• People from the hive and the DMT site were not able to reproduce this reaction. If I remember right, only one confirmation appeared shortly after Drones message.

• I found no single reactions of this type in the Beilstein database or through intensive classical literature search,

• nevertheless, drone stated "... it's based on a procedure found in the literature, but honnestly, I cannot divulge my source.".

• Drone stated several times that his reaction resembled reactions from THIKAL, which is wrong.

• Drones posted long literature lists to the DMT site to corroborate the possibility of practical PTC-ionic and SN2 amine dimethylations. Most of these sources either show the opposite or have nothing to do with this special reactions.

• the pKA for the indolic N-H is lower (if I remember right) than for a substituted amine. Indeed the conditions he gave are the standard conditions for indole-1-alkylation and 1-acylation, many examples exist for that in the literature.

• Drone did not see any overalkylation to quaternary ammonium salts, even if he used an excess of methyl iodide ("Actually, its even better to use a larger excess, and no, quaternization is no problem..."). But the non-PTC SN2 alkylation should be an serious competing reaction.

• Drone never answered the doubts posted here and at the DMT site.

A friend of mine, which I trust in chemical things, did this reaction to show that no  DMT will be produced. He analyzed the products by TLC, color reagents, and comparison with non-prohibited standards. The protocols from his experiments are given below.

His results show that the main product is the quaternary N,N,N-trimethyl-tryptamine salt, together with unchanged tryptamine. Only traces of the intermediate N-methyl-tryptamine were visible.

This is in accordance with a very normal SN2 reaction mechanism, which leads to quaternary ammonium salts. This is because the reaction rate increases with each single alkylation due to the electron donating effect of alkyl groups and the low steric hindrance of methyl groups.

I am not totally sure about the reasons for these discrepancies, so I would be interested to hear  the practical experience of other bees (and newbees from the DMT site ) with this reaction and, especially, a statement from Drone.

Lilienthal

Experimental Part
________________________

Reaction

This reaction closely follows Drone's "breath of hope" recipe, scaled down to 1/50. In a test tube with small magnetic stirring bar 128 mg NaOH (finely ground, 40.0 g/mol, 3.2 mmol, 5.2 equ.), 6 mg benzyltriethylammonium chloride (227.8 g/mol, 0.026 mmol, 0.04 equ.), and 3 ml methylene chloride were combined, ultrasonificated in a ultrasonic bath for 15 s, and allowed to stir at room temperature for 15 min. 100 mg of tryptamine base (160.2 g/mol, 0.62 mmol, 1.0 equ.) was added and the solution was allowed to stir for 1 h. 0.1 ml methyl iodide (141.9 g/mol, 2.28 g/ml, 0.228 g, 1.61 mmol, 2.6 equ) was added and the mixture was allowed to stir for 15 h at room temperature.

Work up and thin layer chromatography

0.1 ml of this suspension was added to 0.4 ml methylene chloride and 0.4 ml ammonia solution and the mixture was thoroughly shaken. The clear phases were separated and extracted with 0.4 ml ammonia solution and 0.4 ml methylene chloride, respectively. The organic and the aquous phases were combined and chromatographed on 0.25 mm silica gel plates with UV indicator with 10% ammonia solution in methanol as solvent. The plates were stained by immersion into Dragendorff's reagent and van Urk's reagent and subsequent heating in a stream of hot air. The following standards were used: tryptamine, N-methyl-tryptamine, N,N-dipropyl-tryptamine, and benzyltriethylammonium chloride. 

Color reagents (vol% for liquids)

Van Urk's / Ehrlich's reagent for 2-unsubstituted indoles: 8% HCl conc. and 1% para-dimethylaminobenzaldehyde in methanol.

Dragendorff's reagent for substituted amines (staining intensity: quaternary > tertiary > secondary > primary amines): 12.5% acetic acid, 0.85% BiONO3 (basic bismuth nitrate), and 20% KI in water. Diluted 1 + 12 with 17% acetic acid in water.

Results

The tryptamine base dissolved very slowly in methylene chloride over 45 min. 1 h after the addition of methyliodide the fine suspension became flocculent and light brown coagulated masses settled. Over the next hours these masses were suspended to yield a thick white milky suspension.

The organic phase gave only one UV absorbing spot of Rf 0.52 (tryptamine), which was strongly stained by van Urk's reagent. Additional very faint van Urk positive spots were seen at Rf  0.04 (N,N,N-trimethyl-tryptamine salt?), Rf 0.39 (N-methyl-tryptamine), Rf 0.72 (?), and Rf 0.95 (?). None of these spots were seen if stained with Dragendorff's reagent.

The aqueous phase gave only one UV absorbing spot of Rf 0.04 (N,N,N-trimethyl-tryptamine salt?) which was strongly stained by van Urk's reagent and Dragendorff's reagent. Additionally a faint van Urk positive spot of Rf 0.52 (tryptamine) and a faint but sharp Dragendorff positive band of Rf 0.95 were seen.

Rf values and staining properties

_Rf__UV__Drag.__Urk_________________

0.95__?____+_____-___?, product in aqueous phase, yellow after heating
0.95__?____-____(+)__?, very minor product in aqueous phase
0.95__?____+_____+___N,N-dipropyl-tryptamine standard
0.72__-____-_____+___?, very minor product in organic phase
0.52__+____-_____+___tryptamine standard /  unchanged educt in organic phase
0.43__+___(+)____+___N-methyl-tryptamine standard / very minor product in organic phase
0.08_(+)__++_____-___benzyltriethylammonium chloride standard
0.04__+___++_____+___N,N,N-trimethyl-tryptamine salt?, main product in aqueous phase

- no staining, (+) faint staining, + staining, ++ strong staining

thanks for bringing this one up again and so neatly  !

I have to add one thing: I dreamt the K2CO3 version in DMF and
it did not work. I was also wondering why drone was so reluctant
in answering anything concerned with this but still he was so kind
to give tips on the (alleged) mechanism.

Sorry to see that still after this time no one could really show
it worked. I was able to dream the opposite (failure) and I trust my
dream lab skills.

Let's see what happens here, otherwise I'll be preparing Leuckart-Wallach's
revenge (a post to come, very soon, as soon as I get the college work off
my back). This, at least shows some promise, either from I-3-AcCHO or T as
starting material. We (psyloxy and my humble self) would appreciate if you could
assist with some mechanistic problems we're having when the time comes.

Laters,

- phaidon

...also wanted to add the ethanolamine reaction...I kept seeing posts in DMT world where it was assumed that the quaternary salt (N,N,N-trimethyltryptamonium ion) was being produced...why not react w/ ethanolamine then to produce dimethyltryptamine? From March's organic chem (and one of the refs that Drone left) it appears that ethanolamine can be used to selectively dealkylate quaternary ammonium ions to tertiary amines...

------------------
'Oh, you can't help that' said the Cat: 'we're all mad here. I'm mad. You're mad.' 'How do you know I'm mad?' said Alice. 'You must be,' said the Cat, 'or you wouldn't have come here.'

1.0 molar equivalents tryptamine and only 0.89 molar equivalents of methyl iodide in chloroform yielded a product distribution of 23 mol% quaternary, 3 mol% tertiary, 14 mol% secondary and 59 mol% primary unchanged tryptamine after a fast reaction which had to be cooled.

Jed Zed: Yes, the SN2 alkylation yields intermediate amounts of tertiary amine with ethyl iodide and good amounts with higher alkyls because of steric hindrance. I am pretty sure that a professional chemist would not even think about dimethylation of amines  via a not steric controlled SN2 mechanism.

Teo: I would say you can't be sure if the right alkyl group will be split off using ethanolamine - you have four possible groups. Indolylacetaldehyde is not easy to handle and with tryptamine / formaldehyde you will get about 100% of a beta-carboline. 

Well, let's take this list one point at a time.

• People from the hive and the DMT site were not able to reproduce this reaction. If I remember right, only one confirmation appeared shortly after Drones message.

  True enough. Well, this is what I can tell you: this is agarden-variety variation of the Sn2 reaction -- a well-studied mechanism, of which I hold no patent to. Why I and a handful of others have had success, while others haven't is peculiar, but nothing unheard of.

• I found no single reactions of this type in the Beilstein database or through intensive classical literature search,
  

  I don't want to be unkind, but really, I call into question your Beilstein searching technique. I ran a substructure query on this matter a couple weeks ago using Beilstein Crossfire, and came up with a rather handsome list of aliphatic amine alkylations using alkyl halides and base.

• nevertheless, drone stated "... it's based on a procedure found in the literature, but honnestly, I cannot divulge my source.".

  ...which I finally did divulge a couple weeks ago. I'll dig them out, and send a list to you in the next 24 hours.

• Drone stated several times that his reaction resembled reactions from THIKAL, which is wrong.
  

  ...unless you're reading the eth-lad, allyl-lad, or pro-lad entries. Really, Lilienthal, take a second look: the N-alkylation of norlysergic acid, as performed by Nichols, is a clear example of n-alkylatin of a tryptamine (albeit a cyclized one.)

• Drones posted long literature lists to the DMT site to corroborate the possibility of practical PTC-ionic and SN2 amine dimethylations. Most of these sources either show the opposite or have nothing to do with this special reactions.

  The long list I provided pertained to, among other things, aliphatic amine alkylations using the lower alkyl halides. Please, tell me which sources you are referring to. Let's get some constructive dialogue going.

• the pKA for the indolic N-H is lower (if I remember right) than for a substituted amine. Indeed the conditions he gave are the standard conditions for indole-1-alkylation and 1-acylation, many examples exist for that in the literature.

  1-alkylation is only achieved when the 3-position is either unsubstituted, or is substituted with an electron-widrawing group.

• Drone did not see any overalkylation to quaternary ammonium salts, even if he used an excess of methyl iodide ("Actually, its even better to use a larger excess, and no, quaternization is no problem..."). But the non-PTC SN2 alkylation should be an serious competing reaction.

  No, I didn't see overalkylation. It should be a competing reactino, but I didn't observe any. I chalked it up to steric hindrance.

• Drone never answered the doubts posted here and at the DMT site.
  

  Actually, I did answer it, but apparently you didn't read it.
  

------------------
-the good reverend drone

Ipsa scientia potestas est

• I am talking about methylations. You are permanently talking about alkylations. All your positive examples are about ethyl or higher alkyls. I allready gave the reason why these SN2 reactions are working for higher alkyls, but not for methyl (see above).

• The mechanism you proposed to explain the absence of overalkylation in the PTC process was not an SN2 mechanism, but an ionic one. I found no reference for that mechanism. But I am strongly interested in your source. Maybe you could post it here again?

• PTC assisted 1-alkylation (and 1-acylation) of indoles also work for 3-alkyl-indoles (if you really want I will give the references).

I have to admit, when I first read the title of this, I was a little hurt. But after reading your responses, I realized your intrest is quite sincere, and the questions you raised are thoughtful. So I'm enthused that we can finally seriously discuss this route, this time with a greater audience.

Now that I got that out of the way, let's get back to business.

quote:

---

I am talking about methylations. You are permanently talking about alkylations. All your positive examples are about ethyl or higher alkyls. I allready gave the reason why these SN2 reactions are working for higher alkyls, but not for methyl (see above).

---

Well, yes I am talking about alkylations in general, rather than specifically methylations, but not to its exclusion. For your reading pleasure, I submit a nice little list of successful, high-yielding, well-documented N-methylations, leading to tertiary amines.

• Tetrahedron: Assymetry 9 (1998) 3667-3675:
  dimethylation using CH3I, NaOH, 6 h, THF; 90% yields

  Now this is nice; really nice.

• Journal of Organic Chemistry (1991), 56 4264-4268:
  EtOH, K2CO3, CH3I; ~45%

• Tetrahedron Letters 33, 13 (1992) 1693-1696:
  Now this one isn't dimethylation in the sense I indicated at all, but is still interesting, simply due to the extrordinary yields. Ethyl chloroformate is allowed to react with the starting material, then the product is reduced with LiAlH4. yield: 99%! While I'm not a reaaly big fan of using lots of LiAlH4, those are pretty much the best yields you can hope for.

• Journal of Medicinal Chemistry(?) vol. 13, March 1970, 254-260:
  An article by Joel G. Whitney, W. A. Gregory, J. C. Kauer, J. R. Roland, J. A. Snyder, R. E. Benson, and E. C. Hermann (sorry, I'm not 100% sure this is JOC, which is why I give the extra refrence information)

  Dimethylation is accomplised in a couple different instances, using the amine, (CH3O)2SO2 (dimethyl sulfate), MeOH, and NaHCO3. Reflux; t= 16 h; yields range around from 54-94%, depending on the starting material.

In addition, I have another ref from JMC '95 that alludes specifically to dimethylation of 5-substituted tryptamines (antimigrane compounds) using dimethyl sulfate, though they also used a couple other methods.

Now before I get too side-tracked, I still want to address the first point in your last post some more. The list I just sumbitted shows quite specifically Sn2 methylations yielding quite respectable amounts of tertiary amines, in virtually all cases the desired compound being the lion's share of final product. The fact is, methyls do offer steric hindrance; not a whle helluva lot, but certainly more than protons. Furthermore, you're going to have to convince me that alkyl groups will act more readily as electron donors than the protons they replace; methyl groups are most certainly more electronegative and electron-withdrawing than mere protons.

quote:

---

The mechanism you proposed to explain the absence of overalkylation in the PTC process was not an SN2 mechanism, but an ionic one. I found no reference for that mechanism. But I am strongly interested in your source. Maybe you could post it here again?

---

The mechanism I provided was most certainly a classic variation of the Sn2 mechanism! The PTC facilitates the base actively deprotonating the amine, resulting in making the amine a much nore active nucleophile for the subsequent reaction with the alkyl halide. The deprotonated amine does a very garden-variety backside-attack on the methyliodide, which causes the iodide to leave, resulting in the desired product. The PTC scoops up the resultant iodide to balance itself out, then does an ion exchange with the solid base. Thus the name: phase-transfer catalysis

quote:

---

PTC assisted 1-alkylation (and 1-acylation) of indoles also work for 3-alkyl-indoles (if you really want I will give the references).

---

Yes, but like I said, usually we're talking about an electron-withdrawing group in the 3-position willing to accept an electron pair to facilitate the mechanism. I know: I've intentionally done this reaction a few times before (this was my intro to the indole world, a long time ago.) Still though, your own results, as well as mine, and everyone else's I've seen, all indicate that 1-alkylation doesn't happen under these sort of conditions. Yes, yes; its possible to alkylate the 1-position -- but not under these conditions.

Wow, this is the longest post I've done in a while! I hope you enjoy it. Let's keep this discussion going.

------------------
-the good reverend drone

Ipsa scientia potestas est

Anyone who wants to mess with trying to dialkylate tryptamine with MeI is free to try, but you *will* get lots of quaternary amine.  Tertiary alkyl amines are significantly more nucleophillic than primary and secondary amines, and the dimethyl is not  that sterically hindering when quaternizing.  Methylating a tertiary amine to give the ammonium is like hitting the broad side of a barn.

Look up one of the original Borch references for the NaBH3CN rxn. and go see the aliens.

The reason I'm not entirely satisfied with reductive amination as a means of dimethylating is the Pictet-Spenger side-reaction that one sees between primary tryptamines and aldehydes, not to mention relying on the undoubtedly-soon-to-be-verboten NaBH3CN.

------------------
-the good reverend drone

Ipsa scientia potestas est

is there a thread on the reaction with NaCNBH3 that
you're talking about? Or could you provide some more
info about the procedure / conditions here or in another thread?

Thank you.

- phaidon

• C. Cardellicchio et al., Tet. Asym. 9, 3667 1998: The paper is about methylation of alpha-phenyl-alpha-(2-hydroxy-napht-1-yl)-methylamine to the methoxy-dimethylamino derivative. This amine is highly hindered.

• K. Soai et al., J. Org. Chem. 56, 4264 1991: This reactions are about norephedrine dialkylation. They prepared  11 higher alkyls using but gave no preparation and no yields for the dimethyl analog.

• J. G. Whitney et al., J. Med. Chem. 13, 254 1970: The reaction follows the sequence 1. quaternization of methyl-amantadine with methyl iodide / NaHCO3 / MeOH, reflux 16 h. 2. reflux of the quaternary salt with ethanolamine for 15 min.

• Your passage "Dimethylation is accomplished in a couple different instances..." does not belong to the paper above nor to any of the other papers.

• R. C. Glen et al., J. Med. Chem. 38, 3566 1995 (this is possibly
  your 1995 source since you gave this paper before in the same context) The route is 1. âmino benzylation with benzaldehyde / NaCNBH4. 2. quaternization with methyl iodide / dimethylsulfate / K2CO3. 3. debenzylation with Pd/C / H2. (It also gives the reaction with NaCNBH4 / formaldehyde / methanol / acetic acid, yield 31%)
  

But you are right, your proposed reaction mechanism is indeed a SN2 reaction with an ionic nucleophile.

And the

• list-tag is quite cool,
• [list]
• isn't it?
• (that's a list in a list  )

"50 mg 3-methylindole (skatole, 131.18 g/mol, 0.381 mmol, 1 eq.), 75 mg NaOH (finely ground, 40.0 g/mol, 1.88 mmol, 4.9 eq.), 5 mg triethylbenzylammonium chloride (227.78 g/mol, 0.022 mmol, 0.06 eq.), and 0.025 ml methyliodide (141.94 g/mol, 2.28 g/ml, 0.0541 mg, 0.381 mmol, 1 eq.)
in 2 ml CH2Cl2 were mixed and stirred at room temperature for 16 h. TLC (silica gel, UV indicator, 50% diethylether / 50% n-hexane, van Urk's reagent as color reagent) indicated the appearance of a new, dark violet spot of Rf 0.81. The skatole spot was blue-violet and had an Rf of 0.67. Their proportion as estimated from UV absorbtion as well as from staining was about 40% skatole and 60 % of the new substance, possibly 1-methyl-skatole."

So 1-methylation seems to be the second sidereaction beside of the (fast?) quaternization. There was no complete reaction. Reasons may be a slow reaction rate, the only equimolar use of methyl iodide, or the partial hydrolysis of the methyl iodide.

Sorry about the delay in my response, but work has kept me busy. You seem unusually quick to dismiss things out-of-hand. I don't know why this is, but allow me to retort.

• quote:

  ---

  C. Cardellicchio et al., Tet. Asym. 9, 3667 1998: The paper is about methylation of alpha-phenyl-alpha-(2-hydroxy-napht-1-yl)-methylamine to the methoxy-dimethylamino derivative. This amine is highly hindered

  ---

  The amine is hindered, but not highly. There's still plenty of space for quaternization to occur, yet apparently this isn't a problem.

• quote:

  ---

  K. Soai et al., J. Org. Chem. 56, 4264 1991: This reactions are about norephedrine dialkylation. They prepared 11 higher alkyls using but gave no preparation and no yields for the dimethyl analog...

  ---

  ...no preparation or yields, except the one labelled "General procedure for the Synthesis of N,N-Dialkylnorephedrines". Giving the general details, and basically that it worked. Admittedly, the yields were omitted, but the concentrations, reagents, and reaction times are given.

• quote:

  ---

  J. G. Whitney et al., J. Med. Chem. 13, 254 1970: The reaction follows the sequence 1. quaternization of methyl-amantadine with methyl iodide / NaHCO3 / MeOH, reflux 16 h. 2. reflux of the quaternary salt with ethanolamine for 15 min.

  ---

  Yes, admittedly the quaternary salt is formed here as an intermediate, but treating it with ethanolamine for 15 minutes is hardly a difficult or unreasonable procedure. I think you're just trying to be difficult here.

• quote:

  ---

  Your passage "Dimethylation is accomplished in a couple different instances..." does not belong to the paper above nor to any of the other papers.

  ---

  Yes; that was my mistake. This quote was supposed to correspond to the article R. C. Glen et al., J. Med. Chem. 38, 3566 1995.

• quote:

  ---

  R. C. Glen et al., J. Med. Chem. 38, 3566 1995 (this is possibly
  your 1995 source since you gave this paper before in the same context) The route is 1. amino benzylation with benzaldehyde / NaCNBH4. 2. quaternization with methyl iodide / dimethylsulfate / K2CO3. 3. debenzylation with Pd/C / H2. (It also gives the reaction with NaCNBH4 / formaldehyde / methanol / acetic acid, yield 31%)

  ---

  You and I must be reading a different articles, or looking at different parts of the experimental details. I've left this one at home, so I'll have to respond to this later.
  

Still though, I'd like to also direct your attention to several additional articles:

• Tetrahedron, vol 51, no 1, pp 165-172, 1995.

  This article details successful dimethylations of aliphatic amines using both methyl iodide and dimethyl sulfate, of which the latter is the superior alkylatin agent. 10 equivalents of dimethyl sulfate a little more than 2 equiv. of Na2CO3, EtOH as the solveent, reflux for 48 h. The result is 74% tertriary amine, 18% of the monomethylated byproduct.

• Phosphorous, Sulfur, and Silicon, Vol. 109-110, pp. 237-240, 1996

  Another mention of successful dimethylation of a aliphatic amine using methyl iodide and K2CO3. No details, other than that this was the combinationed used to successfuly alkylate.

• Tetrahedron Letters, Vol 33, No 13, pp. 1693-1696, 1992

  Another article mentioning successful N,N-dimethylation of norephedrine, using methyl iodide, K2CO3, and EtOH

• Indian Journal of Chemistry Vol. 26B, 1987, pp. 1121-1123

  While not an example of methylation, per se, it does detail diethylation of butan-1-ol-2-amine with ethyl bromide under PTC conditions: (Bu4N)+Br- cat., benzene solvent, anhyd. K2CO3