Sciencemadness Discussion Board - Forest of amines! (saccharin to cyclic and asymmetric amines)

Forest of amines! (saccharin to cyclic and asymmetric amines)

Chordate - 25-3-2012 at 14:38

Hallo chaps, it's been a bit, and I'd like to share a little something I'm currently working on.

Amines!

I read in a paper on alternative gabriel and mitsunobu substrates awhile back that referred to the use of saccharin as an alternative reagent in the Gabriel amine synthesis. It indicated that the sulfamide cleavage step can be performed under different conditions than the hydrolysis of the amide side of the molecule.

That paper is:

http://pubs.acs.org/doi/abs/10.1021/ar00010a001

This naturally offers the possibility of substituting as in the normal gabriel, hydrolyzing the amide, then substituting a second time. So my question is, can I get di-alkylated amines, primary amines, or even cyclic amines starting from saccharin? It is a thoroughly OTC product.

Unfortunately the paper referred in that reference, the pharmacological society of japan paper seems to be lost to me, or at least the open access form of it I found was only the table of contents. So I decided to play, and see if I could recreate the work at least in part and figure out yields and the parameter space.

So far: here is what i have done,

Saccharin is present in sweet'n low as a potassium or sodium salt, the acid form is easily filtered from an aqueous mixture , though it is a sticky fluffy powder that gets EVERYWHERE. Acidify with a little dilute hcl and filter/rinse, further purification is not necessary because the next step involves beating it with a lot of base.

So far for the first round of alkylations I have tried forming benzyl saccharin from benzyl bromide and allyl saccharin from allyl bromide. I mixed molar equivalents of the saccharin to 1.2 molar equivalents of dry potassium carbonate, wetted it down with 2.1 molar equivalents of allyl or benzyl bromide, and set it up on a thermowell. The benzyl bromide being high boiling enough I used a thermometer adapter and reflux condenser so I could keep the temps steady, but the allyl bromide is low boiling enough I just set it up for reflux (use good ventilation, or you will tear gas yourself) and let it sit.

At first, I was incredulous, because I was looking at some 12 grams of slightly damp white powder in each of these flasks, but at around 100 degrees the benzyl bromide/saccharin mix started bubbling casually. I kept the mix at 130-160 and the reaction gradually, over the course of a few hours, turned into an orange oil with some white material at the bottom, presumably residual carbonate. At six hours the flask had turned almost completely into orange syrup, cajolled by occasional swirling to keep the reactants well mixed. Characterization is pending, as I am slowly stripping off the residual allyl bromide under vacuum.

Gentle overhead stirring starting at the point where the mix begins to liquefy would have been great for this, the goo is quite thick and begins as a powder so magnetic stirring would be right out.

Unfortunately the allyl pot did nothing. It sat as white powder for six hours and gradually took on the creamy sheen of oxidized material. I worked it up by driving off the residual bromide and got nothing but burnt saccharin and potassium carbonate. The observations above give a hint as to why this is: the reaction appears to require higher temperatures than the 70 some odd degree allyl reflux temperatures, even with such an active nucleophile. I am making some high boiling n-butyl tosylate and will re-attempt to alkylate in the next few days.

The next few weeks will be me trying to break this reaction, I've some 1,4 dibromobutane I will attempt to make pyrrolidine from, bromocyclopentane to see if I can get it to react with a secondary substrate, and I'm going to attempt to perform all the required cleavages and alkylations in one pot and see if I can't get symmetrical and asymmetric amines from this thing, starting from the benzyl saccharin above.

Any feedback?

[Edited on 25-3-2012 by Chordate]

[Edited on 25-3-2012 by Chordate]

Reaction scheme and bad picture

Chordate - 25-3-2012 at 15:55

Most of the pictures came out terribly. I need a lab camera instead of a camera phone, but here is the reaction scheme I'm working with, and a picture of the benzyl saccharin mixture at roughly 4 hours, so you can see the transition to the orange oil.

Scheme (note, I erroneously though I had propanol but didn't so the propyl substitution is actually butyl):

conditions are:
A) Reflux or Heat with appropriate alkylating agent to yield 2.
B) NaOH reflux to 3
C) R-X, then reflux NaOH(aq) to 4
D) HCl(aq), reflux to amines

Syrup:

[Edited on 26-3-2012 by Chordate]

bbartlog - 25-3-2012 at 16:31

Weird, it's like you're using the saccharine as a protecting group for the nitrogen.

Nicodem - 26-3-2012 at 09:49

Thanks for sharing your experience.
If you give the exact reference for the missing article, I'm sure someone will take the time to retrieve it.

Quote: Originally posted by Chordate

Unfortunately the allyl pot did nothing. It sat as white powder for six hours and gradually took on the creamy sheen of oxidized material. I worked it up by driving off the residual bromide and got nothing but burnt saccharin and potassium carbonate. The observations above give a hint as to why this is: the reaction appears to require higher temperatures than the 70 some odd degree allyl reflux temperatures, even with such an active nucleophile. I am making some high boiling n-butyl tosylate and will re-attempt to alkylate in the next few days.

You are wrong about the saccharinate being such an active nucleophile. It is actually relatively weak. You should try the reaction of allyl bromide with sodium saccharinate (or 0.5 ekv. K2CO3 and saccharin) in a minimum amount of DMF or NMP. Perhaps then it would work at lower temperatures. You can get an estimate of nucleophilicity by considering the pKa of the nucleophile provided it is a hard nucleophile (N-nucleophiles commonly are such). Saccharin has a pKa of about 2 and its hardness is somewhat lower than that of amines (due to the neighboring S and pi-bonds), but must still be higher than the P,S,Cl-nucleophiles (whose nucleophilicity don't correlate to pKa any more). I would thus consider it similarly nucleophilic as the carboxylates are.

Chordate - 26-3-2012 at 21:42

Quote: Originally posted by Nicodem

Gah, I misspoke. I was actually referring to the fairly active Electrophile, allyl bromide. You are quite correct that the saccharinate is not a great nucleophile.

Good idea on using DMF. I must confess that part of my motivation for the project was Sauron whining about people looking for OTC routes to diethylamine. DMF is not quite OTC. Neither is tosyl chloride, really, but diethyl sulfate can be made from OTC materials and the tosylates and dialkyl sulfates have somewhat similar reactivity and high boiling points. I may also have a go at sticking it in a high pressure vessel and wailing on it overnight.

As for the reference, its not a matter of access, the journal is openaccess now the ref is: Sugasawa, S.; Abe, K. J. Pharm. Soc Jpn. 1952 issue 72 p270. The index for these issues is here:
http://ci.nii.ac.jp/vol_issue/nels/AN00284903/ISS0000309280_...

but they don't have the fulltext.

Nicodem - 27-3-2012 at 07:08

Quote: Originally posted by Chordate

As for the reference, its not a matter of access, the journal is openaccess now the ref is: Sugasawa, S.; Abe, K. J. Pharm. Soc Jpn. 1952 issue 72 p270. The index for these issues is here:
http://ci.nii.ac.jp/vol_issue/nels/AN00284903/ISS0000309280_...

but they don't have the fulltext.

Perhaps you just did not look at the right place. Bellow are the links to the whole series of their articles. Most of them are in Japanese, though at least the Japanese chemists for some reason use English chemical nomenclature in the Latin alphabet.

A New Method for the Preparation of Secondary Amines. I
A New Method for the Preparation of Secondary Amines. II
A New Method for the Preparation of Secondary Amines. III
A New Method for the Preparation of Secondary Amines. IV
A New Method for the Preparation of Secondary Amines. V
A New Method for the Preparation of Secondary Amines. VI
A New Method for the Preparation of Secondary Amines. VII
A New Method for the Preparation of Secondary Amines. VIII

There are also several papers for the alkylation of saccharine with alkyl, benzyl and allyl halides or sulfonates in DMF or other such solvents. In most cases a temperature of about 100 °C is used:

J. Am. Chem. Soc. 1939, 61, 15–16. DOI: 10.1021/ja01870a004

J. Am. Chem. Soc. 1954, 76, 302–303. DOI: 10.1021/ja01630a099 ("In our investigations dimethylformamide has been found to be an excellent solvent for the reaction of alkyl halides with sodium saccharin.")

J. Med. Chem. 1992, 35, 4613–4627. DOI: 10.1021/jm00102a016

Eur. J. Org. Chem. 2006, 4483–4489. DOI: 10.1002/ejoc.200600508

Pay attention that you can also use these N-alkyl-saccharine intermediates for other reactions as well. Just like the phthalimides and other imides in general, they are also very susceptible to nucleophilic attack at the carbonyl group and the consequent ring opening. This same property is the one that allows the second alkylation to give the N,N-dialkylsulfonamides you plan to do, but it can also be used to prepare a variety of other products.

Chordate - 30-3-2012 at 16:58

Thanks again Nicodem. Your literature search and suggestions helped out.

So, I did a quick test of Nicodem's above suggestion of adding DMF by taking the pot of saccharin I had attempted to allylate last week and given up on, and throwing 2 ml of DMF into the mix and heating.

Before addition of 2ml DMF it looked like this, mind you this had previously sat under reflux for a little over six hours:

On addition of 2ml of the DMF and heating the mix immediately showed signs of reacting, becoming slurry like and dashes of orange. Surprisingly this occurred at temps well below the 100+ at which the benzyl began reacting.

Here the pot is at T+ 30 minutes, temps have climbed to 70 degrees and the mixture is bubbling.

And at the end of six hours, it looked much like the benzyl pot did.

While letting that run, I finished working up the benzyl saccharin. one of Nicodem's above references suggested steam distillation to remove the excess bromide. So I tried it. Yay, complex apparatus:

A surprising amount of the benzyl bromide came over, this may indicate the need for longer reaction times, stirring, or some other optimization.

At some point, the orange glop starting coming over with the steam and I cut the distillation, though the pot still reeked of benzyl bromide. Then I remembered the steam distillation suggestion was from a paper was published in the 1950s, and they maybe didn't have a vacuum like I did. So I decanted the water, losing a wee bit of solid in the process, but also removing most of the potassium carbonate, and finished stripping bromide under vacuum

This gave me this nice oil, which spontaneously crystallized on standing.

The orange color, it turns out, had nothing to do with the product itself. Recrystallization from methanol (about 125 ml) gave pale orange leaflet crystals, and a second recrystallization gave 3.3 grams of faintly pink material which appears to be rather pure benzyl saccharin. 28% overall yield based on the original saccharin.

This yield should in no way be indicative of what the reaction is capable of, as none of it has been optimized or played with and I was flying blind until nicodem pointed me in the right direction.

[Edited on 31-3-2012 by Chordate]

Nicodem - 31-3-2012 at 06:37

Have you taken the mp of your product? It could as well be unreacted saccharin.
Keep in mind that N-alkylsaccharines relatively easily hydrolyze to the ring-opened carboxylates which are likely water soluble. You used 1.2 molar equivalents of K2CO3 and you have unreacted excess of benzyl bromide which means the post reaction solution is basic enough for the ring-opening to occur (especially upon the steam distillation conditions). This cyclic imides hydrolysis is generally reversible and N-benzylsaccharine should be regenerated by acidification (at least such is the case with N-alkylphthalimides and related imides). Your low yield of the isolated product can therefore also be the consequence of an improper isolation protocol.
For further experiments, I suggest you to use sodium saccharin if you have any. Else, use a 2 : 1 ratio of saccharin vs. K2CO3 (remember the pKa values!). For the work up, removing the excess of the alkyl halide the way you did it is OK, but before you decant the aqueous phase, try to make sure you have an acidic pH there, so acidify first.
You can also do an experiment to see if the ring-opening is a problem at all. Take some presumed N-benzylsaccharine and see if it dissolves in 1M NaOH upon heating. If you get a solution, try regenerating/precipitating the imide by acidification. If such is the case, you can then adapt your isolation protocol to reflect these properties (you can selectively remove BnBr and saccharin by such tricks).

Chordate - 31-3-2012 at 08:02

I have not yet taken the melting point, but I see no unreacted saccharin in the isolate: The alkyl region is clean and there's no lump at 12ppm, which is why I cut them off. Just a little DMSO, H2O, methanol.

<img src="http://i.imgur.com/ecsMJ.jpg" width="800" />

I am a bit spoiled at the moment.

I don't have any sodium saccharin, but that doesn't mean I can't make any, still I am going to forge on with hydrolysis and a second alkylation.

Edit: Gah, I switched E and G on that molecule,

[Edited on 31-3-2012 by Chordate]

[<a href="u2u.php?action=send&username=bfesser">bfesser</a>: reduced image size(s)]

[Edited on 12.8.13 by bfesser]

Nicodem - 31-3-2012 at 08:16

That's an impressive purity, worthy of the two recrystallizations. Have you weighted the crude product and the product after the first recrystallization? Any in-between TLC monitoring of the purity? The reason I ask is because I'm still wondering if the yield is due to low conversion, or a ring-opening reaction during the work up, or the recrystallization losses.

Chordate - 31-3-2012 at 08:50

Quote: Originally posted by Nicodem

That's an impressive purity, worthy of the two recrystallizations. Have you weighted the crude product and the product after the first recrystallization? Any in-between TLC monitoring of the purity? The reason I ask is because I'm still wondering if the yield is due to low conversion, or a ring-opening reaction during the work up, or the recrystallization losses.

Yeah, actually. I got 8.75 grams of the crude product prior to recrystallization. TLC gave Four spots, two light ones that matched the starting materials, a heavy product peak, and a light fourth spot, which may well be the hydrolysis product or whatever the orange goo was. As I mentioned above I was surprised at how much unreacted benzyl bromide was coming over with the steam,

It occurs to me that if I am quite worried about the PH it would not be too hard to remove the unreacted potassium salt from the reaction by washing before heating it and distilling off the excess halide, then one could acidify before distillation and still keep PHs reasonable. This might also be advantageous because it doesn't look like the reaction particularly wants to run to completion under these conditions.

At any rate, I didn't worry too much about yields for the first run. It's more a matter of figuring out the parameter space for later optimization, and for that I thank your suggestions for small experiments to get an idea of what is going on.

[Edited on 31-3-2012 by Chordate]

jon - 31-3-2012 at 10:47

have you thought of throwing a base in the mix to deprotonate that nitrogen? maybe kalium carbonate or stronger.
i know when i have a hard time n-alylating this sometimes helps.

Nicodem - 12-8-2013 at 08:41

An example at Synthetic Pages: N-alkylation of sodium saccharine with Methyl chloroacetate; 1,2-benzisothiazole 1,1-dioxide (DOI: 10.1039/SP295).

Quote:

The anhydrous conditions are extremely important to get best yields.

Their lead reference was Synthetic Communications, 26, 585-591.

karlos³ - 12-11-2019 at 11:38

Sorry for necroposting, but I have to say I just made some phenacyl saccharin from bromoacetophenone and sodium saccharin in DMF, and the DMF really is the key reagent.
The yield(still drying) seems to be almost quantitative(I guess) and that just after an hour of heating.
First, the sodium saccharin(pellets...) went into solution, then later on the solution got milky with precipitate(NaBr I suspect) and then after cooling I added 5x of the reaction volume in water, and the phenacyl saccharin readily precipitated.
I put my nose directly over the büchner while washing the product with copious amounts of water and similar to the synthesis of phenacyl phthalimide from K-phthalimide in DMF, my nose couldn't detect any residues of phenacyl bromide(not a single drop of a tear/snot

).
I know that is not a scientific method at all, but it is still a very effective measure to detect residual bromoacetophenone

I used 75mmol of bromoacetophenone and 80mmol of sodium saccharin, in 30ml of DMF, heated to 110°C for an hour and let it cool down before precipitation with water. Filtered and pressed the residual water out of the filter, then dried with mild heat overnight.
The outcome was after drying 19,5g of a dry fluffy white powder(64,28mmol) with a faint hint of acetophenone, which is a yield of 85,7%.

[Edited on 13-11-2019 by karlos³]

Pumukli - 13-11-2019 at 05:38

Thanks for the report! Useful information. DMSO may also work, it is almost OTC in these days. Just wondering.

karlos³ - 13-11-2019 at 12:20

I would place my bet rather on NMP, at least that is completely OTC.