Sciencemadness Discussion Board

a novel phenylmorpholine synthesis aka preludin

jon - 3-10-2021 at 06:22

the method proposes the akabori momotomi reacrtion of n-hydroxyethyl alanine with benzaldehyde to afford n-hydroxy ethyl nor ephedrine
we discovered the best way to make n-hydroxyl ethyl alanine was the sn2 reaction of 2 bromo propionic acid with 2-amino ethanol which can be distliied from oven cleaners.
reductive aminiation of pyruvic acid with 2-amino ethanol was considered but we found an advantage with the former proposal because 2 bromo propionic acid reacts with 2 mols of 2 amino ethanol without multiple amine substitutions because of the adjacent carboxylic acid group precluding multiple alkylations.
the second step is the concerted cyclization of the hydroxy ethyl amino alcohol with h2so4 we found the reaction is more advantageous with cold h2so4 than with room temperature h2so4.

furthermore we found the best way to make the 2 bromopropionic acid is via the formation of the diazonium salt of alanine followed by displacement with hydrobromic acid.

[Edited on 3-10-2021 by jon]

[Edited on 3-10-2021 by jon]

Newton2.0 - 13-10-2021 at 13:02

The method I'm aware of utilizes a-bromo-propiophenone, cyclization with 2-aminoethanol and reduction of the alcohol to the final phenylmorpholine.

I like it, man! Cool idea!

jon - 16-2-2022 at 16:19

no this idea is much easier.
go on youtube and look up 2 bromopropionic acid synthesis.
they make it from alanine, and 37 percent hbr, and catalytic potassium bromide, and sodium nitrate.
then look on, and look up synthesis of alanine via ammonolysis of 2 bromopropionic acid and ammonia, and use that as a template and replace it with ethanolamine or 2 amino ethanol distilled from oven cleaner.

[Edited on 17-2-2022 by jon]

arkoma - 17-2-2022 at 07:31

any experimental?

jon - 17-2-2022 at 07:33

not as of yet but i have most of the materials in hand

jon - 17-2-2022 at 20:38

another idea i had was to take phenyl propanol amine and react it with 2 iodo ethanol adding it drop/portionwise at a rate that keep its stoichiometry at 1/1 to supress the formation of the bis amino ethanolamine by-product this could be optimized through thin layer chromatography.
in other words do a rate study on this reaction the theory is if the iodo ethanol is added at the rate that it reacts with phenyl propanol amine it wont form the side product making this very practicable.

karlos³ - 21-2-2022 at 21:51

Actually, the "best" approach rather turned out to be the reductive amination of 1-phenylpropane-1,2-dione with an equimolar amount of ethanolamine and then borohydride, aluminium amalgam, or whatever else you have, followed by dehydration of that product.
And that first reaction is so far not documented in the literature, as far as I know.
The yield is typically low with ethanolamine, around 40%, with luck maybe up to 50% but never more(on the other hand, isopropanolamine easily gave twice as much yield the first trial, so did isobutanolamine, so it seems to be some sterical reason, with less hindrance being worse for the outcome).
But it is still the second best approach existing.

Of course, there is that old Manske paper, and probably much of the stuff regarding the reductive amination of benzil(e.g. ephenamine synthesis, there is one paper using borohydride out there, but I could not find it, its maybe not digitalised?).
Then again, the reaction differs with ethanolamine opposed to more "normal" alkylamines, especially in the outcome and, like it is known from the similar reaction with the 1-hydroxyketone, does form, if treated wrong, a gel-like impurity preventing the product's HCl salt from crystallizing.
That bit, by the way, can reliably be prevented by isolating the intermediate as oxalate salt instead.
Also, this salt is preferably be used in the cyclization step, as it won't give off HCl fumes("only" carbon monoxide... but very little only).
It is also advisable to isolate the product from that reaction in turn as oxalate as well, experience has shown that this is far less problematic in formation than any other salt.
Especially the hydrochloride which is a real bitch to crystallize, but the typically used fumarate is not that much better, it does not even nearly form as easily and clean as the oxalate.
These bits resulted from several people's experience and their subsequent improvements, I put much emphasis on it, it is pure gold if you want to avoid frustration.

The 1-hydroxyketone can be used equally well obviously, just that it is not as easily accessed as the diketone is, in my opinion.
But for that process, reductive amination and even including cyclization, there exist a few literature references and it works just as well with the diketone.

In conclusion, this is overall the best way to access these things, even though the reductive amination is somewhat on the low side with their yields, but it is much shorter, much more accessible, and less dangerously toxic (like, do you also have the right kind of gloves, which 2-iodoethanol does not eat through? or is it not as bad as 2-chloroethanol?).
It is the best approach for anyone realistically looking to prepare this diethanolamine(the 2-HydroxyEthylPhenylPropanolAmine, abbreviated HEPPA), when considering all actually possible options and how they did turn out in practice.
Via bromoketone substitution, which is somehow the most widely known and used method for some reason, is a notoriously low yielding approach, significantly lower even than this approach here.
It is not realistically viable, like there is this one recent paper from the McLaughlin-people, who used that and got what, a 2% yield or such, out in the end(then again, these people always get horribly low yields).
Similarly, less sterical hindrance, as it is the case with ethanolamine(but not the other two mentioned alkanolamines), seems to be worst for the outcome, and thus usually circumvented by using a secondary amine instead, like N-benzylethanolamine followed by debenzylation.

And for 2-arylmorpholines without the 3-methyl group, its still a whole another story...
Which I have no answer on yet, and then again, it depends on the chosen amine too, as shown above.
But I believe the approach via epoxide ring opening with ethanolamine or another amine to give the diethanolamine compounds, will function better than what I've tried so far... At least thats the only route (and hope) I have left for these ones, everything else so far was a disappointment.
There exist syntheses for fepradinol and halostachine starting from styrene oxide, and I think that can be applied as easily to other oxiranes without too much of the undesired regioisomer.
i.e. if done like they're doing, so ideally free of solvents, and the oxirane is added slowly dropwise into the amine.
Also, steric hindrance does not really matter there, as the yields from both fepradinol and halostachine in comparison do indicate, it should function just as well for ethanolamine.

Monoamine - 1-3-2022 at 20:49

If preludin is what you want as the end-product, then you may just want to look into obtaining a chlorinated phenmetrazine analogue. These are legal and widely available. Then just make a Grignard reagent from the chlorinated phenmetrazine, destroy it with water and you have your preludin.

clearly_not_atara - 2-3-2022 at 19:22

DocX - 17-3-2022 at 14:01

Great stuff. Needs an experimental though, then maybe post in prepublication?

DocX - 17-3-2022 at 14:02

Great stuff. Needs an experimental though, then maybe post in prepublication?

karlos³ - 20-3-2022 at 08:55

Quote: Originally posted by Monoamine  
If preludin is what you want as the end-product, then you may just want to look into obtaining a chlorinated phenmetrazine analogue. These are legal and widely available. Then just make a Grignard reagent from the chlorinated phenmetrazine, destroy it with water and you have your preludin.

If I look into patent US20130203752A1, I can see that a chlorinated phenmetrazine, at least the 3-substituted one I've heard about from a friend, is several times more potent than phenmetrazine.
What most people don't know is that phenmetrazine is not a very strong substance actually.
Who had it even?
I know not many, and only one of those is so old that he had the real preludin and not synthed it himself :P

Quote: Originally posted by clearly_not_atara

Oh yeah, did I not report back "at home" about that one experiment I did on naphthylglyoxal hydrate, when I tried to reductively aminate that with ethanolamine?
It did not work so well, but I haven't used molsieves, that likely was definitely not beneficial.
I got nothing out actually :(

Then again, great news on the other front!
I tried to make fepradinol, as I mentioned above, styrene, bromohydrin formation with NBS, then treated with base to obtain styrene oxide.
And that stuff reacted with 2-amino-2-methylpropan-1-ol(ok, given that amine is sort of a sterical advantage, bulky I mean, or "thicc", to end up at the right position), a little bit water(20%mol, also equimolar amounts of amine and styrene oxide), and I got around 50% of crude freebase, and a around 42% of the purified oxalate salt of fepradinol.

And then, after throwing it in 70% sulfuric acid, etc... 5,5-dimethyl-2-phenylmorpholine in acceptable yields :)
Next, the same, but on 2-naphthyloxirane instead.
And then, on the same substrate, with ethanolamine.

So yeah this is a viable approach for the shorter C2 phenylmorpholines without the 3-methyl.
Neat :)

Quote: Originally posted by DocX  
Great stuff. Needs an experimental though, then maybe post in prepublication?

Well so far, the production of both N-hydroxyethylalanine, and HEGly(N-hydroxyethylglycine) had for three friends not turned out so great.
Even if it appears easy on paper.

I haven't tried that... but I'm currently messing around purifying a batch sarcosine(N-methylglycine, so a direct analogue), and I can understand their frustration and desire to give up, etc :o
Also, the horrible, horrible yields.

I've got an experimental on 3,6-dimethyl-2-phenylmorpholine, 6-methylphenmetrazine, wait....
There it is:


6g of "dione", the abbreviation we should use for 1-phenylpropane-1,2-dione, was reductively aminated with 1eq. of isopropanolamine(1-aminopropane-2-ol), using NaBH4 in ethanol. The yield of the impure dialkanolamine(abbreviation HIPPA, HydroxyIsopropylPhenylPropanolamine) as HCl salt was 8,41g or 84,7%(opposed to 30-40%, rarely up to 50% in case of ethanolamine). This salt was, after crystallisation attempts were half-assed done(it has the tendency to stay sirupy as HCl salt, same as "HEPPA", which is N-2-hydroxyethylnorephedrine, the phenmetrazine precursor) and since it would have ended up in massive losses, it was instead decided to use it as it is, this is not a problem as this is done like this in quite a few morpholine related papers and patents. It(7,7g left after trying to get it to crystallise) was then cyclised by dissolving in 30ml of methanesulfonic acid and left to stand for 24h, and then poured on ice, basified and extracted. The final product as HCl salt was recrystallised then with a little bit of IPA and DIPE, of which more than 4g could be recovered despite the stupid losses. Melting point of the freebase is 64-65°C and of the HCl salt 169-170°C. The obtained product as HCl salt matched that.

However do not attempt to form their HCl salts, straight go for the oxalate.
Also, the oxalate can be dehydrated in sulfuric acid, or whatever else thats suitable(eg like HCl, which won't give off the HCl attached to the precursors salt), directly the morpholine without nasty fumes :)

But I assumed that you meant jon's route and not the already well worked out diketone amination.
The "dione" is of course, phenylpropane-1,2-dione.

karlos³ - 22-4-2022 at 08:04

Alright so I did try, sort of, atara's proposal, but different :D
And it turned out that this(after doing the ammonolysis with different amines, not just that one, sort of sterically hindered-ish) is actually the versatile route for 2C aminoalcohols I was looking for.
Easy to switch the amines without changing conditions and ratio, unlike with styrene oxide...

Which I also have tried to use, by the way, on 2-naphthyloxirane(made via borohydride reduction of the bromoketone followed by basic cyclization) with isobutanolamine, and cyclized the resulting product with 70% sulfuric acid, because thats what you do with these diols.
This is not a drug forum but lets just say the resulting 5,5-dimethyl-2-(2-naphthyl)morpholine has a very interesting activity, it demands further trials so far, but it is already quite promising.
Triple release/reuptake(some mixed form thereof) likely, but it demands further bioassays.

I tried the reaction of styrene oxide with methylamine too, but it is pretty wasteful with the huge excess required and all.

But the route I did(tried it already each on acetophenone and acetonaphthone, successfully) made sure to take the aldehyde hydrate out first...
I found a preparation of 2,2-dimethoxyacetophenone from acetophenone using iodine in MeOH, and then followed that paper here:
Also tried that with 2-acetonaphthone, and it worked fine, except for the low yield, which is probably to 90% my fault and can be improved.
I used methylamine there too, for other reasons, and I got nice and clean products as their HCl salts out in both cases :)
I could have easily used isobutanolamine instead too, thats what I found appealing, but the preparation is, despite being quite quickly done(I actually ran all three mentioned reactions, the first to the arylmorpholine, and the latter two to the aminoalcohol, over the course of ~four afternoons), quite "expensive" right now.

However, due to a sudden iodine and borohydride scarcity, I was again looking into other routes...
Did you guys know that 1-phenylethane-1,2-diol tosylates selectively on 2-? :D
Did you guys know that you can make styrene oxide almost quantitatively from NBS and base?
And hydrolyse it in hot water to said diol in the high ninety percent?
Not to get into the yields of the tosylation as well, though... :D

And I saw a paper that aminated this tosylate, although in sort of a bomb(a closed bottle with a slight excess of ammonia only, I think alcohol and aq. ammonia) with 80% :o
I think the suspicion is justified that this reaction does really not proceed via the epoxide... or is it already confirmed?
I mean it is a suspicion this is an intermediate in the amination of halohydrins, but not confirmed, or am I not up to date?
Anyways, I have so much styrene and I tried it at 0,2mol, lets see how this works out.
Should work even better with other amines, higher boiling amines in special than methylamine, like ethanolamine.

However, for the OP's compound, this won't be suitable method neither, both of them, according to walden et al.

[Edited on 22-4-2022 by karlos³]