Lysergic acid synthesis (user submission)

[Rhodium]

I got this in the mail...

- - - - -

A very amateur student of chemistry, I came up with a possible idea for a total lysergic acid synthesis. I've no idea whether it would really work and I haven't the means to try, so submit my doodling to the Chemistry Archive.  Perhaps some more experienced chemist will see it and work out whether there's anything good here or not.

The central idea behind this synthesis is that substituted naphthols, and 2-naphthol in particular, will react with dienophiles to form Diels-Alder adducts.  I remember stumbling across a definite instance of this reaction in the literature, many years ago; as I remember, 2-naphthol was reacted with 1,2-ethylenedisulfonic acid (or an ester of that acid.)
The 1,2-ethylenedisulfonic acid here acts a synthon for acetylene--that's an important aspect of this synthesis.  There may be other synthons for acetylene which will work (I vaguely remember that diethyl acetylenedicarboxylate is another such).

The presence of the nitro group in the second step, the reductive amination, is problematic.  An alternative, more time-consuming procedure would be to reduce it to an amino group and then protect it by acetylation or phthalimidation, before carrying out the reductive amination.

The third step is a leap of logic.  (Or of faith.) The hope is that, in the reducing, acidic workup from the ozonolysis, the nitro group will be reduced to an amino group (and then condense with the 4-formyl group to form an indole), and that simultaneously, a Knoevenagel-type condensation of the 1-formyl group closes the tetrahydropyridine ring.

I hope you pardon the very non-standard names:

1.  5-nitro-2-naphthol + acetylene (in the form of a synthon) = 1,2,4,5-tetrahydro-5-nitro-1,4-vinylene-2-naphthalenone (A)
2.  (A) + ethyl 3-methylaminopropionate + reducing conditions = 2-(N-methyl-N-(3-carbethoxypropyl)amino)-1,2,4,5-tetrahydro-5-nitro-1,4-vinylenenaphthalene (B)
3.  (B) + ozone, followed by zinc/acetic acid = [5-amino-2-(N-methyl-N-(3-carbethoxypropyl)amino)-1,4-diformyl-1,2,4,5-tetrahydronapthalene, intermediate] = ethyl isolysergate (C)
4.  (C) + aqueous acid = ethyl lysergate.

http://rhodium.lycaeum.org

[jim]

This is the most creative synthesis I have ever seen on this site.

I can't see anything wrong with it.

I would be very wary about the ozonolysis however and the nitro-group. 

I would also be wary of double bond migration after or during the reductive amination.  I would do this part in the dark at low tempatures just to be sure, especially with that NO2 lurking about...

[oxycodfish]

Interesting synth.  It reminds me of some of my early ideas, although, I freely admit, this seems far more likely to work.  I'd worry primarily about the O3 step -- this may in fact be the easiest way to do it, but I imagine yields (of the actual desired sort, mind) will be low and slow.  Just my opinion.  Something about the reductive amination step looks fishy to me, but I'm no expert.

Looks good -- now I have some bedtime reading material.  Thanks to whomever contributed this synth idea.

[oxycodfish]

Hmm, I didn't even get signed offline and this occurred to me -- what are the chances of the final ring closure occurring as described?  This doesn't look like something that will occur readily, if at all.  But then, it's too late for me to think at 100%, and my molecular model program is broken.  This is more of a question for my own knowledge than a criticism of the synth.

[SarahTonin]

I have to admit this is extremely creative--much better than anything I have ever dreamt up. As for the potential double bond migration, I don't think this would be a problem, as it would lead to a Bredt olefin, which is highly unlikely to form.

As for the ozonolysis step, this should work well provided the proper steps are taken to avoid oxidation following generation of the molozonide. Zn/H+ is the common way to go about reductive cleavage of molozonides. I would watch out for O2 lingering about--probably want to do this under an inert atmosphere.

I can't speak for sure on the reductive amination, as I simply don't know if it'll work. I do know thhat the Knovenagel condensation (cyclization step) should work like  a dream provided one uses an appropriate base. The problem is the availability of the tandem of bases most commonly used--piperidine and pyridine. I've done some work with this reaction in the past condensing malonates with aldehydes and the yields were always very good (>90%).

Someone give this a go and let us know if it works.

[jim]

The more and more I think  about this synthesis the more misgivings I have about the second product.

It seems energically favored for the double bond to migrate towards the ketone, thus creating an enamine.  Doesn't it?

[Rhodium]

The product after the reductive amination? To create an enamine, the double bond would have to travel over two carbons, which I find unlikely unless very basic conditions are used, or transition metal salts are present. Neither is used in this reaction, so that should not present a problem.

http://rhodium.lycaeum.org

[jim]

The double bond would not have to travel over two double bonds.  Look at the second line of the synthesis.  All the double bond has to do is migrate towards the top of the page. 

This probably would happen when it was a ketone still, thus forming an alpha, beta unsaturated ketone.  The nucleophile, the amine, would then not add to the ketone, but rather add across the double bond. 

Does that sound right?

[Lilienthal]

The following is a reaction sequence from H. Plieninger and his co-workers. He tried to synthesize the ergot alkaloid Chanoclavine I and used a very similar approach to build up the benz[cd]indole ring system.

Look at that funny tautomerisms of the keto / aldehyde systems   .

The references are:

H. Plieninger, W. Lehnert, D. Mangold; Chem. Ber. 100, 2421 (1967)
H. Plieninger, W. Lehnert; Chem. Ber. 100, 2427 (1967)
H. Plieninger, W. Lehnert, D. Mangold, D. Schmalz, A. Völkl, J. Westphal; Tet. Lett. 22, 1827 (1975)
H. Plieninger, A. Völkl; Chem. Ber. 109, 2121 (1976)
H. Plieninger, D. Schmalz, J. Westphal, A. Völkl; Chem. Ber. 109, 2126 (1976)
H. Plieninger, D. Schmalz; Chem. Ber. 109, 2140 (1976)

Synthesis of 2-hydroxy-5-nitro-2-naphtaline:
C. D. Hurd, O. E. Fancher, W. A. Bonner; J. Org. Chem. 12, 369 (1974)
P. Friedländer, S. Szymanski, Chem. Ber. 25, 2076 (1892)

[PolytheneSam]

I found this US patent on lysergic acid synthesis looking though 564/67:  2796419

http://www.geocities.com/dritte123/PSPF.html

http://www.privacy.org/article.php?sid=620&mode=thread&order=0&thold=0

http://www.privacy.org

[Lilienthal]

That's the Kornfeld synthesis, the first total synthesis of lysergic acid. There are many improved (regarding to steps and yields) total syntheses of lysergic acid. Maybe I will compile a short overview over all published syntheses (later...)   .

[smiley_boy]

I have to say this scheme looks a bit more than a little sketchy. There's no regioselective control of the first step, which would make stereoselectivity during the reductive amination (of which there is none) very difficult.

The ozonolysis looks reasonable, and I'm even willing to suspend disbelief for the Knoevenagel condensation step. Still, the selective reduction of the nitro group in the presence of those unprotected aldehydes? That I'd need convincing for.

The thing though that sticks out most for me is the last step; that double bond migration. We're talking about a seriously uphill battle here. By moving it from the 8,9- to the 9,10 position, a huge amount of ring-strain would be induced. Thermodynamically, I can't see this happening.

Its creative, I'll give it that. Somehow though, it almost reminds me of something spat out by a computer. I'm probobly being a bit harsh, since I'm sure this is more of a rough, rough, rough draft than anything. Its a nifty idea, and I'd bet a fine jumping-off point for some even more clever ideas.