Does anyone have further information on hydroxymethylation of aromatic ethers?


From the patent:
-------------------

EXAMPLE 1:

Preparation of 4,5-methylenedioxy-2-propylbenzyl acetate

A mixture of dihydrosafrole, 25 g, in acetic acid (25 g) was added, with stirring over 3 hours to a mixture of acetic acid (230 g), paraformaldehyde (45 g) and sulphuric acid (10%, 1.6 g) held at 70°C. After the reaction had been shown to be complete by glc, the excess paraformaldehyde was removed by filtration and the filtrate poured into water (450 ml). Sodium hydroxide solution (20 ml of a 10% solution) was then added. Ether extraction, followed by drying over magnesium sulphate filtration and evaporation of solvent gave an oil. Methanol (30 ml) was added and after cooling to 0°C a colourless precipitate of "dimer" (7) was removed by filtration. Evaporation of the filtrate gave 12.7 g of an oil which was purified by distillation to give 7.3 g of colourless oil as 4,5-methylenedioxy-2-propylbenzyl acetate. (b.p. 176°-185°C at 0.4 mmHg)


EXAMPLE 4

Preparation of 4,5-methylenedioxy-2-propylbenzyl acetate (

Dihydrosafrole (20g), acetic acid (100 g) and water (10 g) were added to a mixture of acetic acid (250 ml), bexane (200 ml), acetic anhydride (40 g) and paraformaldehyde (40 g) to which 1.7 g of 40% sulphuric acid had been added. The reaction mixture was held at 60°C for 5 hours and was then cooled to 20°C and excess paraformaldehyde removed by filtration. Saturated brine (200 ml) was then added followed by ether (100 ml). The mixture was ether extracted (3 times). Follwoing removal of solvent under reduced pressure the material was distilled under vacuum to give 4,5-methylenedioxy-2-propylbenzyl acetate ( as a colourless oil (b.pt 154°-165° C at 2.0 mm Hg)... Yield 59%.

[GC_MS]

I thought that reacting a brominated aromatic compound with NaOMe was one of the easiest methods. There is an example on Rh's page, where 5-halovanillin is reacted with NaOMe to obtain 3,5-dimethoxy-4-hydroxybenzaldehyde. I usually perform alkylation of phenols via DMS though.

[bottleneck]

Hmmm... are you sure this reaction is related to the electrophilic hydroxymethylation of aromatic rings?


Anyway, I found something which might be an alternative to hydroxymethylation. In the aromatic.aldehyde.synthesis.pdf on rhodium.ws is a note on condensing phenols with chloral hydrate.

In

Patent GB445263

("Manufacture of trichloromethyloxyaryl carbinols and of oxy-acids and oxy-aldehydes therefrom"), they get dammned good yields of the benzaldehydes, 70-90%, with the ortho-disubstituted aromatics mentioned such as pyrocatechol.

First the phenol is condensed with chloral or chloral hydrate and a basic catalyst such as sodium phosphate. After a couple of days of standing (try a little chloral to pass the time, maybe?) at about 70 degrees C, the trichloromethylcarbinol is oxidized in situ to the aldehyde by bubbling air through the mixture with a little copper salt as catalyst. Easy peasy, looks like.

If this condensation happens in the remaining para position in toluhydroquinone instead of meta to the methyl group, this does not seem a bad way of getting the benzaldehyde.

Otherwise, perhaps it will also work with aromatic ethers, like 2,5-dimethoxytoluene. If it did, then certainly it would go in the para position to the methyl.


From the patent:
-------------------------


Example 4

124.1 parts of molten guaiacol are stirred with 2 parts of anhydrous sodium guaiacolate.  165.4 parts of molten chloral hydrate are then cautiously added whilst stirring. The whole is stirred for 5 days at 70-75°C, whereupon the reaction mixture is strongly diluted with water, 200 parts of prepared chalk are added and the whole is boiled in a reflux apparatus for 48 hours whilst air is passed through it.  The escaping air is passed through a solution of sodium bisulphite.  From the cooled reaction mixture and from the bisulphite solution the vanillin which has been formed is isolated in the known manner by extraction with ether.  The yield is about 80-90 per cent of the theoretical.
  The duration of the boiling with chalk can be reduced to 24 hours by the addition of 5 parts of copper sulphate

[bottleneck]

I can't believe no one is interested in this chloral condensation stuff!

It beats all other formylation methods I have seen on both mildness, simplicity and non-toxicity, and even yield. Plus, chloral isn't exactly difficult to get a hold of.

Maybe you guys just don't realize how nasty Vilsmeier is and how elaborate chloromethylation is.

[Antoncho]

Dear Bottleneck!

What you have found is indeed awesome. Double awesome, in fact.

I think your mistake was not using the proper markup for the patents - it is just so much easier to click on a link rather than doing all that elaborate work  of finding a patent on Espacenet

And, generally, it is considered bon ton on the Hive to supply some examples in addition to patent links.


Well, i have read both patents and i must say they offer a real breakthrough! I have also ventured to dig up all of the patents mentioned in them and they also have some real interesting tidbits.

The first one: it is the 1st patent ever to describe methylolation of a fully etherified phenol. The yield ain't all that great, but the procedure is EASY and also non-toxic. Which is makes it markedly different from chloromethylations. Together with azomethinic formylation (as in

Post 382264

(Antoncho: "Zealot: azomethinic (Haack) formylation", Novel Discourse)) this is currently one of the so few kitchen-chem methods that can bee used directly on phenol ethers.

Apart from that ... since Haack formylation also gives yields in 50-60% range, the method you've found is very valuable!  A little more involved in terms of duration, but doesn't require gassing w/HCl and pre-preparation of reagents.


The second method does require pre-preparation of reagents (namely, chloral, which can bee made by simply chlorinating EtOH - of course, working w/chlorine in the kitchen is YUCK YUCK YUCK - but nonetheless quite realistic, much more bearable than with bromine ). And it works only on phenols. And it prefers being unhydrous.

BUT. It's also very, very nice. It is the only selective hydroxymethylation process i've seen thus far.

It's esp. interesting in relation to piperonal.

-OH group isn't that deactivating at all, which means it would bee easy to methylolate catechol, methylenate w/DCM in DMF (with GOOD yields) and then oxidize to aldehyde.

It would definitely bee SWIM's method of choice if he ever decided to make piperonal from scratch. Preferrable to vanillinic methods, i dare say.





Well, in addition to that i advise everyone interested to take a look at

Patent GB344675

which describes in detail how to oxidize BzOH's to BA's with copper salts. The would also bee definitely THE method of choice in kitchen conditions.




Bottleneck! Could you, maybee, correct the patent links in your above posts and supply the most relevant examples from the patents you've found so that they can bee rated as excellent and eventually make it to Rh's site.




Sincerely Yours,


Antoncho

[bottleneck]

Thanks Antoncho.


> I think your mistake was not using the proper markup for the patents - it is just so much easier to click on a link rather than doing all that elaborate work of finding a patent on Espacenet
> And, generally, it is considered bon ton on the Hive to supply some examples in addition to patent links.

Maybe you're right. I'll try to be a bit neater in the future.


> The first one: it is the 1st patent ever to describe methylolation of a fully etherified phenol. The yield ain't all that great, but the procedure is EASY and also non-toxic. Which is makes it markedly different from chloromethylations.

I am a bit worried about some oddities in the patent though... Check out the boiling points.


> Apart from that ... since Haack formylation also gives yields in 50-60% range, the method you've found is very valuable!  A little more involved in terms of duration, but doesn't require gassing w/HCl and pre-preparation of reagents.

This method you mention sounds very nice! It also seems to be mentioned in the

https://www.thevespiary.org/rhodium/Rhodium/pdf/aromatic.aldehyde.synthesis.pdf

(mentions nitrosobenzene), but there is only reference to phenols and anilines. It is great to hear it also worths with the ethers!


> The second method does require pre-preparation of reagents (namely, chloral, which can bee made by simply chlorinating EtOH - of course, working w/chlorine in the kitchen is YUCK YUCK YUCK - but nonetheless quite realistic, much more bearable than with bromine). And it works only on phenols.

Are you sure about only working with phenols? In

Patent US2426417

, it is claimed, as far as I can tell, that alkoxybenzenes are condensed with chloral in the presence of AlCl3 or H2SO4 or other Lewis acid catalyst, but gives no references.


> Bottleneck! Could you, maybee, correct the patent links in your above posts and supply the most relevant examples from the patents you've found so that they can bee rated as excellent and eventually make it to Rh's site.

Thanks, but I am not sure if they are excellent or not, with the oddities in the first patent, and as regards the chloral condensation, I still haven't seen any specific examples of 1,2,5-substituted substrates reacting nor am I sure of the product with hydroquinone, so I am not sure of the usefulness.



Most of these are references from the article

https://www.thevespiary.org/rhodium/Rhodium/pdf/aromatic.aldehyde.synthesis.pdf

Condensation with formaldehyde:

Patent DE103578

Patent DE105103

Patent DE105798

Patent DE118567

Patent GB161679

Patent GB399723

Condensation with chloral:

Patent FR791818

Patent DE475918

Patent DE598652

Patent GB219676

Patent GB344675

Patent GB445263

Patent GB453482

Miscellaneous:

Patent GB1032879

Another lewis acid catalyzed chloral condensation, this time to create a,a-methoxyphenylacetic acids. Weird.

[Vitus_Verdegast]

Are you sure about only working with phenols? In Patent US2426417, it is claimed, as far as I can tell, that alkoxybenzenes are condensed with chloral in the presence of AlCl3 or H2SO4 or other Lewis acid catalyst, but gives no references.

I've read about the chloral condensation in a Chem.Rev. paper on the Reimer-Tiemann reaction (60 (1960) p.169-184), the authors claim that this is a variation of the latter. It would be great if it appeared to work fine on phenol ethers, but I don't have such a good feeling about it.

Migrdichian et al., Org. Syntheses p 1336, 1957 use dichromate to oxidize the trichlorocarbinol.

Still, this is really great, considering that the overall yield of vanillin in the patent (80-90%), it sure kicks Reimer-Tiemann's butt.

Chloral hydrate is frequently used in microscopy (to ppt. cellulose from plant matter I believe), I wonder if there isn't any denatured form available.
  

[bottleneck]

> I've read about the chloral condensation in a Chem.Rev. paper on the Reimer-Tiemann reaction (60 (1960) p.169-184), the authors claim that this is a variation of the latter.

> It would be great if it appeared to work fine on phenol ethers, but I don't have such a good feeling about it.

Why is that? The yield is already higher than with formaldehyde, no? Is there some reason you know of it should it not work on ethers?

> Chloral hydrate is frequently used in microscopy (to ppt. cellulose from plant matter I believe), I wonder if there isn't any denatured form available.

There are supposedly low yields (30%?) to be had by oxidizing trichloroethylene in bleach.

Patent US2759978

, but the procedure at least avoids gaseous chlorine.

Otherwise, with proper glassware, chlorination of ethanol is probably preferable, though the patents mentioned in Merck Index don't exactly seem to be brimming with info on yields.

Patent DE133021

Patent DE734723

[Vitus_Verdegast]

Why is that? The yield is already higher than with formaldehyde, no? Is there some reason you know of it should it not work on ethers?

The Reimer-Tiemann works only on phenols. If this is a variation on it then there is a good chance that it also only works on phenols.

This is an aldol condensation between a phenol and chloral, isn't it?

[bottleneck]

> This is an aldol condensation between a phenol and chloral, isn't it?

I think so.

> The Reimer-Tiemann works only on phenols. If this is a variation on it then there is a good chance that it also only works on phenols.

But how would one explain the benzyl acetate product in of piperonylpropylbenzene in

Patent US5315018

? Or even the condensation in

Patent US2426417

between "phenyl ethers" (phenol ethers I guess)  and chloral? Okay, they claim a Lewis acid catalyst necessary, but this sure doesn't look like a Friedel-Crafts reaction.

[Vitus_Verdegast]

from

Patent US2426417

Alternatively these carbinols may be prepared by the reaction of chloroform with an appropriate ether derivative of a hydroxybenzaldehyde in presence of an alkaline catalyst.

Strange, this is the first time I hear of a Reimer-Tiemann being performed on phenol ethers. It doesn't give any reaction details either.

[bottleneck]

> "Alternatively these carbinols may be prepared by the reaction of chloroform with an appropriate ether derivative of a hydroxybenzaldehyde in presence of an alkaline catalyst."

> Strange, this is the first time I hear of a Reimer-Tiemann being performed on phenol ethers. It doesn't give any reaction details either.

It isn't Reimer-Tiemann. The chloroform condenses with the aldehyde moeity to form the trichloromethylcarbinol.

But as you can see here in

Patent GB942057

, aromatic ethers condense with aldehydes, including formaldehyde just like the normal Bakelit-reaction. Surely this could mean they will also work for most of the other aldol condensations?

And as you can also see, in

Patent US5315018

it is argued that because formaldehyde can condense with phenol it should also condense with aromatic ethers.

[Vitus_Verdegast]

Yes you are right about this,

one thing, if you condense an aromatic ether with chloral you will most likely end up with a mixture of both o- and p- isomers, isn't it?

[bottleneck]

One would think so. Except, the para-product seems to be the main product usually.

But lots of unknowns

1,2,3-Trimethoxybenzene might be a pretty good substrate to test this on as it is both very activated and completely unhindered, sterically, in the para-position.

It would be nice if hydroquinone worked and the trichloromethylcarbinol moiety turned out to be deactivating enough to yield predominantly 2,5-dihydroxyphenyltrichloromethylcarbinol.

[Vitus_Verdegast]

I'll see if I can find some interesting things in the library about chloral condensations

[bottleneck]

> I'll see if I can find some interesting things in the library about chloral condensations

It would be great if you found something. I am sure there are hundreds of articles with useful information, because DDT was made by condensation with chloral, only difference beig the product is of couse a diphenylmethane so they add a lot of acid to catalyze the formation of the bicycle.

Actually, come to think of it, the fact that DDT was made like this (from chlorobenzene) kind of gives a, well, let's just call it "a very strong hint" that you don't even need ethers on there to condense aromatics with aldehydes

[Vitus_Verdegast]

from Ind. Chim. Belge 1954(1), p.120-134 :


Benzene and chloral, in presence of AlCl₃, give different products in different reaction conditions.
To obtain the phenyl trichloromethyl carbinol, 200 gr of anhydrous chloral is dissolved in 1L dry benzene. To this stirring mixture there is added, over a 3 hour period, 40 gr of anhydrous AlCl₃ and the mixture is stirred for 12 hours. Ice-water is then added, the organic layer is separated and dried over CaCl₂. The solvent is evaporated and the residue is distilled in vacuo.

starting material  -  product  -  bp.  -  mp.  -  yield

benzene  -  C₆H₅CHOHCCl₃  -  145°/154mm  -  37°  -  70%
toluene  -  p-CH₃C₆H₄CHOHCCl₃  -  154-56°/13.5mm  -  63-64°  -  57% (53%)
p-xylene  -  p-(CH₃)₂C₆H₃CHOHCCl₃  -  163-64°/14mm  -  61-61.5°  -  55%
o-xylene  -  1,2-(CH₃)₂-4-C₆H₃CHOHCCl₃  -  183-89°/20mm  -  73-74°  -  56% (53%)
anisol  -  p-(CH₃O)C₆H₄CHOHCCl₃  -  184-86°/16mm  -  55-56°  -  58%
ethylbenzene  -  p-(CH₃CH₂)C₆H₄CHOHCCl₃  -  175°/25mm 
dimethylaminobenzene  -  p-N(CH₃)₂C₆H₄CHOHCCl₃  -  109°
chlorobenzene  -  p-Cl-C₆H₄CHOHCCl₃  -  159-162°/14mm
bromobenzene  -  p-Br-C₆H₄CHOHCCl₃  -  173-74°/12-14mm

(For the last 4 entries no melting point or yield was given.)

When sulfuric acid is used as the Friedel-Crafts catalyst instead of AlCl₃ the initially formed phenyl trichloromethyl carbinol reacts with another mole of benzene to yield chiefly beta,beta,beta-trichloro-alpha,alpha-diphenylmethane.


An equimolar mixture of a phenol, chloral and anhydrous K₂CO₃ yields, after stirring for a variable amount of time, a trichloromethyl carbinol :

starting phenol  -  rxn time (days)  -  obtained product  -  yield  -  mp.

phenol  -  42  -  p-HO-C₆H₄CHOHCCl₃  -  60-70%  -  87°
p-cresol  -  56  -  HO-CH₃-C₆H₃CHOHCCl₃  -  50%  -  147-48° (orientation unknown)
guaiacol  -  21  -  3-MeO-4-HO-C₆H₃CHOHCCl₃  -  72%  -  118-19°
pyrocatechol  -  60(*)  -  2,3 or 4,5-(OH)₂-C₆H₃CHOHCCl₃  -  64-77%  -  128-29°
resorcin  -  3(*)  -  1,3-(OH)₂-C₆H₃CHOHCCl₃  -  50%  -  176°

(*) The phenol and chloral are heated until complete dissolution, followed by the addition of anhydrous K₂CO₃


If 4 mol of 10% aq. K₂CO₃ solution is heated with 1 mol of phenyl trichloromethyl carbinol at 120° in the presence of trimethylamine, benzaldehyde is the only product that is obtained, without the amine it yields a mixture of benzaldehyde and mandelic acid.

[Vitus_Verdegast]

From Bull. Soc. Chim. Fr. (1952) p.932-934 :


3,4-dimethoxy-alpha-hydroxy-beta,beta,beta-trichloroethyl benzene

A mixture of 147 gr (1 mol) of anhydrous chloral, 138 gr (1 mol) of veratrole and 200 gr petroleum ether were placed in a roundbottom flask, placed in an ice-bath and equipped with a mechanical stirrer. Anhydrous AlCl₃, 30 gr, was added in small portions and the mixture is left stirring at RT for 12 hours. Water is added and the organic phase is decanted from the solid mass. The residue is recrystallised from benzene, yield 200 gr, mp. 144°

When this reaction was performed with sulfuric acid as the catalyst, it was impossible to stop it at the hydroxyalcoylation stage.